# Deployment Checklist Source: https://docs.datarm.ai/admin-guide/deployment-checklist Prerequisites and steps for deploying AVA # Deployment Checklist **Coming Soon**: This section is under development. Check back soon for a comprehensive deployment checklist and prerequisites guide. ## What You'll Learn * Azure subscription requirements * Microsoft Entra ID admin access prerequisites * Required Azure services and permissions * Custom domain configuration (optional) * Deployment timeline and milestones * Post-deployment validation steps ## In the Meantime Review our [Deployment Architecture](/deployment/architecture) documentation for detailed information on AVA's Azure infrastructure and deployment process. ## Related Topics * [Azure Prerequisites](/deployment/architecture#deployment-process) - What you need before deploying * [Security Architecture](/deployment/security) - Understanding the security model * [Entra ID Setup](/admin-guide/entra-id-setup) - Identity configuration *** *Last Updated: January 2025* # Entra ID Setup Source: https://docs.datarm.ai/admin-guide/entra-id-setup Configuring Microsoft Entra ID Enterprise Application # Entra ID Setup **Coming Soon**: This section is under development. Check back soon for step-by-step Entra ID configuration instructions. ## What You'll Learn * Creating the Enterprise Application * Configuring App Registration * Setting up delegated permissions * Assigning users and groups * Conditional Access policies * Multi-factor authentication configuration ## In the Meantime Learn about AVA's identity model in our [Security Architecture](/deployment/security) documentation, specifically the [Identity & Access Management](/deployment/security#identity--access-management) section. ## Related Topics * [Security Model](/deployment/security#microsoft-entra-id-integration) - Understanding delegated permissions * [User Provisioning](/admin-guide/user-provisioning) - Adding users * [Deployment Architecture](/deployment/architecture#identity--access-layer) - Identity flow *** *Last Updated: January 2025* # MCP Server Setup Source: https://docs.datarm.ai/admin-guide/mcp-servers Configuring integrations with enterprise applications # MCP Server Setup **Coming Soon**: This section is under development. Check back soon for detailed MCP server configuration guides. ## What You'll Learn * Understanding MCP server architecture * Configuring Microsoft 365 integration (automatic) * Setting up Salesforce Connected App * Configuring Jira OAuth application * Setting up GitHub App integration * NetSuite integration setup * Troubleshooting connection issues ## In the Meantime Explore our [Integrations](/integrations/overview) documentation for information on each MCP server and how they connect to enterprise applications. ## Related Topics * [Microsoft 365 Integration](/integrations/microsoft-365) - Calendar, Mail, Files, Teams * [Salesforce Integration](/integrations/salesforce) - CRM setup * [Deployment Architecture](/deployment/architecture#mcp-server-layer) - MCP server technical details *** *Last Updated: January 2025* # Monitoring & Health Checks Source: https://docs.datarm.ai/admin-guide/monitoring Monitoring AVA deployment health and performance # Monitoring & Health Checks **Coming Soon**: This section is under development. Check back soon for monitoring and observability guidance. ## What You'll Learn * Azure Monitor integration * Application Insights configuration * Container Apps logs and metrics * Database performance monitoring * Key Vault access logging * Setting up alerts and notifications * Cost analysis and optimization ## In the Meantime See the [Deployment Architecture](/deployment/architecture#monitoring--observability) documentation for information on Azure Monitor integration and observability features. ## Related Topics * [Deployment Architecture](/deployment/architecture) - Monitoring & observability section * [Security Audit Logging](/deployment/security#audit-logging) - Security event monitoring * [Deployment Sizing](/deployment/architecture#deployment-sizing) - Performance optimization *** *Last Updated: January 2025* # Admin Guide Overview Source: https://docs.datarm.ai/admin-guide/overview Deployment and administration guide for IT teams # Admin Guide Overview Welcome to the AVA Admin Guide! This section provides IT administrators and deployment teams with the information needed to deploy, configure, and maintain AVA in your Azure tenant. **Note**: Some pages in this section are still under development. Check back soon for additional deployment and administration guides. ## What You'll Find Here Prerequisites and deployment steps Configuring Enterprise Application Setting up integrations Adding and managing users Health checks and Azure Monitor Full deployment architecture ## Core Documentation * [Deployment Architecture](/deployment/architecture) - Complete Azure deployment details * [Security Model](/deployment/security) - Identity, permissions, and compliance * [Integrations](/integrations/overview) - MCP server configuration ## Need Help? Get help from our support team Request deployment assistance *** *Last Updated: January 2025* # User Provisioning Source: https://docs.datarm.ai/admin-guide/user-provisioning Adding and managing users in AVA # User Provisioning **Coming Soon**: This section is under development. Check back soon for user provisioning and management guidance. ## What You'll Learn * Adding users via Entra ID groups * User access levels and permissions * Removing user access * Knowledge Search corpus permissions * Prompt Library sharing permissions * Monitoring user activity ## In the Meantime Review the [Security Architecture](/deployment/security) documentation to understand how AVA uses delegated permissions and Entra ID for user access control. ## Related Topics * [Entra ID Setup](/admin-guide/entra-id-setup) - Configuring identity * [Security Model](/deployment/security#identity--access-management) - User permissions * [Knowledge Search Access](/features/knowledge-search#corpus-permissions) - Document permissions *** *Last Updated: January 2025* # Champion Program Source: https://docs.datarm.ai/adoption/champion-program Building internal AVA advocates # Champion Program **Coming Soon**: This section is under development. Check back soon for guidance on building an AVA champion program. ## What You'll Learn * Identifying potential champions * Champion roles and responsibilities * Training champions effectively * Creating champion networks * Recognizing and rewarding champions * Scaling champion influence ## Champion Responsibilities * **Early Adoption**: Be first to try new features * **Peer Training**: Help colleagues learn AVA * **Feedback**: Provide insights to IT and leadership * **Best Practices**: Share successful workflows * **Advocacy**: Promote AVA benefits ## In the Meantime Identify power users in your organization who can help drive adoption. Provide them with advanced training using our [User Guide](/user-guide/overview) and [Use Cases](/essentials/use-cases) documentation. ## Related Topics * [Training Resources](/adoption/training-resources) - Materials for champions * [Rollout Strategy](/adoption/rollout-strategy) - Champion role in rollout * [Use Cases](/essentials/use-cases) - Examples champions can share *** *Last Updated: January 2025* # Measuring Success Source: https://docs.datarm.ai/adoption/measuring-success Usage metrics and KPIs for AVA adoption # Measuring Success **Coming Soon**: This section is under development. Check back soon for guidance on measuring AVA adoption and ROI. ## What You'll Learn * Key performance indicators (KPIs) for AVA adoption * User engagement metrics * Time savings calculations * ROI measurement frameworks * Azure Monitor dashboards for usage analytics * Reporting templates ## Suggested KPIs * **Adoption Rate**: % of users actively using AVA * **Engagement**: Sessions per user per week * **Feature Usage**: Chat vs Tasks vs Knowledge Search * **Time Savings**: Estimated hours saved per user * **User Satisfaction**: Survey scores and feedback * **Support Tickets**: Reduction in IT/support requests ## In the Meantime Use [Azure Monitor](/admin-guide/monitoring) to track basic usage metrics for your AVA deployment. ## Related Topics * [Admin Monitoring](/admin-guide/monitoring) - Azure Monitor integration * [Rollout Strategy](/adoption/rollout-strategy) - Setting success criteria * [Deployment Architecture](/deployment/architecture#monitoring--observability) - Observability *** *Last Updated: January 2025* # Adoption & Training Overview Source: https://docs.datarm.ai/adoption/overview Strategies for successfully rolling out AVA in your organization # Adoption & Training Overview Successfully deploying AVA is just the beginning. This section provides guidance on introducing AVA to your organization, training users, and measuring success. **Note**: Some pages in this section are still under development. Check back soon for additional adoption strategies and training materials. ## What You'll Find Here How to introduce AVA to your organization Materials for training your team Usage metrics and KPIs Building internal advocates ## Getting Started with Adoption Define timeline, pilot users, and success criteria Identify and train power users who can help others Start with a small group before org-wide rollout Collect user feedback and iterate Roll out to all users with ongoing support ## Need Help? Get adoption guidance from our team Request customer success assistance *** *Last Updated: January 2025* # Rollout Strategy Source: https://docs.datarm.ai/adoption/rollout-strategy How to introduce AVA to your organization # Rollout Strategy **Coming Soon**: This section is under development. Check back soon for comprehensive rollout planning guidance. ## What You'll Learn * Defining rollout phases and timeline * Identifying pilot user groups * Creating a communication plan * Managing change and expectations * Addressing common concerns * Scaling from pilot to organization-wide ## Key Rollout Phases 1. **Planning & Preparation**: Define goals, success criteria, timeline 2. **Pilot Launch**: Small group of early adopters 3. **Feedback & Iteration**: Gather insights, refine approach 4. **Phased Rollout**: Gradual expansion to departments 5. **Organization-Wide**: Full deployment with ongoing support ## In the Meantime Contact [avasupport@datarm.ai](mailto:avasupport@datarm.ai) for personalized rollout planning assistance. ## Related Topics * [Training Resources](/adoption/training-resources) - Materials for your team * [Champion Program](/adoption/champion-program) - Building advocates * [Use Cases](/essentials/use-cases) - Department-specific examples *** *Last Updated: January 2025* # Training Resources Source: https://docs.datarm.ai/adoption/training-resources Materials for training your team on AVA # Training Resources **Coming Soon**: This section is under development. Check back soon for training materials, templates, and guides. ## What You'll Find * Training presentation templates * Quick reference guides * Role-specific training paths * Self-paced learning modules * Certification programs (future) * Train-the-trainer materials ## Available Resources Now 5-minute getting started guide for new users Comprehensive end-user documentation Real-world examples by role and industry Video tutorials (coming soon) ## In the Meantime Use our existing documentation to create custom training materials for your organization. ## Related Topics * [Rollout Strategy](/adoption/rollout-strategy) - Planning your launch * [Champion Program](/adoption/champion-program) - Training advocates * [Resources](/resources/overview) - Templates and guides *** *Last Updated: January 2025* # AVA 1.10.0 Source: https://docs.datarm.ai/changelog/About AVA 1.10.0 Release notes for AVA version 1.10.0 # Release Notes for AVA Update 1.10.0 We're excited to bring you a patch update with the latest fixes and enhancements in AVA Update 1.10.0, designed to improve your experience and address recent issues. > For complete User Documentation, go to [AVA User Documentation](https://datarm.notion.site/AVA-User-Documentation-d522aab7765b4f4aaf38aeaf0186e66e). ## General * **New Settings Modal**: A new settings modal has been added to the top-right corner of the interface. This modal contains: * AVA Help Documents * AVA Release Notes * User Settings * Contact DataRM Link * Logout button * **Save to Home Screen**: iOS users can now save AVA to their home screen. * **Dark Mode**: Toggle dark mode on/off in AVA Settings. * **AVA New User Walkthrough**: The new user walkthrough has been updated and improved. * **PowerPoint Support**: AVA now supports Powerpoint files from SharePoint, OneDrive and your own computer. * **Custom Instructions**: New custom instructions field that allows users to give AVA their own system prompt. ## Tasks * **Disable Superprompt**: You can now disable the Tasks Superprompt if you'd like AVA to only use your task description. * **Task Library Relocation**: The Task Library has been moved to the top of the tasks page. ## Knowledge Search * **View Unauthorized Sharepoint Sites**: Users can click the "Can't find a Sharepoint site?" button to view a list of sites that the user doesn't have permission to access. * **View Data Source**: Added a "View Data Source" button that directs to the specific Sharepoint site for the selected corpus. ## Application Improvements * Uploaded images now display tiny images in file thumbnail. * Fixed a scrolling bug when looking at a large output inside of AVA Chat. * Added a login loading bar. *** \ These updates are designed to enhance user experience by providing easier access to settings, improving navigation, and introducing new features like custom instructions and corpus creation. # AVA 1.10.1 Source: https://docs.datarm.ai/changelog/About AVA 1.10.1 Release notes for AVA version 1.10.1 # Release Notes for AVA Update 1.10.1 We're excited to bring you a patch update with the latest fixes and enhancements in AVA Update 1.10.1, designed to improve your experience and address recent issues. > For complete User Documentation, go to [AVA User Documentation](https://datarm.notion.site/AVA-User-Documentation-d522aab7765b4f4aaf38aeaf0186e66e). ## General * **New Settings Modal**: A new settings modal has been added to the top-right corner of the interface. This modal contains: * AVA Help Documents * AVA Release Notes * User Settings * Contact DataRM Link * Logout button * **Save to Home Screen**: iOS users can now save AVA to their home screen. * **Dark Mode**: Toggle dark mode on/off in AVA Settings. * **AVA New User Walkthrough**: The new user walkthrough has been updated and improved. * **PowerPoint Support**: AVA now supports Powerpoint files from SharePoint, OneDrive and your own computer. * **Custom Instructions**: New custom instructions field that allows users to give AVA their own system prompt. ## Tasks * **Disable Superprompt**: You can now disable the Tasks Superprompt if you'd like AVA to only use your task description. * **Task Library Relocation**: The Task Library has been moved to the top of the tasks page. ## Knowledge Search * **View Unauthorized Sharepoint Sites**: Users can click the "Can't find a Sharepoint site?" button to view a list of sites that the user doesn't have permission to access. * **View Data Source**: Added a "View Data Source" button that directs to the specific Sharepoint site for the selected corpus. ## Application Improvements * Uploaded images now display tiny images in file thumbnail. * Fixed a scrolling bug when looking at a large output inside of AVA Chat. * Added a login loading bar. ## HOTFIX 1.10.1 * Fixed a bug where app would crash when the user accessed the settings page if the user's profile did not have a profile picture. *** \ These updates are designed to enhance user experience by providing easier access to settings, improving navigation, and introducing new features like custom instructions and corpus creation. # AVA 1.3.6 Source: https://docs.datarm.ai/changelog/About AVA 1.3.6 Release notes for AVA version 1.3.6 # Updates in AVA Julius 1.3.6 This update to the Beta release introduces several exciting new features > For complete User Documentation, go to [AVA User Documentation](https://datarm.notion.site/AVA-User-Documentation-d522aab7765b4f4aaf38aeaf0186e66e). ## General * **Logout**: Select the logout option from the settings menu to log out of AVA. ## Chat * **Sharing Chats**: Click the share icon on a chat in session history to share your chat with others on your team. * **Fine-tune AVA**: Tune your AVA response with the new Analytic/Balanced/Creative switch for different chat modes. * **Stop Button**: Ask the wrong question? You can interrupt AVA by clicking the stop button in Chat. [Chat Documentation](https://datarm.notion.site/AVA-Chat-7799bf2a84a842cb8ddc164303f4f41b) ## Tasks * **New Look**: Upgraded Tasks now include a new look and feel, with more intuitive layout and improved navigation. * **Stop Button**: Stop a response from AVA by clicking the stop button in the Task window. [Tasks Documentation](https://datarm.notion.site/AVA-Tasks-b59f920b85884e9f98cb3c2cba14d9e4) ## Knowledge Search * **Secure Search**: Securely search your organizations private knowledge bases using natural language queries powered by AI Vector Similartity Search! [Knowledge Search Documentation](https://datarm.notion.site/AVA-Knowledge-Search-In-Continued-Development-9b4904139f224574a8d4995f2d0ca333) ## Workspace * **New Look**: Workspace now has a new look and feel, with a more intuitive layout and improved navigation. * **Formatting**: New Workspaces now have more formatting options. * **Edit in Workspace**: You can still edit with markdown if you prefer. * **File Viewer**: View files cited in Knowledge Search chats, or review documents before adding them to the chat/task from the search modal. ## Bug Fixes 1. Workspace content now saves in tasks correctly. 2. Adding too much text to an AVA chat previously broke chat. This is now resolved. ## Known Issues 1. Files selected from "My Computer" are not display in the fileviewer. 2. Text-based files from "My Computer" such as .md, .js, .py, etc. can be selected, but AVA cannot read or upload them … yet. 3. The stop button will sometimes fail to stop AVA from chattering on undeterred by your wishes. 4. On mobile devices, the "+" icon in the "Add to Chat" button is not rendering correctly. # AVA 1.3.7 Source: https://docs.datarm.ai/changelog/About AVA 1.3.7 Release notes for AVA version 1.3.7 # Release Notes for AVA Update 1.3.7 We’re back with another round of upgrades and tweaks to keep your AVA experience smooth. > For complete User Documentation, go to [AVA User Documentation](https://datarm.notion.site/AVA-User-Documentation-d522aab7765b4f4aaf38aeaf0186e66e). ## General * **More AI Models**: Choose from a variety of region-specific AI models in Tasks to improve AVA's output. * **Swift Login**: Enjoy a faster login experience that gets you where you need to be in record time. * **Extended Timeout**: AVA now waits through a full hour of patient inactivity before bidding you adieu. ## Chat * **Improved Notifications**: See what AVA is doing with better notifications when uploading files and sharing chats. [Chat Documentation](https://datarm.notion.site/AVA-Chat-7799bf2a84a842cb8ddc164303f4f41b) ## Knowledge Search * **Enhanced Knowledge Search**: You can now search your knowledge bases and files with greater accuracy. * **Tailored Prompts**: Knowledge Search now serves up specific prompts for a more personalized experience. [Knowledge Search Documentation](https://datarm.notion.site/AVA-Knowledge-Search-In-Continued-Development-9b4904139f224574a8d4995f2d0ca333) ## Known Issues * Text-based files from "My Computer" such as .md, .js, .py, etc. can be selected, but AVA cannot read or upload them … yet. * The stop button will sometimes fail to stop AVA from chattering on undeterred by your wishes. * On mobile devices, the "+" icon in the "Add to Chat" button is not rendering correctly. ## Application Improvements * **Optimized State Handling**: We've fine-tuned state management for a smoother ride. * **Feature Flag Flexibility**: Access available models with ease from OAI direct or Azure, thanks to feature flag enhancements. * **Better Token Estimation**: AVA now predicts your token needs with even greater accuracy. *** \ All in, this host of enhancements to make your AVA interactions more delightful than ever. Stay tuned for more exciting updates in the pipeline! # AVA 1.7.0 Source: https://docs.datarm.ai/changelog/About AVA 1.7.0 Release notes for AVA version 1.7.0 # Release Notes for AVA Update 1.7.0 We’re back with another round of upgrades and tweaks to keep your AVA experience smooth. > For complete User Documentation, go to [AVA User Documentation](https://datarm.notion.site/AVA-User-Documentation-d522aab7765b4f4aaf38aeaf0186e66e). ## Knowledge Search * **Embedded File Viewer Updates:** Bugfixes to ensure files open as expected within the file viewer. [Knowledge Search Documentation](https://datarm.notion.site/AVA-Knowledge-Search-In-Continued-Development-9b4904139f224574a8d4995f2d0ca333) ## Workspace * **Workspace Behavior:** AVA now opens the workspace by default on login. ## Application Improvements * **Testing Framework:** Added a testing framework to ensure the ever more complex application continues to work correctly as we add new and exciting features. * **About Ava Relocation:** AVA now accesses release notes from a central location. *** \ Overall, this host of enhancements intends to make your AVA interactions more delightful than ever. Stay tuned for more exciting updates in the pipeline! # AVA 1.8.0 Source: https://docs.datarm.ai/changelog/About AVA 1.8.0 Release notes for AVA version 1.8.0 # Release Notes for AVA Update 1.8.0 We’re back with another round of upgrades and tweaks to keep your AVA experience smooth. > For complete User Documentation, go to [AVA User Documentation](https://datarm.notion.site/AVA-User-Documentation-d522aab7765b4f4aaf38aeaf0186e66e). ## General * **AVA Mobile Experience** AVA now supports mobile devices, allowing users to access the platform on the go. * **Search Modal Meeting Transcripts** Users can now search for recorded Teams meeting transcripts within the search feature and add those transcripts to both chats and tasks. * **Job Title Bugfix:** A bug that prevented users from accessing AVA when their job title was over 50 characters has been fixed. ## Tasks * **Task Output** Users can now view the output of tasks in the workspace * [Shown here](https://datarm.notion.site/Getting-Started-082019aba9bc4a6991eea1de8e8f5dc6#58494d36eb974900815047c68199395d). [Tasks Documentation](https://datarm.notion.site/AVA-Tasks-b59f920b85884e9f98cb3c2cba14d9e4) ## Workspace * **Workspace Refactored:** The workspace has been refactored to improve performance and user experience. Users can now adjust the size of the workspace to suit their needs * [Workspace Overview](https://datarm.notion.site/Getting-Started-082019aba9bc4a6991eea1de8e8f5dc6#58494d36eb974900815047c68199395d) * **Search Feature Relocated to Workspace** Users can now search for meeting transcripts within the workspace search feature. * **Search Modal Updates** The search feature that allows users to add files from Office365 or their device has been relocated to the workspace as its own tab. *** \ All in, this host of enhancements to make your AVA interactions more delightful than ever. Stay tuned for more exciting updates in the pipeline! # AVA 1.9.0 Source: https://docs.datarm.ai/changelog/About AVA 1.9.0 Release notes for AVA version 1.9.0 # Release Notes for AVA Update 1.9.0 We're excited to bring you the latest enhancements and fixes in AVA Update 1.9.0, designed to make your experience even smoother and more efficient. > For complete User Documentation, go to [AVA User Documentation](https://datarm.notion.site/AVA-User-Documentation-d522aab7765b4f4aaf38aeaf0186e66e). ## General * **Snackbar Notifications:** Snackbar notifications now indicate whether a PDF was OCR'ed. * **Removed File Upload Limit:** The 5MB limit on uploaded files has been removed. * **Enhanced Header Actions:** Users can now share, delete, and edit the title of chats, tasks, and searches directly from the header. * **Feature Visibility:** The header now displays which feature you're using. * **Microsoft Profile Integration:** AVA now recognizes you and basic information from your Microsoft profile. * **Tabular Data Display:** AVA can now display tabular data in a proper table format. * **UI Updates and Bug Fixes:** Various small UI updates and bug fixes to enhance overall user experience. * **New 4o Mini Feature:** Added new default model gpt-4o-mini to enhance functionality and reduce costs. * **Meeting Transcript Storage:** Updates to how meeting transcripts are stored for better management. ## Tasks * **Task Redesign:** Tasks have been redesigned to use a single user prompt and an AVA-generated superprompt. * **Task Library Relocation:** The task library has been moved below tasks for easier access. * **Task Library Redesign:** Redesigned the task library for better usability. * [Task Interface Guide](https://datarm.notion.site/AVA-Tasks-b59f920b85884e9f98cb3c2cba14d9e4#84906de458fb48a781d28f2a8d6840c7) * **Search Tab Highlight:** If the search tab is open, clicking the "Add Documents and Data" button will highlight that tab to guide new users. * **Superprompt:** AVA now generates a SuperPrompt based on your input to help you get the most out of your tasks! * [Superprompt Guide](https://datarm.notion.site/AVA-Tasks-b59f920b85884e9f98cb3c2cba14d9e4#1038845bbd0680da9423f8f237ac7a55) [Tasks Documentation](https://datarm.notion.site/AVA-Tasks-b59f920b85884e9f98cb3c2cba14d9e4) ## Knowledge Search * **SharePoint Integration:** Clicking list items in the Embedded Files Viewer now navigates to that item in SharePoint. [Knowledge Search Documentation](https://datarm.notion.site/AVA-Knowledge-Search-In-Continued-Development-9b4904139f224574a8d4995f2d0ca333) ## Workspace * **Fetch Past Meeting Transcripts:** Added a button to fetch past meeting transcripts. * **File Search Results Update:** Visual update to file search results on Desktop, now located in its own tab in the workspace. ## Application Improvements * **Enhanced Token Counting:** Improved token counting for better accuracy and performance. * **Automated Testing Suite:** Introduced a Playwright testing suite for automated testing. * **Improved Token Counting:** Enhanced token counting for backend processes. *** \ These updates are designed to make your interactions with AVA more efficient and enjoyable. Stay tuned for more exciting features and improvements in future releases! # AVA 1.9.1 Source: https://docs.datarm.ai/changelog/About AVA 1.9.1 Release notes for AVA version 1.9.1 # Release Notes for AVA Update 1.9.1 We're excited to bring you a patch update with the latest fixes and enhancements in AVA Update 1.9.1, designed to improve your experience and address recent issues. > For complete User Documentation, go to [AVA User Documentation](https://datarm.notion.site/AVA-User-Documentation-d522aab7765b4f4aaf38aeaf0186e66e). ## General * **Recent Files Reload:** Recent files will now reload when switching back to the "All" search option. * **Support for 4o-0806 Model:** Added support for the 4o-0806 model to enhance functionality. ## Tasks * **Task Title Save Fix:** Fixed an issue where the task title was not saving in the Save modal. * **Tooltip Fix:** Resolved an issue where tooltips were coming through the Save modal in tasks. [Tasks Documentation](https://datarm.notion.site/AVA-Tasks-b59f920b85884e9f98cb3c2cba14d9e4) ## Workspace * **Workspace Button Visibility:** The workspace button will no longer be hidden when the library is open. *** \ These updates are designed to address recent issues and improve your interactions with AVA. Stay tuned for more exciting features and improvements in future releases! # AVA 2.0.0 Source: https://docs.datarm.ai/changelog/About AVA 2.0.0 Release notes for AVA version 2.0.0 # Release Notes for AVA Update 2.0.0 Welcome to AVA 2.0.0! In this release, we’re introducing powerful new features and enhancements designed to make your experience smoother and more efficient. From advanced task management to seamless login improvements, AVA 2.0.0 is here to support your workflow with greater functionality and reliability. > For complete User Documentation, go to [AVA User Documentation](https://datarm.notion.site/AVA-User-Documentation-d522aab7765b4f4aaf38aeaf0186e66e). ## General * **Login Improvements:** The login process has been enhanced to support longer sessions and faster access to AVA. * **New User Walkthrough Button:** The new user walkthrough can now be triggered by a button in the top-left of AVA's header. * **Notification Bell:** The notification bell has been removed from the header. ## Tasks * **Multi-Step Tasks:** The Tasks page has been revamped! AVA Tasks now supports multi-step functionality, allowing you to add multiple stages within a task and share linear progress seamlessly. * **Superprompt Toggle:** The Superprompt can now be toggled on and off. * **Sharing Tasks:** You can now use the share buttons to share Tasks with team members, just like you do with AVA Chat and Knowledge Search sessions. * **Task Library Updates:** The task library is now specific to your saved task templates. * **Automatic Task Title Generation:** If you don't change the name of your task when saving, a title will be generated for you, instead of "New Task". [Tasks Documentation](https://datarm.notion.site/AVA-Tasks-b59f920b85884e9f98cb3c2cba14d9e4) ## Knowledge Search * **Create a Corpus:** You can now create your own corpus within AVA's Knowledge Search! * **Add Files to Corpus:** You can now add files directly to a Knowledge Base from your Corpus SharePoint site or your computer. [Knowledge Search Documentation](https://datarm.notion.site/AVA-Knowledge-Search-In-Continued-Development-9b4904139f224574a8d4995f2d0ca333) ## Application Improvements * **Bug Fixes:** We've resolved various issues to improve AVA's stability, performance, and overall reliability. *** \ These updates are designed to enhance user experience by introducing multi-step task management, streamlining login, and adding new capabilities for managing knowledge data. With these improvements, AVA 2.0.0 aims to provide smoother workflows, easier access to key features, and an overall more intuitive experience. # AVA 2.0.1 Source: https://docs.datarm.ai/changelog/About AVA 2.0.1 Release notes for AVA version 2.0.1 # Release Notes for AVA Update 2.0.1 Welcome to AVA 2.0.0! In this release, we’re introducing powerful new features and enhancements designed to make your experience smoother and more efficient. From advanced task management to seamless login improvements, AVA 2.0.0 is here to support your workflow with greater functionality and reliability. > For complete User Documentation, go to [AVA User Documentation](https://datarm.notion.site/AVA-User-Documentation-d522aab7765b4f4aaf38aeaf0186e66e). ## General * **Login Improvements:** The login process has been enhanced to support longer sessions and faster access to AVA. * **New User Walkthrough Button:** The new user walkthrough can now be triggered by a button in the top-left of AVA's header. * **Notification Bell:** The notification bell has been removed from the header. ## Tasks * **Multi-Step Tasks:** The Tasks page has been revamped! AVA Tasks now supports multi-step functionality, allowing you to add multiple stages within a task and share linear progress seamlessly. * **Superprompt Toggle:** The Superprompt can now be toggled on and off. * **Sharing Tasks:** You can now use the share buttons to share Tasks with team members, just like you do with AVA Chat and Knowledge Search sessions. * **Task Library Updates:** The task library is now specific to your saved task templates. * **Automatic Task Title Generation:** If you don't change the name of your task when saving, a title will be generated for you, instead of "New Task". [Tasks Documentation](https://datarm.notion.site/AVA-Tasks-b59f920b85884e9f98cb3c2cba14d9e4) ## Knowledge Search * **Create a Corpus:** You can now create your own corpus within AVA's Knowledge Search! * **Add Files to Corpus:** You can now add files directly to a Knowledge Base from your Corpus SharePoint site or your computer. [Knowledge Search Documentation](https://datarm.notion.site/AVA-Knowledge-Search-In-Continued-Development-9b4904139f224574a8d4995f2d0ca333) ## Application Improvements * **Bug Fixes:** We've resolved various issues to improve AVA's stability, performance, and overall reliability. ## HOTFIX 2.0.1 * Fixed a small empty state bug when getting corpus information. * Initially corpus embeddings were happening internally. We updated AVA to do those embeddings within the customer's own Azure tenant. * Fixed a bug that cause a crash if the user didn't have a phone number. *** \ These updates are designed to enhance user experience by introducing multi-step task management, streamlining login, and adding new capabilities for managing knowledge data. With these improvements, AVA 2.0.0 aims to provide smoother workflows, easier access to key features, and an overall more intuitive experience. # AVA 2.0.2 Source: https://docs.datarm.ai/changelog/About AVA 2.0.2 Release notes for AVA version 2.0.2 # Release Notes for AVA Update 2.0.2 Welcome to AVA 2.0.0! In this release, we’re introducing powerful new features and enhancements designed to make your experience smoother and more efficient. From advanced task management to seamless login improvements, AVA 2.0.0 is here to support your workflow with greater functionality and reliability. > For complete User Documentation, go to [AVA User Documentation](https://datarm.notion.site/AVA-User-Documentation-d522aab7765b4f4aaf38aeaf0186e66e). ## General * **Login Improvements:** The login process has been enhanced to support longer sessions and faster access to AVA. * **New User Walkthrough Button:** The new user walkthrough can now be triggered by a button in the top-left of AVA's header. * **Notification Bell:** The notification bell has been removed from the header. ## Tasks * **Multi-Step Tasks:** The Tasks page has been revamped! AVA Tasks now supports multi-step functionality, allowing you to add multiple stages within a task and share linear progress seamlessly. * **Superprompt Toggle:** The Superprompt can now be toggled on and off. * **Sharing Tasks:** You can now use the share buttons to share Tasks with team members, just like you do with AVA Chat and Knowledge Search sessions. * **Task Library Updates:** The task library is now specific to your saved task templates. * **Automatic Task Title Generation:** If you don't change the name of your task when saving, a title will be generated for you, instead of "New Task". [Tasks Documentation](https://datarm.notion.site/AVA-Tasks-b59f920b85884e9f98cb3c2cba14d9e4) ## Knowledge Search * **Create a Corpus:** You can now create your own corpus within AVA's Knowledge Search! * **Add Files to Corpus:** You can now add files directly to a Knowledge Base from your Corpus SharePoint site or your computer. [Knowledge Search Documentation](https://datarm.notion.site/AVA-Knowledge-Search-In-Continued-Development-9b4904139f224574a8d4995f2d0ca333) ## Application Improvements * **Bug Fixes:** We've resolved various issues to improve AVA's stability, performance, and overall reliability. ## HOTFIX 2.0.2 * In certain instances, AVA would crash when a user deleted a task. She doesn't do that now. * Fixed a bug where descriptions in a task step were not saving properly. *** \ These updates are designed to enhance user experience by introducing multi-step task management, streamlining login, and adding new capabilities for managing knowledge data. With these improvements, AVA 2.0.0 aims to provide smoother workflows, easier access to key features, and an overall more intuitive experience. # AVA 2.0.3 Source: https://docs.datarm.ai/changelog/About AVA 2.0.3 Release notes for AVA version 2.0.3 # Release Notes for AVA Update 2.0.3 Welcome to AVA 2.0.0! In this release, we’re introducing powerful new features and enhancements designed to make your experience smoother and more efficient. From advanced task management to seamless login improvements, AVA 2.0.0 is here to support your workflow with greater functionality and reliability. > For complete User Documentation, go to [AVA User Documentation](https://datarm.notion.site/AVA-User-Documentation-d522aab7765b4f4aaf38aeaf0186e66e). ## General * **Login Improvements:** The login process has been enhanced to support longer sessions and faster access to AVA. * **New User Walkthrough Button:** The new user walkthrough can now be triggered by a button in the top-left of AVA's header. * **Notification Bell:** The notification bell has been removed from the header. ## Tasks * **Multi-Step Tasks:** The Tasks page has been revamped! AVA Tasks now supports multi-step functionality, allowing you to add multiple stages within a task and share linear progress seamlessly. * **Superprompt Toggle:** The Superprompt can now be toggled on and off. * **Sharing Tasks:** You can now use the share buttons to share Tasks with team members, just like you do with AVA Chat and Knowledge Search sessions. * **Task Library Updates:** The task library is now specific to your saved task templates. * **Automatic Task Title Generation:** If you don't change the name of your task when saving, a title will be generated for you, instead of "New Task". [Tasks Documentation](https://datarm.notion.site/AVA-Tasks-b59f920b85884e9f98cb3c2cba14d9e4) ## Knowledge Search * **Create a Corpus:** You can now create your own corpus within AVA's Knowledge Search! * **Add Files to Corpus:** You can now add files directly to a Knowledge Base from your Corpus SharePoint site or your computer. [Knowledge Search Documentation](https://datarm.notion.site/AVA-Knowledge-Search-In-Continued-Development-9b4904139f224574a8d4995f2d0ca333) ## Application Improvements * **Bug Fixes:** We've resolved various issues to improve AVA's stability, performance, and overall reliability. ## HOTFIX 2.0.3 * Added handling for non-US phone numbers. * Created a smoother experience when loading recent files in the Search tab *** \ These updates are designed to enhance user experience by introducing multi-step task management, streamlining login, and adding new capabilities for managing knowledge data. With these improvements, AVA 2.0.0 aims to provide smoother workflows, easier access to key features, and an overall more intuitive experience. # AVA 2.1.0 Source: https://docs.datarm.ai/changelog/About AVA 2.1.0 Release notes for AVA version 2.1.0 # Release Notes for AVA Update 2.1.0 Welcome to AVA 2.1.0! This release brings a host of exciting new features and enhancements to both the frontend and backend, designed to improve your experience with AVA. From UI updates to advanced integration capabilities, AVA 2.1.0 is here to streamline your workflow and enhance functionality. * **Pretty UI Updates:** Enjoy a more visually appealing interface with our latest UI enhancements. * **@ Mention Menu:** In chats, users can now select an @ mention from a menu to call functions, including AVA help functions and Salesforce queries. * **Knowledge Search (KS) Enhancements:** * Perform KS against a subset of documents rather than the entire corpus by selecting checkboxes. * Users can now delete files from a KS corpus as needed. * Improved toast notifications for adding files to a corpus. * **Chain of Thought:** A new accordion in KS responses displays Chain of Thought information to give additional insight into how AVA got the response you're seeing. * **Results Tab:** A new Results tab in the workspace shows all results of a KS similarity search, ranked from most relevant to least. * **Scalability Enhancements:** Increased app scalability for file uploads, allowing more users to include files in chats and tasks without affecting server stability. * **Task Image Support:** Users can now add images to tasks and fetch images from SharePoint. * **Performance Improvements:** Experience smoother performance with several minor optimizations for better reliability and maintainability. These updates are crafted to provide a more intuitive, efficient, and powerful experience with AVA. Enjoy the new capabilities and enhancements that AVA 2.1.0 has to offer! # AVA 2.1.1 Source: https://docs.datarm.ai/changelog/About AVA 2.1.1 Release notes for AVA version 2.1.1 # Release Notes for AVA Update 2.1.1 Hotfix Welcome to the AVA 2.1.1 Hotfix! This update addresses several critical issues and introduces minor enhancements to ensure a smoother experience with AVA. Here's what's been fixed and improved: * **Task Creation Default Setting:** The enableSuperPrompt feature is now set to false by default when creating new tasks or steps, ensuring a more streamlined task setup process. * **Session History Stability:** Resolved an issue where deleting an unselected task from the Session History would cause the app to crash, enhancing overall stability. * **Delete Popup Fix:** Corrected the issue where the DeletePopup was not appearing when clicking the trash icon in the Header to delete a session, ensuring proper functionality. * **@ Mention Guidance:** Added helpful text to the new chat screen to inform users about the support for @ mentions, improving user awareness and ease of use. * **Date and Time Awareness:** AVA now accurately recognizes the current date and time, enhancing its contextual understanding. * **PowerPoint Upload Fix:** Resolved a bug that was causing PowerPoint uploads to fail, ensuring seamless file handling. * **Handle Sharepoint Throttling:** Fix for errors when AVA fetches Sharepoint files too quickly. * **Standardize Fallback LLM Models:** Fix errors related to outdated or disparate fallback LLM models. These fixes and improvements are designed to enhance your experience with AVA, ensuring more reliable and efficient use of its features. Thank you for your continued support and feedback! # AVA 2.2.0 Source: https://docs.datarm.ai/changelog/About AVA 2.2.0 Release notes for AVA version 2.2.0 # Release Notes for AVA Update 2.2.0 Welcome to AVA 2.2.0! This release introduces powerful new features and enhancements to improve your experience with AVA. From expanded file support (e.g., Excel/CSV), to exporting content out in various formats, to advanced integration capabilities, AVA 2.2.0 is designed to streamline workflows, enhance collaboration, and provide deeper insights. Explore the latest updates below. *** ### **General Improvements** * **Location-Specific Knowledge Searches:** Perform Knowledge Searches tailored to specific locations for more targeted results. * **Improved Online Meeting Handling:** Enhanced handling of online meetings ensures smoother integration within AVA, ensuring that you get new meeting transcripts as soon as possible, and don't lose old ones. * **Tips and Tricks:** Peep the new widget at the bottom of the sidebar for helpful tips and tricks to maximize your use of AVA. *** ### **@ Mentions and Connected Data** * **Expanded @ Mentions Support:** Use **@Salesforce** and **@Jira** mentions to query and retrieve data directly from these platforms. (*This feature requires admin setup*) * **Sticky @ Mentions in Chat:** Mentions now persist within chat for easier follow-up queries. * **Call KS Corpuses in @ Mentions:** Seamlessly reference Knowledge Search corpuses via @ mentions for instant access to relevant data. * **@ Mentions in Tasks:** Use @ mentions in Tasks to quickly access connected data and Knowledge Source corpuses for enhanced workflow automation. *** ### **File and Export Enhancements** * **Excel/CSV/Log File Support:** Users can now upload and analyze PowerPoint, Excel, CSV, and log files within AVA. * **Export Options:** Export content from chat and the Workspace editor directly to MS Word, PDF, TXT, or Outlook for greater flexibility. * **Small Prompt Library for Uploaded Files:** When adding files to chat, you can now add from a list of curated prompts to help you get the most out of your files. (*Currently supported file types: Excel, CSV, and Meeting Transcripts*) *** ### **UI and Search Enhancements** * **New UI Component for In-Progress Meetings:** Quickly see in AVA when you have an active meeting, join the meeting with a click, or add that meeting's transcript to chats or tasks, even while it is still happening. * **Consolidated Search Source Options:** We've simplified the Search tab so you can quickly search your Microsoft SharePoint files and online meetings, and filter results by file type. * **Location and Date Columns in Search:** Added a **Location** column to files in the search tab and a **Date** column for embedded files in Knowledge Search for extra context. *** ### **Performance and Technical Updates** We’ve supercharged AVA behind the scenes! With smarter processing, faster responses, and a sprinkle of extra security, AVA is now smoother, quicker, and more reliable than ever. You’ll notice faster logins, better handling of complex tasks, and an overall boost in performance—so you can focus on getting work done without a hitch. *** These updates are crafted to provide a more intuitive, efficient, and secure experience with AVA. We’re excited for you to explore the new capabilities and enhancements in version 2.2.0. Thank you for your continued support and feedback! # AVA 2.2.1 Source: https://docs.datarm.ai/changelog/About AVA 2.2.1 Release notes for AVA version 2.2.1 # Release Notes for AVA Update 2.2.1 Hotfix Welcome to the AVA 2.2.1 Hotfix! This update focuses on optimizing performance, refining user experience, and addressing minor bugs to ensure AVA runs smoothly. Here's what's been updated: * **Session History Performance Improvement:** Users with extensive session histories were experiencing delays during login. AVA now fetches only the most recent sessions upon login, significantly improving load times. Users can click "Show All Sessions" at the bottom of Session History to access older sessions if needed. * **Release Notes Display Enhancement:** Updated the handling of release notes in User Settings. Users will now see only the release notes relevant to their version of AVA, ensuring clarity and relevance. * **Minor Bug Fixes:** Resolved several minor bugs to enhance stability and overall functionality. These updates are designed to improve your experience with AVA, providing faster performance and a more seamless interface. Thank you for your ongoing feedback and support! # AVA 2.3.0 Source: https://docs.datarm.ai/changelog/About AVA 2.3.0 Release notes for AVA version 2.3.0 # Release Notes for AVA Update 2.3.0 Welcome to AVA 2.3.0! This release introduces a handful of small, but mighty updates to improve your experience with AVA. * **Exporting to PowerPoint:** Users can now export content from chats or the Notes tab in the Workspace to PowerPoint. AVA will intelligently convert your text content into slides, complete with a title slide and a slide for each section of content. You can even choose from uploaded templates to customize the appearance of your presentation. * **Small Prompt Library:** We've updated a small Prompts button that appears above the chat box when you add a spreadsheet or meeting transcript to a chat. We've also given you some quick options to get you relevant prompts for file handling faster. * **Updates to Fetching Online Meetings:** Some users were experiencing issues fetching online meetings. We've improved that by paginating your meetings. This helps get you to your recent meetings faster, and you can click the new "load more" button to download the next set of meetings. * **Improved App Performance:** Thanks to a freak accident involving a lightning strike and chemicals in our laboratory, AVA is now significantly faster and more performant. We hope you like the new speed. These updates are designed to make your AVA experience more streamlined, efficient, and enjoyable. We hope you enjoy them! # AVA 2.3.2 Source: https://docs.datarm.ai/changelog/About AVA 2.3.2 Release notes for AVA version 2.3.2 # Release Notes for AVA Update 2.3.2 Welcome to AVA 2.3.2! Not a whole lot of changes this time around. Just a few small updates to improve your experience. * Improved handling for retrieving online meeting transcripts. * A couple small bug fixes around user settings. * Improved real-time updates for active meetings. Thank you for using AVA! 😊 # AVA 2.4.0 Source: https://docs.datarm.ai/changelog/About AVA 2.4.0 Release notes for AVA version 2.4.0 # Release Notes for AVA Update 2.4.0 Welcome to AVA 2.4.0! 🎉 This release makes AVA faster, smarter, and more reliable than ever. Here's what's new: *** ### **Key Updates** * **Support for New AI Models:** Reasoning models such as OpenAI's o1 and o3-mini and extremely large context models like GPT-4.1 are available in AVA Chat. * **Default to New Chat Experience:** All logins start in a new chat. This also applies when deleting the current session. * **Enhanced Task Flexibility:** You can now select any previous step in an AVA Task (not just the immediate prior step) as input, boosting workflow efficiency. * **Expanded Corpus Management:** Corpus owners can now manage files added by any user, creating more flexible organization-wide control. * **Improved Meeting Visibility:** The AVA Workspace now only shows meetings in your organization for which you have transcript access, eliminating unwanted clutter. * **Real-Time Status for Knowledge Search Files:** Instantly see the progress of file embedding in Knowledge Search, keeping you informed every step of the way. * **Rearchitected File Processing:** File handling has been redesigned to better support complex workflows, ensuring smooth performance and laying the groundwork for future capabilities. * **Performance Overhaul:** * AVA's frontend has been refactored for faster performance across the app. * Chats and Tasks have been changed to handle even larger data loads with ease. * Task execution now draws on this enhanced memory store, making AVA more responsive and eliminating issues with lag or crashes in Tasks. *** ### **Bug Fixes and Improvements** * Disabled chatboxes no longer submit messages on Enter, keeping workflows clean and intentional. * Refreshed the chat file context accordion for improved clarity and usability. * Client secrets are now protected from exposure in the browser to further enhance security. * Resolved issues where user settings could fail to fetch, maintaining a smooth configuration experience. * Improved retry logic for file prompt fetching, reducing errors and interruptions. * @ mentions now persist reliably across sessions and logins for seamless communication. * Addressed rare crashes during data comparison in tasks, improving reliability. * Fixed issues that could cause tasks to fail when using super prompts. * Enhanced Knowledge Search windows with clearer loading states and improved empty state messaging. * Long file names are now truncated in toast notifications, creating a cleaner display. * Success toasts for deleting Knowledge Search files have been restored. * Users without recent meetings can now reliably access and review their older meetings. *** These updates reinforce AVA's position as a reliable and innovative tool for your business needs. Thank you for your continued feedback and support—stay tuned for even more exciting developments! 🚀 # AVA 2.4.1 Source: https://docs.datarm.ai/changelog/About AVA 2.4.1 Release notes for AVA version 2.4.1 # Release Notes for AVA Update 2.4.1 AVA 2.4.1 is here! This release brings smarter models, smoother workflows, and key fixes to keep everything running seamlessly. *** ### **Highlights** * **New AI Models:** Support for additional models—4.1 mini, 4.1 nano, and Deepseek R1. Admin configuration may be required. * **Smarter SharePoint Management:** Batch and recursive file fetching for faster, more efficient file handling. * **Improved Tasks:** Fixed metadata fetching to ensure smoother workflows. *** ### **Fixes & Improvements** * **File Handling Tweaks:** SharePoint files load faster, and upload progress tracking is now precise and transparent. * **Meeting Transcript Embedding:** Fixed embedding issues for seamless meeting insights. * **Quota Exceeded Fix:** IndexedDB implementation prevents storage errors. * **Shared File Icons:** Consistent "shared with me" icons across tasks and chats. *** We’re always leveling up to make AVA better for you. Thanks for your feedback—exciting things are on the horizon! 🚀 # AVA 2.5.0 Source: https://docs.datarm.ai/changelog/About AVA 2.5.0 Release notes for AVA version 2.5.0 # Release Notes for AVA Update 2.5.0 AVA 2.5.0 has arrived! 🎉 This update brings exciting new features, smarter workflows, and improvements designed to enhance your experience. *** ### **Highlights** * **Prompt Library with RBAC:** Create reusable prompts, and share them with whole teams for smoother collaboration. Prompts use MS Teams access control. * **Filetype Specific Prompts:** Write and use new prompts specific to the file types added to chats. This can help you analyze images, Word, Excel files and more with ease. * **Support for Code File Types:** Upload, process, and analyze code files like `.py`, `.js`, and `.ts` effortlessly in Chats and Tasks. * **Sticky Workspace Divider:** Keep your workspace layout consistent with a divider that stays in place across sessions. You can customize your panel sizes in User Settings. * **Prompt History Navigation:** Use the up arrow in chat to recall your past messages in that session. Must be enabled in Settings. (Only applies to new sessions in 2.5.0 and beyond.) * **Sharing Notifications:** Send and receive email notifications when a session is shared, including access links and custom messages. * **Simplified Theme Options:** AVA now defaults to light and dark modes making theme selections cleaner, simpler, and more user-friendly. *** We’re always working to make AVA better for you, and your feedback drives us forward. Thanks for being part of the journey—more great updates are just around the corner! 🚀 # AVA 2.6.0 Source: https://docs.datarm.ai/changelog/About AVA 2.6.0 Release notes for AVA version 2.6.0 # Release Notes for AVA Update 2.6.0 Small but mighty—AVA 2.6.0 delivers targeted enhancements to boost your productivity and streamline your experience. *** ### **Highlights** * **Refined Dark/Light Modes:** Additional tweaks for even smoother, more comfortable theme transitions. * **Salesforce @ Mentions Enhanced:** Enjoy a major boost in speed and reliability when using @ mentions with Salesforce. * **Smarter Microsoft @ Mentions:** The old `@Microsoft` mention has been replaced with focused options—call out calendar, email, Teams, Microsoft Defender, or perform a general search across people, files, and SharePoint sites with better control. * **Brand-New `@Web` Mention:** Instantly search the web for up-to-date information, directly in your chats and tasks. * **New Reasoning Models in Tasks:** Access the latest advanced reasoning models in Tasks for deeper analysis and smarter outcomes. *** Thanks for helping us shape AVA into the next must-have business productivity tool. Your feedback keeps us moving forward—stay tuned for what’s next! 🚀 # AVA 2.6.3 Source: https://docs.datarm.ai/changelog/About AVA 2.6.3 Release notes for AVA version 2.6.3 # Release Notes for AVA Update 2.6.3 Small but mighty—AVA 2.6.3 delivers targeted enhancements to boost your productivity and streamline your experience. *** ### **Highlights** * **Refined Dark/Light Modes:** Additional tweaks for even smoother, more comfortable theme transitions. * **Salesforce @ Mentions Enhanced:** Enjoy a major boost in speed and reliability when using @ mentions with Salesforce. * **Smarter Microsoft @ Mentions:** The old `@Microsoft` mention has been replaced with focused options—call out calendar, email, Teams, Microsoft Defender, or perform a general search across people, files, and SharePoint sites with better control. * **Brand-New `@Web` Mention:** Instantly search the web for up-to-date information, directly in your chats and tasks. * **New Reasoning Models in Tasks:** Access the latest advanced reasoning models in Tasks for deeper analysis and smarter outcomes. *** Thanks for helping us shape AVA into the next must-have business productivity tool. Your feedback keeps us moving forward—stay tuned for what’s next! 🚀 # AVA 2.7.0 Source: https://docs.datarm.ai/changelog/About AVA 2.7.0 Release notes for AVA 2.7.0 # AVA 2.7.0 **Release Date:** September 5, 2025 ## Overview AVA 2.7.0 introduces comprehensive GDPR automation features, enhanced third-party risk management, and significant improvements to the data catalog. This release strengthens privacy compliance capabilities and regulatory reporting. *** ## New Features * **Data Subject Request Management**: Automated workflows for access, rectification, erasure, and portability requests * **Consent Management**: Track and manage user consent across all systems * **Right to be Forgotten**: Automated data deletion workflows with verification * **Data Processing Records (Article 30)**: Maintain GDPR-compliant processing activity records * **Privacy Impact Assessments**: Guided DPIA workflows with templates * **Breach Notification**: 72-hour breach notification workflow automation * **Vendor Risk Scoring**: Automated risk assessment for vendors and suppliers * **Contract Management**: Store and track data processing agreements (DPAs) * **Continuous Monitoring**: Ongoing vendor risk evaluation with alerts * **Due Diligence Questionnaires**: Customizable questionnaire templates * **Risk Register**: Centralized third-party risk inventory * **Vendor Portal**: Self-service portal for vendor assessments * **Smart Recommendations**: AI-powered suggestions for data classification * **Data Dictionary**: Centralized business glossary with term relationships * **Advanced Search Filters**: 20+ new filter options for precise discovery * **Catalog Analytics**: Usage insights and data asset popularity metrics * **Collaborative Curation**: Team-based data stewardship with comments * **API Documentation**: Auto-generated API documentation from metadata * **Pre-built Templates**: 40+ regulatory report templates (GDPR, CCPA, HIPAA, SOC 2) * **Schedule and Distribute**: Automated report generation and distribution * **Evidence Collection**: Automatic gathering of compliance evidence * **Audit Trails**: Comprehensive logs for regulatory audits * **Executive Dashboards**: C-level compliance status visualization *** ## Improvements * 50% faster scanning for large databases * Support for additional data sources: Oracle, SAP HANA * Improved pattern recognition for PII detection * Incremental scanning to reduce resource usage * Fine-grained permissions at the column level * Temporary access grants with auto-expiration * Access request workflow improvements * Integration with ServiceNow for approvals * Customizable notification templates * Digest mode for reduced email volume * Integration with Slack and Microsoft Teams * Mobile push notifications (iOS and Android) * Immutable audit logs with blockchain verification * Enhanced search and filtering in audit viewer * Compliance score calculation improvements * Automated evidence package creation *** ## Bug Fixes Fixed issue where data classification rules would not apply to newly discovered assets Resolved performance degradation in dashboard loading for users with 1000+ assets Corrected calculation of retention policy expiration dates Fixed duplicate notifications being sent for the same event Addressed SSO timeout issues with Okta integration Resolved CSV import failures for files with BOM (Byte Order Mark) Fixed incorrect user attribution in audit logs for API operations Corrected display issues in risk heatmaps for certain screen resolutions *** ## API Changes New API endpoints introduced. All existing endpoints remain backward compatible. ### New Endpoints * `POST /api/v2/dsr/requests` - Create data subject requests * `GET /api/v2/vendors/{id}/risk-score` - Retrieve vendor risk assessment * `POST /api/v2/catalog/recommend` - Get AI-powered classification recommendations * `GET /api/v2/compliance/reports/{type}` - Generate compliance reports ### Enhanced Endpoints * `GET /api/v2/assets` - New filters for consent status and GDPR classification * `POST /api/v2/policies/evaluate` - Now supports batch evaluation *** ## Security Enhancements * TLS 1.3 now required for all connections * Enhanced API rate limiting with IP-based throttling * Added support for Hardware Security Modules (HSM) for key storage * Improved secrets management with HashiCorp Vault integration * Security headers updated to align with OWASP recommendations *** ## Breaking Changes **PostgreSQL 12 End of Support**: PostgreSQL 12 is no longer supported. Upgrade to PostgreSQL 13 or higher before installing this version. *** ## Migration Notes * Estimated migration time: 5-10 minutes for typical deployments * New indexes will be created for performance optimization * Backup your database before upgrading * Review GDPR settings in Admin Console after upgrade * Configure notification channels for DSR workflows * Update retention policies if using legacy format *** ## Documentation * [GDPR Automation Guide](/compliance/gdpr) * [Third-Party Risk Management](/guides/tprm) * [Data Catalog User Guide](/guides/data-catalog) * [API Reference](/api-reference) # AVA 2.7.1 Source: https://docs.datarm.ai/changelog/About AVA 2.7.1 Release notes for AVA 2.7.1 # AVA 2.7.1 **Release Date:** September 22, 2025 ## Overview AVA 2.7.1 is a maintenance release addressing critical bugs and performance issues identified in version 2.7.0. This release includes important stability improvements and security patches. *** ## Bug Fixes **Critical**: Fixed data corruption issue in PostgreSQL metadata store during concurrent writes **Critical**: Resolved authentication bypass vulnerability in API v2 endpoints (CVE-2024-XXXXX) Corrected calculation errors in compliance score aggregation Fixed memory leak in background data discovery jobs Resolved UI freezing issue when viewing large policy lists (500+ policies) Addressed incorrect timezone handling in audit log timestamps Fixed broken links in email notification templates Corrected SAML SSO redirect issues with certain identity providers Resolved data export failures for datasets with null values in certain fields *** ## Performance Improvements * Improved index usage for metadata queries (40% faster) * Optimized connection pooling for high-concurrency scenarios * Reduced lock contention in write operations * Fixed rendering delays for data catalog pages * Improved pagination performance for large result sets * Reduced initial page load time by 25% *** ## Security Updates **Important**: This release includes critical security patches. Immediate upgrade strongly recommended. * Patched authentication bypass vulnerability (CVSS 9.1) * Updated OpenSSL to version 3.0.11 * Enhanced input sanitization for all user-supplied data * Improved password policy enforcement *** ## Configuration Changes No configuration changes required for this release. Upgrade is backward compatible with 2.7.0. *** ## Known Issues * Data lineage visualization may not display correctly in Firefox ESR (workaround available in KB article #1247) * Scheduled reports may be delayed by up to 5 minutes during peak hours (optimization planned for 2.8.0) *** ## Upgrade Instructions ```bash theme={null} docker pull datarm/ava:2.7.1 docker-compose down docker-compose up -d ``` ```bash theme={null} helm upgrade ava datarm/ava --version 2.7.1 kubectl rollout status deployment/ava ``` 1. Download the installer from the customer portal 2. Run the upgrade script with administrator privileges 3. Restart the AVA services 4. Verify the upgrade in Settings > About *** ## Support For assistance with this upgrade, contact: * **Email**: [avasupport@datarm.ai](mailto:avasupport@datarm.ai) * **Portal**: [https://support.datarm.ai](https://support.datarm.ai) * **Emergency**: Available for Enterprise tier customers # AVA 2.8.0 Source: https://docs.datarm.ai/changelog/About AVA 2.8.0 Release notes for AVA 2.8.0 # AVA 2.8.0 **Release Date:** October 10, 2025 ## Overview AVA 2.8.0 delivers enhanced integration capabilities, advanced data lineage visualization, and improved governance workflows. This release emphasizes enterprise connectivity and streamlined compliance operations. *** ## New Features * **Column-level Lineage**: Trace data transformations at the column level across entire data pipelines * **Impact Analysis**: Instantly visualize downstream impacts of schema or data changes * **Interactive Lineage Graph**: Zoom, filter, and explore complex data flows with improved UI * **Lineage Export**: Export lineage diagrams as SVG, PNG, or interactive HTML * **Historical Lineage**: Track how data flows have changed over time * **SharePoint Connector**: Scan and classify documents in SharePoint Online * **Teams Integration**: Governance notifications and approvals directly in Microsoft Teams * **OneDrive Scanning**: Automatic discovery of sensitive data in OneDrive for Business * **Outlook Add-in**: Data classification and DLP warnings in email composition * **Power BI Integration**: Governance metadata for Power BI datasets and reports * **Visual Workflow Designer**: Build complex governance workflows with drag-and-drop * **Conditional Logic**: Create sophisticated approval chains with branching logic * **SLA Management**: Define and track service level agreements for governance requests * **Workflow Templates**: Pre-built templates for common governance scenarios * **Parallel Approvals**: Support for concurrent approval paths * **Automated Quality Checks**: Schedule and run data quality validations * **Quality Metrics Dashboard**: Monitor completeness, accuracy, consistency, and timeliness * **Anomaly Detection**: AI-driven detection of data quality issues * **Quality Alerts**: Configurable notifications for quality threshold violations * **Remediation Tracking**: Track data quality improvement initiatives *** ## Improvements * Policy versioning with rollback capabilities * Bulk policy application to multiple assets * Policy effectiveness analytics * Template library with 50+ pre-built policies * 30+ new compliance report templates * Custom report builder with drag-and-drop * Scheduled report distribution via email * Export to PDF, Excel, and PowerPoint formats * REST API performance improvements (35% faster) * New connectors: Snowflake, Databricks, MongoDB * Enhanced error handling with detailed diagnostics * Connector health monitoring dashboard * Role-based access control granularity improvements * User activity analytics and insights * Self-service user onboarding workflows * Enhanced group management with nested groups *** ## Bug Fixes Fixed issue where large metadata imports would timeout Resolved display issues in Safari for data catalog search Corrected permission inheritance for nested organizational units Fixed duplicate entries in data steward assignment lists Addressed slow loading times for assets with extensive metadata Resolved export failures for reports with more than 100,000 rows Fixed incorrect date range calculations in compliance dashboards *** ## Security Updates This release includes important security patches. Update recommended for all users. * Updated dependencies to address CVE-2024-XXXX * Enhanced input validation for API endpoints * Improved session management and token rotation * Additional logging for security-relevant events *** ## Deprecation Notice The legacy metadata import API will be deprecated in AVA 3.0. Please migrate to the new bulk import API introduced in this version. *** ## Known Issues * Lineage visualization may be slow for graphs with more than 5,000 nodes (optimization planned for 2.9.0) * Teams integration requires admin consent for certain tenants * Power BI connector requires Power BI Pro or Premium licenses *** ## Upgrade Notes Database migrations may take 15-30 minutes for large deployments. Schedule during maintenance windows. # AVA 2.9.0 Source: https://docs.datarm.ai/changelog/About AVA 2.9.0 Release notes for AVA 2.9.0 # AVA 2.9.0 **Release Date:** November 15, 2025 ## Overview AVA 2.9.0 introduces advanced AI-powered data classification, enhanced risk scoring algorithms, and comprehensive DORA compliance features. This release focuses on automation, regulatory compliance, and enterprise-scale performance improvements. *** ## New Features * **Automated Data Discovery**: Machine learning models automatically identify and classify sensitive data across structured and unstructured sources * **Custom Classification Models**: Train organization-specific classifiers using your own data patterns * **Confidence Scoring**: Each classification includes a confidence score for audit and review purposes * **Real-time Classification**: Classify data as it enters your systems with sub-second latency * **Multi-language Support**: Enhanced NLP capabilities for 40+ languages * **ICT Risk Management**: Built-in templates and workflows aligned with EU Digital Operational Resilience Act * **Third-Party Risk Assessment**: Automated vendor risk scoring for critical ICT service providers * **Incident Reporting**: DORA-compliant incident classification and reporting workflows * **Testing Framework**: Structured approach to operational resilience testing * **Regulatory Dashboards**: Real-time compliance status monitoring * **Multi-dimensional Risk Analysis**: Evaluate risk across 12+ dimensions including data sensitivity, access patterns, and threat intelligence * **Predictive Risk Modeling**: AI-driven predictions of emerging risks based on historical patterns * **Custom Risk Weights**: Configure risk calculation weights specific to your organization * **Risk Trends**: Visualize risk evolution over time with interactive charts * **Automated Risk Remediation**: Suggested actions based on risk severity * **Semantic Search**: Find conceptually similar content even without exact keyword matches * **Multi-modal Search**: Search across documents, images, and structured data simultaneously * **Search Analytics**: Track search patterns to identify knowledge gaps * **Relevance Tuning**: Customize search ranking based on organizational priorities *** ## Improvements * 40% faster dashboard loading times * Improved query optimization for large datasets (10M+ records) * Reduced memory footprint by 25% * Enhanced caching strategies for frequently accessed data * Redesigned navigation with improved accessibility (WCAG 2.1 AA) * Contextual help integrated throughout the application * Customizable dashboard layouts with drag-and-drop * Dark mode performance optimizations * GraphQL API now supports field-level permissions * Improved rate limiting with burst capacity * Enhanced webhook delivery with automatic retries * New batch operations for bulk data processing * Support for FIDO2/WebAuthn passwordless authentication * Enhanced audit logging with tamper detection * Improved encryption for data at rest (AES-256-GCM) * Security headers aligned with OWASP best practices *** ## Bug Fixes Fixed issue where data lineage graphs would not render for circular dependencies Resolved memory leak in real-time dashboard updates for long-running sessions Corrected timezone handling in scheduled report generation Fixed race condition in concurrent policy evaluations Addressed CSV export formatting issues with special characters Resolved SSO session timeout inconsistencies with Azure AD Fixed notification delivery delays for high-priority alerts *** ## Breaking Changes **API Version Update**: The legacy v1 API endpoints are now deprecated and will be removed in AVA 3.0 (Q2 2025). Migrate to v2 API endpoints before March 2025. **Risk Score Calculation**: The risk scoring algorithm has been updated. Historical risk scores remain unchanged, but new calculations may differ from previous versions. Review custom risk policies if you have specific thresholds configured. *** ## Migration Guide For detailed migration instructions, see our [Migration Guide](/guides/migration/2-9-0) or contact support at [avasupport@datarm.ai](mailto:avasupport@datarm.ai) *** ## What's Next Looking ahead to AVA 2.10.0 (December 2025): * AI-powered policy recommendations * Enhanced data quality scoring * Advanced workflow automation * Integration with Microsoft Purview # Azure Deployment Architecture Source: https://docs.datarm.ai/deployment/architecture Understanding AVA deployment in your Azure tenant ## Overview AVA is deployed exclusively as a **single-tenant application within your Azure tenant**. Unlike traditional SaaS offerings, AVA runs entirely in your Azure environment, giving you complete control over data residency, security, and compliance. AVA Architecture Diagram **Single-Tenant Only**: AVA is never deployed as a multi-tenant SaaS. Each customer gets their own isolated deployment in their Azure tenant. ## Core Azure Components AVA leverages managed Azure services to provide a secure, scalable, and maintainable enterprise AI platform: Frontend React application hosting Backend Python services and MCP servers Primary data store with pgVector for embeddings Secrets management and configuration AI model deployment and orchestration Identity, authentication, and authorization ## Architecture Layers ### 1. Identity & Access Layer **Microsoft Entra ID** provides the foundation for security: * **Enterprise App**: Registered in your Entra ID for SSO * **App Registration**: Defines API permissions and scopes * **Managed Identity**: Backend services use managed identity to access Key Vault * **Delegated Permissions**: AVA operates with user permissions (no elevated access) AVA Identity Flow ### 2. Frontend Layer **Azure App Service** hosts the React-based web application: * Static web hosting with CDN integration * TLS/SSL termination * Custom domain support (e.g., `ava.yourcompany.com`) * Automatic scaling based on user load AVA Frontend Interface ### 3. Backend Services Layer **Azure Container Apps** runs the Python-based backend services: * RESTful API for frontend communication * Business logic orchestration * AI model interaction * Task execution engine * Knowledge Search processing **Key features**: * Auto-scaling based on CPU/memory/request metrics * Zero-downtime deployments * Built-in load balancing * Integrated with Azure Monitor for observability AVA Chat Interface ### 4. MCP Server Layer **Model Context Protocol (MCP) servers** run as dedicated Container Apps: Each integration (Microsoft 365, Salesforce, Jira, GitHub, NetSuite, etc.) is deployed as a separate container app that: * Uses your existing SSO configuration per application * Operates with delegated user permissions * Scales independently based on usage * Communicates securely with backend services AVA Connected Data Sources **Example MCP Servers**: * `mcp-microsoft365`: Calendar, Mail, Files, Teams integration * `mcp-salesforce`: CRM data access * `mcp-jira`: Project management * `mcp-github`: Repository and code access * `mcp-netsuite`: Business records ### 5. Data Layer **Azure PostgreSQL Flexible Server** with pgVector extension: * Primary application database * Vector embeddings storage for Knowledge Search (RAG) * High availability with zone redundancy * Automated backups with point-in-time restore * Private endpoint for network isolation AVA Tasks Interface ### 6. AI Provider Layer **Azure AI Foundry** (and optional AWS Bedrock, GCP Vertex): * Model deployment and hosting * Token usage tracking * Content filtering * Multi-model support (GPT-5 series, GPT-image-1, Claude 4 series, Gemini 2.5, etc.) * Cost optimization through model selection AVA Model Selection ## Security Architecture ### Managed Identity Pattern AVA backend services use **Azure Managed Identity** to access resources: ``` Backend Container Apps → Managed Identity → Azure Key Vault → Secrets ↓ Azure PostgreSQL (via managed identity) ↓ Azure AI Foundry (via managed identity) ``` **Benefits**: * No credentials stored in code or configuration * Automatic credential rotation * Azure AD authentication for all services * Audit trail through Azure Monitor ### Network Security * **Private Endpoints**: Database and Key Vault accessible only within VNet * **No Public Internet Exposure**: Backend services communicate via internal networking * **TLS Everywhere**: All communication encrypted in transit * **Azure DDoS Protection**: Standard protection for public-facing services ### Data Protection * **Encryption at Rest**: Azure Storage/Database encryption enabled by default * **Encryption in Transit**: TLS 1.3 for all communications * **Key Management**: All keys stored in Azure Key Vault with FIPS 140-2 compliance * **Data Residency**: Data never leaves your Azure tenant AVA Security Features ## Deployment Sizing AVA deployment is **right-sized to your organization** based on: * Number of users * Expected usage patterns * Connected data sources * Knowledge Search corpus size * Performance requirements ### Small Deployment (\< 100 users) * **App Service**: P1V3 (2 vCPU, 8 GB RAM) * **Container Apps**: * Backend: 1-2 replicas, 0.5 vCPU, 1 GB RAM each * MCP servers: 1 replica each, 0.25 vCPU, 0.5 GB RAM * **PostgreSQL**: B\_Standard\_B1ms (1 vCPU, 2 GB RAM, 32 GB storage) ### Medium Deployment (100-500 users) * **App Service**: P1V3 (2 vCPU, 8 GB RAM) * **Container Apps**: * Backend: 2-4 replicas, 0.5 vCPU, 1 GB RAM each * MCP servers: 1-2 replicas each, 0.5 vCPU, 1 GB RAM * **PostgreSQL**: B\_Standard\_B2s (2 vCPU, 4 GB RAM, 64 GB storage) ### Large Deployment (500+ users) * **App Service**: P2V3 (4 vCPU, 16 GB RAM) with autoscale * **Container Apps**: * Backend: 4-8 replicas, 1 vCPU, 2 GB RAM each * MCP servers: 2-3 replicas each, 0.5 vCPU, 1 GB RAM * **PostgreSQL**: GP\_Standard\_D2s\_v3 (2 vCPU, 8 GB RAM, 128 GB storage) **Performance Tuning**: All deployment sizes can be scaled up or down based on actual usage patterns and performance requirements. ## Scaling Strategy ### Horizontal Scaling AVA automatically scales based on metrics: * **Frontend (App Service)**: CPU and memory-based autoscale rules * **Backend (Container Apps)**: Request count, CPU, memory triggers * **MCP Servers**: Independent scaling per integration based on usage ### Performance Optimization As usage grows, AVA can be optimized by: 1. **Increasing Container App Replicas**: More concurrent request handling 2. **Scaling Database Tier**: Better query performance for large datasets 3. **Adding Read Replicas**: Offload read traffic from primary database 4. **Optimizing AI Model Selection**: Balance cost vs performance per use case ## Monitoring & Observability All Azure resources integrate with **Azure Monitor**: * Application Insights for frontend/backend telemetry * Container Apps logs and metrics * Database performance metrics * Key Vault access logs * Cost analysis and optimization recommendations ## Deployment Process * Azure subscription with Owner or Contributor role * Microsoft Entra ID admin access * Custom domain (optional but recommended) DataRM provisions all Azure resources in your tenant using Infrastructure as Code (Bicep and ARM templates): * Resource Group creation * App Service and Container Apps * PostgreSQL database with pgVector * Key Vault for secrets * Managed identities and role assignments Configure Microsoft Entra ID: * Create Enterprise Application * Register App Registration with API permissions * Configure delegated permissions * Assign users/groups Deploy and configure MCP servers: * Microsoft 365 (automatic for Entra users) * Salesforce (requires Salesforce Connected App) * Jira (requires Jira OAuth app) * GitHub (requires GitHub App) * Other integrations as needed Connect to AI providers: * Azure AI Foundry (primary) * AWS Bedrock (optional) * GCP Vertex (optional) * Configure model access and quotas * User authentication testing * Data source connectivity verification * Performance baseline testing * Security configuration audit * DNS configuration for custom domain * User onboarding and training * Monitoring and alerting setup * Handoff to customer admin team ## Maintenance & Updates ### Automated Updates AVA backend services (Container Apps) support **blue-green deployment**: * New version deployed alongside existing version * Traffic gradually shifted to new version * Zero-downtime updates * Automatic rollback on errors ### Database Migrations * Schema changes applied via automated migration scripts * Backward-compatible migrations ensure zero downtime * Point-in-time restore available for rollback ### Security Patching * Azure platform handles OS-level patching * Container images rebuilt monthly with security updates * Critical vulnerabilities patched within 48 hours ## Cost Optimization ### Estimated Monthly Cost **AVA Pricing**: Monthly costs range from **$500 to $2,500** depending on your organization size and number of users. Pricing includes all Azure infrastructure, AI model usage, and platform features. **Cost varies based on:** * Number of active users * AI model usage patterns * Connected data sources * Knowledge Search corpus size * Performance requirements **AI Token Costs**: The largest variable cost is AI model usage. Optimize by choosing appropriate models per task (e.g., GPT-5 Mini/Nano, Claude Haiku 4.5, or Gemini 2.5 Flash-Lite for simple tasks, GPT-5 Pro/Claude Opus 4.1 for complex reasoning). ### Optimization Strategies 1. **Auto-scaling**: Scale down during off-hours 2. **Reserved Instances**: 1-year commitment saves 30-40% 3. **Model Selection**: Use GPT-5 Mini/Nano, Flash, or Haiku models for routine tasks; reserve GPT-5 Pro/Opus for complex analysis 4. **Corpus Optimization**: Limit Knowledge Search corpora to essential documents 5. **Resource Right-Sizing**: Monitor and adjust compute resources based on actual usage ## Disaster Recovery ### Backup Strategy * **Database**: Automated daily backups with 7-day retention (configurable to 35 days) * **Point-in-Time Restore**: Restore to any point within retention window * **Configuration Backup**: Key Vault secrets backed up to separate storage account * **Code Repository**: Infrastructure as Code stored in source control ### High Availability * **Zone Redundancy**: PostgreSQL and Container Apps deployed across availability zones * **Automatic Failover**: Database failover in \< 2 minutes * **Load Balancing**: Built-in load balancing for Container Apps * **Health Checks**: Automatic unhealthy instance replacement ### Business Continuity * **RTO (Recovery Time Objective)**: \< 4 hours * **RPO (Recovery Point Objective)**: \< 15 minutes * **Geo-Replication**: Optional for enterprise customers (additional cost) ## Compliance & Security ### Leverage Your Existing Azure Certifications **Inherit Your Tenant's Compliance**: Because AVA runs entirely within your Azure tenant, you automatically inherit all existing Azure certifications and compliance frameworks your organization has already achieved. **Key Benefits:** * **No Separate Compliance Audit**: AVA lives within your already-certified Azure environment * **Your Security Policies**: All existing Azure security policies, governance, and controls apply to AVA * **Data Residency**: Data stays within your tenant's geographic boundaries and compliance zones * **Existing Certifications**: Leverage your Azure tenant's SOC 2, ISO 27001, HIPAA, FedRAMP, GDPR, PCI DSS, and other certifications **How It Works:** 1. **Single-Tenant Architecture**: AVA deploys as resources in your Azure subscription 2. **Azure's Shared Responsibility Model**: Infrastructure compliance handled by Azure, application security configured during deployment 3. **Your Governance**: AVA respects all Azure policies, RBAC, and compliance frameworks you've implemented 4. **Audit Trail**: All AVA activity logged through your existing Azure Monitor and Application Insights **Simplified Compliance**: By running in your Azure tenant, AVA doesn't introduce new compliance challenges—it operates within your already-certified cloud environment. ## Next Steps Detailed technical flow and MCP integration Deep dive into security architecture Request deployment for your organization Learn how to use AVA after deployment # Security Architecture Source: https://docs.datarm.ai/deployment/security Understanding AVA security model and delegated permissions ## Security Philosophy AVA's security model is built on a fundamental principle: **delegated user permissions**. AVA never operates with elevated privileges or service accounts. Instead, it uses your Microsoft Entra ID identity and accesses only what you can access. **Zero Trust by Design**: Every action AVA performs is authenticated through your Entra ID identity with delegated permissions. No backdoors, no elevated access, no shadow IT. ## Identity & Access Management ### Microsoft Entra ID Integration AVA leverages your existing Microsoft Entra ID (formerly Azure AD) for all authentication and authorization: **Single Sign-On (SSO)**: * Users sign in with their corporate Microsoft account * Multi-factor authentication (MFA) enforced if configured in Entra ID * Conditional Access policies honored * Session management controlled by Entra ID policies **No Separate Credentials**: * No passwords stored in AVA * No separate user database to manage * User provisioning/deprovisioning managed via Entra ID **How It Works**: 1. User authenticates with Entra ID 2. AVA requests delegated permissions (OAuth 2.0) 3. User grants consent (one-time) 4. AVA receives access token with user's permissions 5. All API calls use user's identity **What This Means**: * AVA can only access files the user can access * AVA can only read emails the user can read * AVA can only query Salesforce data the user can query * Audit logs show the actual user, not a service account **Enterprise Application**: * Registered in your Entra ID * Visible in Azure Portal → Enterprise Applications * Permissions visible and auditable * Can be disabled/removed by admins **Required Permissions** (all delegated, not application): * Microsoft Graph API: User.Read, Mail.Read, Calendars.Read, Files.Read * Additional permissions per connected MCP server * All permissions require user consent or admin consent ## Data Security ### Encryption **Azure Storage Encryption**: * All data encrypted with AES-256 * PostgreSQL encrypted at rest * Encryption keys managed by Azure Key Vault * FIPS 140-2 validated encryption modules **TLS 1.3 Everywhere**: * All API communication encrypted * Frontend ↔ Backend: TLS 1.3 * Backend ↔ Database: TLS 1.3 * MCP Servers ↔ External APIs: TLS 1.3 **Azure Key Vault**: * All secrets stored in Key Vault * No hardcoded credentials * Automatic secret rotation * Access logged and auditable **Your Tenant Only**: * Data never leaves your Azure tenant * No cross-tenant data sharing * Compliance with data residency requirements * Full control over data location ### Data Protection **What AVA Stores**: * User session metadata (who logged in, when) * Conversation history (chat messages, task definitions) * Knowledge Search corpora metadata and embeddings * User preferences and custom instructions * Audit logs **What AVA Does NOT Store**: * User passwords (handled by Entra ID) * Raw email or calendar data (fetched on-demand via delegated permissions) * Salesforce/Jira/GitHub data (fetched on-demand via MCP servers) * Any data AVA doesn't need to function **Data Retention**: * Conversation history: Configurable (default 90 days, can be set to unlimited) * Knowledge Search embeddings: Until corpus is deleted * Audit logs: Configurable (default 365 days, can export to Azure Storage) * Session tokens: Expire per Entra ID policy (typically 1-24 hours) ## Network Security ### Architecture ``` Internet → Azure Front Door (DDoS Protection) ↓ App Service (Frontend - Public) ↓ (TLS) Container Apps (Backend - Private VNet) ↓ (Private Endpoint) PostgreSQL (Database - Private VNet) ↓ (Managed Identity) Key Vault (Secrets - Private VNet) ``` ### Network Isolation * **PostgreSQL**: Accessible only within VNet, no public endpoint * **Key Vault**: Private endpoint for secret access * **Container Apps**: Backend services in private VNet subnet * **No Direct Database Access**: Database not exposed to internet * All backend services deployed in customer's VNet * Network Security Groups (NSGs) control traffic * Azure Firewall optional for additional control * VNet peering with customer networks supported * Azure DDoS Protection Standard enabled * Automatic mitigation of layer 3/4 attacks * Rate limiting at application layer * WAF rules for common attack patterns **Inbound**: * Users → Azure Front Door → App Service (HTTPS only) * App Service → Container Apps (internal HTTPS) **Outbound** (from backend): * Container Apps → Microsoft Graph API (via delegated permissions) * Container Apps → MCP Servers (internal, same VNet) * MCP Servers → External APIs (Salesforce, Jira, etc. via HTTPS) * All outbound traffic logged ## Managed Identity & RBAC ### How AVA Backend Accesses Resources AVA backend services use **Azure Managed Identity** instead of credentials: Each Container App is assigned a system-managed identity automatically by Azure The managed identity is granted specific Azure RBAC roles: * **Key Vault Secrets User**: Read secrets from Key Vault * **PostgreSQL Database User**: Connect to database * **Storage Blob Data Contributor**: (if using Azure Storage for exports) When backend needs to access a resource: 1. Requests access token from Azure Instance Metadata Service 2. Azure validates the managed identity 3. Returns short-lived access token 4. Backend uses token to access resource **Benefits**: * No passwords or keys in configuration * Automatic credential rotation * Audit trail via Azure AD logs * Cannot be compromised or leaked ### User-Level RBAC Within AVA, access is controlled by: 1. **Microsoft 365 Permissions**: Users see only their own email, calendar, files 2. **Salesforce Permissions**: Delegated access respects Salesforce sharing rules 3. **Jira Permissions**: Users see only projects/issues they have access to 4. **GitHub Permissions**: Repository access follows GitHub permissions 5. **Knowledge Search**: Users can only search corpora they're granted access to **No Shadow RBAC**: AVA doesn't create a separate permission system. It uses your existing access controls in each system. ## Audit & Compliance ### Audit Logging All actions in AVA are logged with: * **Who**: User identity (UPN from Entra ID) * **What**: Action performed (query, export, corpus access, etc.) * **When**: Timestamp (UTC) * **Where**: Source IP, device info * **Result**: Success/failure, error codes **Log Destinations**: * Azure Monitor / Application Insights * Azure Storage (for long-term retention) * Azure Sentinel (for SIEM integration) * Export to customer's log management system ### Compliance Frameworks AVA supports compliance with: Security, availability, confidentiality controls Information security management system EU data protection and privacy Healthcare data security (via Azure compliance) Federal government (via Azure Government) Payment card data security (via Azure) **Shared Responsibility Model**: Azure provides infrastructure compliance (physical security, network, encryption). AVA provides application-level compliance (access controls, audit logs, data handling). Customer is responsible for user access management and data classification. ### Data Subject Rights (GDPR) AVA supports GDPR data subject rights: * **Right to Access**: Export all user data via admin API * **Right to Erasure**: Delete user and all associated data * **Right to Portability**: Export data in machine-readable format (JSON) * **Right to Rectification**: Users can edit/delete their conversations * **Data Retention Policies**: Configurable retention periods ## Security Best Practices ### For Administrators Configure in Entra ID: * Require MFA for all users accessing AVA * Use Conditional Access policies * Consider hardware tokens for privileged users * Audit Enterprise Application permissions quarterly * Review user access to AVA * Check MCP server configurations * Validate Knowledge Search corpus permissions * Set up alerts for suspicious activity * Review failed login attempts * Monitor data export events * Check for unusual query patterns * Only create corpora for documents users should access * Don't include sensitive documents unless necessary * Use corpus-level access controls * Regularly review corpus contents Entra ID Conditional Access policies apply to AVA: * Require MFA from untrusted locations * Block legacy authentication * Require compliant devices * Restrict access by IP range ### For End Users * AVA stores conversation history * Don't paste passwords, API keys, or secrets * Be mindful of PHI, PII, financial data * Use Knowledge Search for sensitive documents (more controlled) * Review PowerPoint/Word exports before sharing externally * Exports may contain aggregated data from multiple sources * Check for sensitive information before sending * Custom instructions are stored in database * Don't include sensitive information * Keep instructions professional and task-focused * Contact your IT admin if you see unusual behavior * Report suspected unauthorized access * Security team email: [avasupport@datarm.ai](mailto:avasupport@datarm.ai) ## Incident Response ### Security Incident Handling * Azure Monitor alerts trigger automatically * Security Operations Center (SOC) monitoring * User reports via support channel * Review audit logs and telemetry * Determine scope and impact * Identify affected users/data * Disable compromised accounts (via Entra ID) * Revoke access tokens * Isolate affected Container Apps if needed * Block IP addresses if attack detected * Patch vulnerabilities * Reset credentials if needed * Restore from backup if data corruption * Update security configurations * Notify affected users within 72 hours (GDPR) * Report to management/board if major incident * Document incident in compliance logs * Post-mortem and lessons learned ### Breach Notification In the unlikely event of a data breach: * **Timeline**: Notification within 72 hours of discovery (GDPR requirement) * **Channels**: Email to admin contacts, in-app notification * **Information Provided**: What data affected, timeline, remediation steps * **Support**: Dedicated incident support team assigned ## Security Certifications AVA maintains the following security certifications: * **SOC 2 Type II**: Annual audit by independent CPA firm * **ISO 27001**: Information Security Management System certification * **Penetration Testing**: Quarterly by third-party security firm * **Vulnerability Scanning**: Continuous automated scanning * **Responsible Disclosure Program**: [security@datarm.ai](mailto:security@datarm.ai) **Request Security Documentation**: Contact [avasupport@datarm.ai](mailto:avasupport@datarm.ai) to receive: * SOC 2 Type II report (requires NDA) * ISO 27001 certificate * Penetration test summary * Security whitepaper ## Common Security Questions **No.** AVA uses delegated permissions. It can only access what you can access with your Microsoft work account. **No.** AVA fetches email data on-demand via Microsoft Graph API using your permissions. Only the AI's response is stored in conversation history. **No by default.** Conversations are private to each user. However, admins can enable organization-wide audit logging to comply with legal/regulatory requirements. AVA sessions are tied to your Entra ID authentication. If you report device theft, IT admin can: * Revoke device access in Entra ID * Force sign-out from all sessions * Your conversations remain secure in Azure **Key Differences**: * AVA data never leaves your Azure tenant * Uses your identity and permissions * No training on your data * Enterprise-grade security and compliance * Audit logs and governance controls **No.** AVA uses AI models (GPT-4, Claude, etc.) but your data is never used to train or improve these models. All prompts are sent with opt-out flags to prevent training. ## Next Steps Understand AVA deployment architecture Technical flow and MCP integration Monitor and review security events Questions about security? Email us # Core Capabilities Source: https://docs.datarm.ai/essentials/core-capabilities Powerful features available across all of AVA—AI that works, not just talks # Core Capabilities: AI That Works™ AVA isn't just another AI chatbot—it's your AI super assistant that **completes real work** across your Microsoft 365 environment. These core capabilities are available everywhere in AVA, whether you're in Chat, Tasks, or Knowledge Search. **The Power of Microsoft Graph API**: AVA leverages Microsoft Graph API to take action on your behalf—scheduling meetings, creating tasks, sharing files, and much more. The possibilities are nearly endless. ## Real-Time Web Search with @Web Access current information from the web in real-time—perfect for research, news, and up-to-date data. ### How @Web Works **Get current data AVA's training doesn't include** **Examples:** * "**@Web** what are the latest developments in AI?" * "**@Web** find recent articles about GPT-5" * "**@Web** current stock price of Microsoft" * "**@Web** today's weather forecast for Seattle" **What AVA does:** * Searches the web in real-time * Finds current, relevant information * Summarizes results with source citations * Provides links to original sources **Stay informed with latest information** **Use cases:** * Market research and competitor analysis * News monitoring and current events * Technical documentation and tutorials * Product information and reviews * Academic research and papers **Example:** "**@Web** find the latest research papers on quantum computing from 2025" **Mix web data with your enterprise data** **Example:** 1. "**@Web** find latest industry trends in SaaS" 2. "**@Salesforce** show our pipeline for this quarter" 3. "Compare our positioning to industry trends" AVA combines web research with your business data for powerful insights. Use **@Web** when you need information beyond AVA's training data or your organizational data—news, current events, real-time data, and recent developments. ## Work with Images Add visual context to your conversations or generate new images—AVA handles both. **Upload images to provide visual context** **What you can do:** * Upload screenshots for troubleshooting * Add diagrams for discussion * Include photos for analysis * Attach images from your computer **How it works:** 1. Click the image/attachment icon in any chat or task 2. Select an image from your computer 3. The image is added to the conversation 4. AVA can analyze, describe, or reference the image **Use cases:** * **Design feedback**: "Review this mockup and suggest improvements" * **Error troubleshooting**: "What's wrong in this error screenshot?" * **Data visualization**: "Analyze the trends in this chart" * **Document review**: "Extract text from this image" **Supported formats:** * JPG, PNG, GIF, WebP * Max size: 10MB per image * Multiple images per conversation **Create images from text descriptions** **Available models:** * **GPT-image-1**: Latest model with improved quality and detail * **DALL-E 3**: High-quality generation with natural/vivid styles **How it works:** 1. Ask AVA to generate an image 2. Provide a detailed description 3. Choose style preferences (natural vs vivid) 4. AVA creates the image using your selected model **Examples:** ``` "Generate an image of a modern office workspace with natural lighting" "Create a diagram showing a cloud architecture with Azure services" "Design a logo concept for a tech startup focused on AI" ``` **Use cases:** * **Presentation graphics**: Generate custom visuals for slides * **Concept visualization**: See ideas before building * **Marketing materials**: Create unique images for content * **Documentation**: Illustrate technical concepts **Tips for better results:** * Be specific and detailed in descriptions * Mention style, lighting, colors, composition * Specify "photorealistic" or "illustration" style * Iterate and refine based on results ## Microsoft 365 Actions via Graph API AVA doesn't just read your data—it takes action on your behalf using Microsoft Graph API. ### Schedule Meetings & Manage Calendar **Ask AVA to schedule meetings for you** **Examples:** * "Schedule a 30-minute meeting with Sarah tomorrow at 2pm to discuss Q4 planning" * "Find a time when John, Lisa, and I are all free this week and book an hour" * "Create a recurring weekly standup every Monday at 9am" **What AVA does:** * Checks attendee availability * Finds optimal meeting times * Sends calendar invitations * Adds meeting details and agenda * Sets up Teams meeting link **Advanced:** * "Schedule a meeting with the Sales team next week—find the best time when most people are available" * "Book a room for our quarterly review next Friday afternoon" **Modify existing calendar events** **Examples:** * "Move my 2pm meeting to 3pm" * "Cancel tomorrow's standup" * "Add Mike to the Q4 planning meeting" * "Change the Teams link for Friday's meeting" **What AVA does:** * Finds the specified event * Makes requested changes * Notifies attendees of updates * Reschedules or cancels as needed **Find free time for meetings** **Examples:** * "When am I free tomorrow afternoon?" * "Show me everyone's availability for a meeting this week" * "Find a 2-hour block when the team is all available" **What AVA does:** * Checks calendars (yours and others if shared) * Identifies available time slots * Suggests optimal times * Considers time zones and working hours ### Create & Manage Tasks **Turn action items into tracked tasks** **Examples:** * "Create a task to review the contract by Friday" * "Add 'Prepare Q4 presentation' to my todo list" * "Make a task list from the action items in today's meeting" **What AVA does:** * Creates tasks in Microsoft To Do * Sets due dates and reminders * Adds task details and notes * Organizes into lists * Assigns priorities **Delegate tasks through AVA** **Examples:** * "Assign a task to John to review the proposal by Wednesday" * "Create a task for the team to complete the survey" **What AVA does:** * Creates task with assignee * Sets deadlines * Sends notifications * Tracks assignment **Update task status and details** **Examples:** * "Mark my 'Review contract' task as complete" * "Move the presentation deadline to next Tuesday" * "Show me all my overdue tasks" **What AVA does:** * Updates task status * Modifies due dates * Changes priorities * Provides task overviews ### Share & Collaborate on Files **Share OneDrive/SharePoint files with others** **Examples:** * "Share the Q4 Budget spreadsheet with Sarah" * "Give the Sales team read access to the proposal document" * "Share this presentation with editing permissions to John" **What AVA does:** * Finds the specified file * Sets appropriate permissions * Sends share notification * Provides share link **Update file sharing and access** **Examples:** * "Remove John's access to the contract" * "Change Sarah's permission to view-only on the budget" * "Make this document accessible to everyone in the company" **What AVA does:** * Updates sharing permissions * Revokes access when needed * Modifies permission levels * Manages organizational access **Generate shareable links** **Examples:** * "Create a share link for this presentation" * "Generate a view-only link to the report that expires in 7 days" **What AVA does:** * Creates share link * Sets expiration dates * Configures access levels * Provides link for distribution ### Send Emails & Communications **Compose and send emails** **Examples:** * "Draft a follow-up email to the client about our proposal" * "Send an email to the team with today's meeting notes" * "Compose a response to Sarah's question about the timeline" **What AVA does:** * Drafts email content * Uses conversation context * Formats professionally * Can send on your behalf (with confirmation) **Respond to existing emails** **Examples:** * "Reply to John's email about the budget with approval" * "Forward Sarah's message to the finance team" **What AVA does:** * Locates original email * Crafts appropriate response * Maintains email thread * Handles forwarding with context ## Copy to Workspace Notes Capture insights from any AVA conversation and build comprehensive documents. ### How It Works Have a conversation with AVA in Chat, run a Task, or perform a Knowledge Search Click the "Copy to Notes" button on any AVA response you want to save Content is added to your Workspace Notes where you can organize and format it Export your compiled notes to PowerPoint, Word, or copy to clipboard ### Use Cases **Gather insights from multiple conversations** 1. Ask AVA questions across different topics 2. Copy useful responses to Notes 3. Organize by sections with headings 4. Build comprehensive research document 5. Export to Word for final report **Example:** Research about competitors → Copy findings to Notes → Add insights from sales data → Organize into sections → Export report **Compile meeting materials** 1. Query @Salesforce for account status 2. Copy account summary to Notes 3. Ask for recent email threads 4. Copy key discussion points to Notes 5. Add your talking points 6. Export to PowerPoint for presentation **Example:** Account history → Recent communications → Deal status → All compiled in Notes → Professional presentation **Build documents iteratively** 1. Start with outline in Notes 2. Ask AVA to draft each section 3. Copy good content from chat to Notes 4. Refine and edit in Notes 5. Continue building until complete 6. Export to final format **Example:** Blog post outline → Section by section drafting → Copy best responses → Refine → Complete article **Save insights across sessions** 1. Have conversations over days/weeks 2. Copy valuable insights to Notes 3. Build knowledge base 4. Organize by topic 5. Reference later or share with team **Example:** Daily learnings → Copy key insights → Organize weekly → Build knowledge repository ### Notes Features * **Rich text formatting**: Headers, bullets, bold, italic, links * **Organization**: Structure content with sections * **Persistence**: Notes saved across sessions * **Export options**: PowerPoint, Word, clipboard * **Search**: Find content in your notes * **Share**: Share notes with teammates ## Share Conversations Collaborate with teammates by sharing your AVA conversations—complete with context and history. ### Why Share Conversations? Share ongoing analysis instead of explaining from scratch Get input on complex problems with full context Share research findings with complete background Everyone sees the same information and reasoning ### How to Share Have a conversation in AVA Chat, run a Task, or complete a Knowledge Search Click the "Share" button in the conversation header * Select individual users * Choose Microsoft 365 groups * Set access permissions (view/edit) Recipients receive notification and can access the full conversation ### Sharing Options **Recipients can see but not modify** * View entire conversation history * See all context and data sources * Cannot add messages or change content * Perfect for sharing insights or reports **Recipients can continue the conversation** * View full conversation * Add their own messages * Continue the analysis * Build on existing work * Perfect for team problem-solving ### Use Cases **Share analysis for immediate input** **Scenario**: You've analyzed Q4 sales data and want your manager's opinion **Action**: Share the conversation with your manager → They see all analysis → Provide feedback → You continue based on input **Bring colleagues up to speed** **Scenario**: New teammate joining a project **Action**: Share your research conversations → They get full context → No need to re-explain → They're productive immediately **Team problem-solving** **Scenario**: Troubleshooting a technical issue **Action**: Share your troubleshooting conversation → Team members add their insights → Collective intelligence → Faster solution **Preserve decision-making process** **Scenario**: Made important strategic decision using AVA **Action**: Share the conversation → Stakeholders see full reasoning → Transparent decision process → Create institutional knowledge ## PowerPoint Generation Transform your AVA conversations and notes into professional presentations. ### Current Capability **Export from Notes to PowerPoint** 1. Create content in Workspace Notes 2. Use headings (H1) for slide titles 3. Use bullets for slide content 4. Click "Export to PowerPoint" 5. AVA generates formatted presentation **Features:** * Automatic slide layout selection * Formatted bullets and text * Proper heading hierarchy * Professional design * Editable in PowerPoint after export ### Version 3 (Coming Soon) **Enhanced PowerPoint Generation** is coming in AVA Version 3 with powerful new capabilities! **What's New in Version 3:** Generate presentations directly from Chat conversations—not just Notes AI selects optimal slide layouts based on content type Automatically includes relevant images and charts Apply your organization's PowerPoint templates Advanced text formatting and visual hierarchy "Turn this conversation into a presentation" - done! **Version 3 Example:** ``` You: I just finished analyzing Q4 sales with @Salesforce data. Turn this entire conversation into a presentation. AVA: I've created a 12-slide presentation: - Executive summary slide - Sales performance overview with charts - Top deals breakdown - Regional analysis - Recommendations All formatted with your company template! ``` ## Export Capabilities Generate Office documents from any AVA conversation, task, or knowledge search. **Presentations from conversations** * Export Notes to slides * V3: Direct from chat * Professional formatting * Editable in PowerPoint **Formatted documents** * Conversation transcripts * Research reports * Meeting summaries * Proper heading structure **Data and tables** * Export data tables * Analysis results * Formatted spreadsheets * Formulas preserved **Shareable documents** * Export to PDF * Perfect for distribution * Print-ready format **Quick sharing** * Copy any content * Paste anywhere * Formatted or plain text **Save to cloud** * Export directly to OneDrive * Automatic file naming * Organized storage ### Export Workflows 1. Conduct research across multiple sources (@Salesforce, @Files, @Web) 2. Copy key findings to Notes 3. Organize with headings and bullets 4. Export to Word document 5. Final edits in Word 6. Save to SharePoint 1. Analyze data in Chat 2. Generate insights and recommendations 3. Copy to Notes 4. Structure with slide titles 5. Export to PowerPoint 6. Present to leadership 1. Query data sources 2. Ask AVA for analysis 3. Generate data tables 4. Export to Excel 5. Further analysis or charting 6. Share with team 1. Get answer or insight from AVA 2. Copy to clipboard 3. Paste into email, Slack, Teams 4. Instant sharing ## The Power of Microsoft Graph API All these capabilities are possible because AVA integrates deeply with Microsoft Graph API—the unified API for Microsoft 365. ### What is Microsoft Graph API? Microsoft Graph API provides a single endpoint to access data and intelligence across Microsoft 365, including: * **Calendar & Events**: Create, update, manage meetings * **Mail & Messages**: Read, send, organize email * **Files & Documents**: Access, share, collaborate * **Tasks & Planner**: Create, assign, track work * **Teams & Channels**: Messages, meetings, collaboration * **Users & Groups**: Directory, profiles, org structure * **Security & Compliance**: Policies, alerts, audit ### Why This Matters AVA acts on your behalf using **your** permissions. It can only do what you can do—no elevated access, no backdoors. One conversation can span multiple Microsoft services seamlessly. Check calendar, send email, share file—all without switching contexts. All actions logged and audited. Respects your organization's security policies. Complies with data governance rules. As Microsoft adds capabilities to Graph API, AVA automatically gains new powers. The platform grows without AVA updates. ## Next Steps Experience these capabilities in conversations Use core capabilities in multi-step workflows Master Notes, Search, and File Viewer Start using AVA today *** **AI That Works™** - AVA is more than a chatbot. It's your AI super assistant, taking action across your entire Microsoft 365 environment to help you get real work done. # How AVA Works Source: https://docs.datarm.ai/essentials/how-it-works Understanding AVA architecture, identity flow, and technical implementation ## Architecture Overview AVA is deployed as a private, single-tenant application within your Azure environment. This architecture ensures your data never leaves your tenant while providing enterprise-grade AI capabilities through a secure, identity-based access model. Complete AVA architecture diagram showing identity flow, backend services, MCP server hub, enterprise apps, and AI providers ## Identity and Access Flow The foundation of AVA's security model is **delegated permissions** through Microsoft Entra ID: Users sign in to AVA using their Microsoft work account (same credentials used for Office 365) Microsoft Entra ID authenticates the user and establishes their identity and permissions AVA operates with the user's permissions—it can only access what the authenticated user can access **Critical**: AVA doesn't maintain separate permission systems. Delegated permissions MATCH existing access controls **No Elevated Access**: AVA never has broader permissions than the user. If you can't access a file in SharePoint, neither can AVA on your behalf. ## System Architecture ### Frontend Layer **AVA Front End App Service** (Azure App Service) * React-based web application * Handles user interface and interactions * Communicates with backend via secure APIs * Session management and state ### Backend Layer Two primary backend services running as Azure Container Apps: **Core orchestration and AI integration** * Chat conversation management * Task workflow execution * Model Context Protocol (MCP) orchestration * AI provider integration (AI Foundry, AWS Bedrock, GCP Vertex) * User context and session handling **Knowledge Search and RAG capabilities** * Document embedding generation * Vector similarity search * Corpus management * SharePoint synchronization * Search result ranking and retrieval ### MCP Server Hub **Container Apps running MCP servers** Model Context Protocol (MCP) servers act as standardized connectors to enterprise applications: * Outlook (Mail & Calendar) * OneDrive (Files) * SharePoint (Sites) * Teams (Meetings & Messages) * Read and write CRM data * Query opportunities, leads, cases * Update records * Issue management * Project queries * Status updates * Repository access * Code search * Pull request queries **Additional MCP Servers:** * NetSuite integration * Microsoft Security (Defender) * Web Scraping & Crawling * Custom MCP servers (extensible) ### Extend AVA to Any Enterprise Application **Leverage Your Existing SSO Infrastructure**: AVA can integrate with virtually any application already configured as an Enterprise Application in your Microsoft Entra ID—if it uses OAuth or SAML for authentication. **How It Works:** Your organization has likely already configured single sign-on (SSO) for dozens of enterprise applications in Microsoft Entra ID. AVA can leverage these existing OAuth/SAML configurations to create new MCP integrations without additional authentication setup. **The Power of MCP Extensibility:** 1. **Use What You Already Have**: Enterprise apps already in your Entra ID tenant can become AVA data sources 2. **OAuth/SAML Ready**: If your app supports OAuth or SAML (most modern SaaS apps do), it can connect to AVA 3. **No Duplicate Auth**: Leverages the SSO configuration you've already completed 4. **Delegated Permissions**: Same identity-based security model as all AVA integrations **Examples of What's Possible:** * **HR Systems**: Workday, BambooHR, ADP - access employee data, org charts, PTO balances * **Project Management**: Monday.com, Asana, ClickUp - query tasks, projects, timelines * **Customer Support**: Zendesk, Freshdesk, Intercom - access tickets, customer conversations * **Marketing Platforms**: HubSpot, Marketo, Mailchimp - campaign data, analytics, contacts * **Finance Systems**: QuickBooks, Xero, Bill.com - financial data, invoices, reports * **Custom Internal Tools**: Any internal app with OAuth/SAML SSO configured in Entra ID **Requirements:** ✅ Application registered as Enterprise Application in your Entra ID ✅ OAuth 2.0 or SAML 2.0 authentication configured ✅ SSO working for your users ✅ API access available (most modern apps provide REST APIs) **The possibilities expand with your enterprise ecosystem**: As you add new Enterprise Applications to Entra ID, AVA's potential integrations grow—creating a truly unified AI assistant across your entire technology stack. ### Data Layer **Relational data storage** * User context and sessions * Chat conversation history * Task definitions and state * Prompt library * Activity logs * Corpus metadata **Vector storage for RAG** * Document embeddings for Knowledge Search * Semantic search capabilities * Fast similarity matching * Embedded alongside relational data **File and binary storage** * Uploaded documents * Temporary file storage * Export artifacts (PowerPoint, Word, Excel) ### AI Providers AVA supports multiple AI providers for model flexibility: **Primary provider** * GPT-4, GPT-4 Turbo * GPT-3.5 Turbo * o1, o3-mini models * Embeddings (text-embedding-ada-002) **Alternative provider** * Claude Sonnet * Claude Opus * Other Bedrock models **Alternative provider** * Gemini Pro * PaLM models Users can select different models based on: * Task complexity * Response speed requirements * Cost considerations * Context window size needs ### Supporting Azure Resources Application monitoring, telemetry, and performance tracking Secure storage for secrets, API keys, and connection strings Passwordless authentication between Azure services ## How a Conversation Works Let's trace a typical user interaction: User types a message in AVA Chat: "@Salesforce show me my open opportunities" Front End App Service sends request to Back End App Service with user context Backend validates user identity via Entra ID and retrieves delegated access token Backend identifies @Salesforce mention and routes request to Salesforce MCP server Salesforce MCP server: * Uses delegated user token * Queries Salesforce with user's permissions * Returns only data user can access Backend sends user question + Salesforce data to selected AI provider (e.g., GPT-4 via Azure AI Foundry) AI model generates response based on retrieved data and user context Response flows back through backend → frontend → user's browser ## How Knowledge Search Works Knowledge Search uses Retrieval Augmented Generation (RAG): User creates a corpus pointing to a SharePoint site/document library Services App Service: * Connects to SharePoint using user's delegated permissions * Downloads documents user can access * Chunks documents into manageable segments * Each chunk is sent to embedding model (e.g., text-embedding-ada-002) * Vector embeddings are generated * Embeddings stored in PostgreSQL with pgVector If enabled, AVA periodically checks SharePoint for: * New documents * Modified documents * Deleted documents Updates embeddings accordingly When user searches: * Query is converted to vector embedding * pgVector performs similarity search * Top matching chunks are retrieved * Retrieved chunks are sent to AI model as context * Model generates response based on actual document content * Cites source documents in response ## How Tasks Work Multi-step task execution: User creates task with multiple steps, each with its own instructions User can add documents from workspace (OneDrive, SharePoint, or local files) Backend processes each step sequentially: * Evaluates step instructions * Includes previous step results as context * Can use @ mentions to pull live data * SuperPrompt (if enabled) enhances AI guidance Task state stored in PostgreSQL: * Current step * Step results * Overall progress Successful tasks can be saved as templates for reuse ## Security Architecture ### Authentication & Authorization ``` User → Entra ID (SAML/OAuth) → AVA Front End ↓ Delegated Access Token ↓ Back End App Service ↓ Uses User's Token for All MCP Calls ↓ Enterprise Apps (M365, Salesforce, etc.) ``` **Key Principles:** * ✅ **Delegated Permissions**: AVA uses SAML/OAuth to obtain delegated access * ✅ **User-Scoped Access**: Every API call uses the authenticated user's token * ✅ **No Privilege Escalation**: AVA cannot access data the user cannot access * ✅ **Audit Trail**: All actions logged with user identity ### Data Security * All data stays within your Azure tenant * Processing happens in your tenant's region * No external SaaS data transfer * You control data location and retention * **In Transit**: TLS 1.3 for all communications * **At Rest**: Azure-managed encryption for database and blob storage * **Secrets**: Stored in Azure Key Vault * Container Apps in isolated virtual network * Managed identities for service-to-service auth * No hard-coded credentials * Azure Private Link support (optional) ## Deployment Model AVA is deployed as a **single-tenant, private deployment** for each customer: * Deployed entirely within your subscription * You control all Azure resources * No shared infrastructure * Complete data isolation * Registered in your Entra ID * Uses your identity provider * Integrates with your policies * Respects your conditional access **This means:** * ✅ Data sovereignty: Your data, your tenant, your control * ✅ Compliance-friendly: Easier to meet regulatory requirements * ✅ Customizable: Can be tailored to your environment * ✅ Secure: No multi-tenant security risks ## Scalability & Performance AVA is built to scale: * **Azure Container Apps**: Auto-scales based on demand * **PostgreSQL Flexible Server**: Can scale vertically and add read replicas * **Blob Storage**: Virtually unlimited storage capacity * **AI Providers**: Load balanced across regions **Performance Characteristics:** * Chat responses: 2-10 seconds (depending on model and complexity) * Knowledge Search: Sub-second vector search, 3-8 seconds total with AI generation * Task execution: Varies by steps and data sources * Concurrent users: Scales automatically with container app scaling ## Monitoring & Observability * Request/response telemetry * Performance metrics * Dependency tracking * Custom events and metrics * Failure analysis * Structured application logs * User activity tracking * MCP server call logs * AI provider interaction logs * Export to Log Analytics * Service health monitoring * Performance degradation alerts * Error rate thresholds * Custom metric alerts ## Technical Stack Summary | Layer | Technology | | ---------------- | ----------------------------------------- | | **Frontend** | React, TypeScript, Azure App Service | | **Backend** | Python, FastAPI, Azure Container Apps | | **Database** | PostgreSQL Flexible Server with pgVector | | **Storage** | Azure Blob Storage | | **Identity** | Microsoft Entra ID (SAML/OAuth) | | **Secrets** | Azure Key Vault | | **Monitoring** | Application Insights, Log Analytics | | **AI Models** | Azure AI Foundry, AWS Bedrock, GCP Vertex | | **Integrations** | Model Context Protocol (MCP) | | **Embedding** | text-embedding-ada-002, pgvector | ## Next Steps Explore what you can do with AVA Learn how to deploy AVA Start using AVA today # Key Features Source: https://docs.datarm.ai/essentials/key-features Comprehensive overview of AVA capabilities across Chat, Tasks, Knowledge Search, and Workspace ## AVA Chat Features Your conversational AI interface with connected data access. Pull data from any connected source by typing @ - simple, intuitive access to all your business systems Combine information from multiple sources in a single conversation without switching contexts Access all your past conversations, organized by recency with search and filtering Share chat sessions with team members, including full context and history ### Connected Data Sources Access live data from enterprise applications: * **@Mail**: Query your Outlook inbox and sent items * **@Calendar**: Check schedule, find meetings, see availability * **@Files**: Search and access OneDrive and SharePoint documents * **@Teams**: Access Teams messages and channel information * **@ToDo**: View and manage your task lists * **@Salesforce**: Query opportunities, leads, cases, accounts, contacts * Update CRM records directly from chat * Create new records with AI assistance * Generate reports and forecasts * **@GitHub**: Search repositories, check PR status, review code * **@Jira**: Query issues, update tickets, check sprint status * Link commits to conversations * Generate development summaries * **@NetSuite**: Business records and transactions * **@Security**: Microsoft Defender alerts and incidents * **@Web**: Web scraping and content crawling * **Custom MCP servers**: Extend with your own integrations ### Conversation Capabilities * **Context Awareness**: AVA remembers earlier in the conversation and maintains context across @ mentions * **Multi-Turn Dialogues**: Natural back-and-forth conversations with follow-up questions * **Code Generation**: Generate and explain code in multiple languages * **Data Analysis**: Analyze data from connected sources and generate insights * **Content Creation**: Draft emails, documents, presentations with business context ## AVA Tasks Features Multi-step workflow automation with intelligent orchestration. Create tasks with multiple sequential steps, each with its own instructions Save successful workflows as templates and share with your team Toggle enhanced AI guidance for better results on complex tasks Add documents from workspace for context in any step ### Task Capabilities Each step runs in order, with results from previous steps available as context for subsequent steps Use @ mentions within task steps to pull fresh data from connected sources Visual progress indicators show which steps are complete, in progress, or pending Share task templates with: * Individual team members * Microsoft 365 groups * Entire organization Export task results to Office formats (Word, PowerPoint, Excel) or plain text ### Common Task Use Cases * **Research & Analysis**: Multi-step research workflows with synthesis * **Report Generation**: Automated report creation from multiple data sources * **Document Processing**: Batch processing of documents with consistent steps * **Content Creation**: Structured content creation pipelines * **Data Transformation**: ETL-style workflows for data manipulation ## Knowledge Search Features RAG-powered semantic search across your SharePoint documents. Query multiple document collections simultaneously in a single search Create custom knowledge bases from any SharePoint site or document library Automatically update corpus when documents are added, modified, or deleted Extract custom fields from documents for enhanced search capabilities ### Knowledge Search Capabilities * Understands intent, not just keywords * Finds relevant content even without exact matches * Handles questions in natural language * Returns ranked results by relevance * Every answer includes source citations * Direct links to source documents * Displays file names and locations * Shows relevant excerpts * View all embedded files in a corpus * Add or remove specific documents * Edit corpus descriptions * Configure auto-sync settings * Set field extraction rules * Respects SharePoint permissions * Users only search documents they can access * Corpus owners control sharing * Delegated permissions model ### Perfect For * **Contract Libraries**: Search all contracts for specific clauses, terms, or conditions * **HR Documentation**: Find policies, procedures, and employee information * **Technical Documentation**: Query manuals, specifications, and guides * **Customer Service**: Search support articles, FAQs, and troubleshooting guides * **Legal Documents**: Research case law, precedents, and legal opinions * **Compliance Materials**: Find regulatory documents and compliance procedures ## Workspace Features Integrated panel for search, notes, and file viewing. ### Search Tab Search OneDrive and SharePoint for documents to add to conversations Access recorded Teams meetings with transcripts and summaries Upload files from your local device to reference in chat **Features:** * Full-text search across your Microsoft 365 files * Filter by file type (Word, Excel, PowerPoint, PDF, etc.) * Preview files before adding to conversation * Drag-and-drop file upload * Meeting filter by date and attendees ### Notes Tab Rich text editor for capturing ideas and creating content. **Capabilities:** * Full formatting toolbar (bold, italic, headers, lists, etc.) * Create structured documents collaboratively with AVA * Copy notes to clipboard * **Export to PowerPoint**: Generate presentations directly from notes * Organize ideas from multiple chat sessions * Save notes for later reference **Use Cases:** * Meeting notes while chatting with AVA * Brainstorming and idea capture * Document drafting with AI assistance * Presentation content development * Structured note-taking ### File Viewer Tab View and edit documents from SharePoint and OneDrive. **Features:** * **View Mode**: Preview documents without editing * **Edit Mode**: Make changes using Microsoft Office Online integration * Multi-tab support for multiple documents * Direct access to user's files * Synchronized with SharePoint/OneDrive **Benefits:** * Reference documents while chatting about them * Make edits without leaving AVA * Keep context while working on files * Side-by-side comparison of documents ## Prompt Library Features Create, organize, and share custom prompts across your organization. Build custom prompts with instructions, connected data references, and file type filters Share prompts with individuals, teams, or Microsoft 365 groups Filter prompts by connected data sources, file types, or sharing scope Use pre-built prompts as starting points for your own ### Prompt Capabilities Include specific data sources in your prompts: * @Salesforce for CRM queries * @Microsoft365 for file and calendar access * @Jira for development workflow prompts * Any combination of connected sources Specify which file types the prompt applies to: * Excel spreadsheets only * Word documents only * PowerPoint presentations * PDF documents * Email messages * Meeting transcripts Control who can use your prompts: * **Personal**: Only you * **Team**: Specific Microsoft 365 group * **Organization**: Everyone in your company * **Individual**: Specific people Prompts maintain history: * See when prompts were created/modified * Track who created or last updated * Revert to previous versions if needed ### Example Prompts ``` Name: "Weekly Sales Pipeline Review" Connected Data: @Salesforce File Types: Excel Sharing: Sales Team Prompt: "Review all opportunities closing this week. Create a summary table with Company, Deal Value, Close Date, Stage, and Risk Assessment. Sort by deal value descending. Export to Excel." ``` ``` Name: "Contract Clause Extraction" Connected Data: @Contracts Demo (Knowledge Search) File Types: PDF, Word Sharing: Legal Team Prompt: "Extract all payment terms, termination clauses, and renewal provisions from this contract. Create a structured summary with section references." ``` ## Model Selection & AI Features Choose the right AI model for each task. ### Available Models **Text & Reasoning:** * **GPT-5**: Latest flagship model for logic-heavy and multi-step tasks * **GPT-5 Pro**: Extended reasoning capabilities for complex analysis * **GPT-5 Codex**: Advanced coding model for development tasks * **GPT-5 Chat**: Natural, multimodal, context-aware conversations * **GPT-5 Mini/Nano**: Lightweight, cost-effective variants for speed **Image Generation:** * **GPT-image-1**: Latest image generation with improved quality and detail * **DALL-E 3**: High-quality image generation with natural/vivid styles * **Claude Opus 4.1**: Most intelligent model, industry leader for coding and agents * **Claude Sonnet 4.5**: Most intelligent for complex agents, 1M token context * **Claude Haiku 4.5**: Fast, cost-effective near-frontier intelligence * **Claude Sonnet 4**: Hybrid reasoning with instant or deep thinking modes * **Claude Opus 4**: Next-gen model for autonomous AI agents * **Gemini 2.5 Pro**: Most advanced model with Deep Think mode for complex reasoning * **Gemini 2.5 Flash**: Best price-performance, rapid creative workflows * **Gemini 2.5 Flash-Lite**: Speed and cost-optimized variant * **Gemini Pro**: Standard model for general-purpose tasks ### Model Selection Criteria Choose based on: * **Task Complexity**: Use GPT-5, GPT-5 Pro, Claude Opus 4.1, or Gemini 2.5 Pro for complex reasoning and multi-step logic * **Speed Requirements**: Use GPT-5 Nano, Claude Haiku 4.5, or Gemini 2.5 Flash-Lite for quick responses * **Coding Tasks**: GPT-5 Codex or Claude Opus 4.1 excel at code generation and debugging * **Context Window**: Claude Sonnet 4.5 (1M tokens) for large document analysis * **Cost**: Balance performance needs with budget - GPT-5 Mini/Nano, Flash, and Haiku models offer best price-performance * **Conversations**: GPT-5 Chat and Gemini 2.5 models excel at natural, context-aware dialogue * **Image Generation**: GPT-image-1 for latest quality, DALL-E 3 for creative visual content ### Custom Instructions Personalize AVA's behavior globally: ``` Example Custom Instructions: "I'm a VP of Sales at a B2B SaaS company. When analyzing opportunities: - Always show ARR (Annual Recurring Revenue) - Include sales cycle length - Highlight competitive threats - Flag deals over $100K Format financial data in tables with currency symbols. My timezone is Pacific Time (PT). My working hours are 8am-6pm weekdays." ``` AVA applies these preferences to all conversations and tasks. ## Export & Integration Features Generate Office documents from your AI work. From conversations, notes, or task results Formatted documents with proper styling Data tables and analysis results Simple text file exports ### PowerPoint Generation Special capability to create presentations: * **From Chat**: Select conversation turns to include as slides * **From Notes**: Convert notes directly to presentation format * **Smart Formatting**: Auto-applies appropriate slide layouts * **Image Support**: Includes images and charts where relevant * **Customizable**: Edit generated slides before export ### Export Options All exports support: * Copy to clipboard for pasting elsewhere * Download directly to local device * Save to OneDrive/SharePoint * Share via link with team members ## Collaboration Features Work together with your team using AVA. Share entire conversations with team members including full context Share successful workflow templates across teams Organizational prompt repository for best practices Shared corpus access for team knowledge search ### Sharing Capabilities * **Granular Permissions**: Control who can view vs edit * **Microsoft 365 Integration**: Share with existing groups and teams * **Link-Based Sharing**: Generate secure share links * **Access Tracking**: See who has accessed shared resources ## Core Capabilities Available Everywhere **AI That Works™** - These powerful features work across all of AVA (Chat, Tasks, Knowledge Search), making AVA your AI super assistant. **Add and generate images** **Upload Images:** * Add screenshots, diagrams, photos to any conversation * Supported formats: JPG, PNG, GIF, WebP * Available in Chat, Tasks, and Knowledge Search * AVA analyzes and references images in context **Generate Images:** * Create images using GPT-image-1 or DALL-E 3 * Simply describe what you want to see * Perfect for presentations, concepts, documentation * Generate professional visuals on demand **Use Cases:** * Design feedback on mockups * Troubleshoot errors with screenshots * Visualize data and concepts * Create presentation graphics * Illustrate documentation [Learn more about image capabilities →](/essentials/core-capabilities#work-with-images) **Take action via Microsoft Graph API** **Schedule Meetings:** * "Schedule a 30-minute meeting with Sarah tomorrow at 2pm" * AVA checks availability and sends invitations * Creates Teams meeting links automatically **Create Tasks:** * "Add 'Review Q4 budget' to my todo list by Friday" * Tasks created in Microsoft To Do * Can assign to others with deadlines **Share Documents:** * "Share this presentation with the Sales team" * AVA sets permissions and sends notifications * Manages access levels (view/edit) **Send Emails:** * "Draft a follow-up email to the client" * AVA composes and can send on your behalf * Uses conversation context for relevant content **The possibilities are nearly endless** with Microsoft Graph API integration. [Explore all Microsoft actions →](/essentials/core-capabilities#microsoft-365-actions-via-graph-api) **Build documents from conversations** **One-Click Copy:** * Click "Copy to Notes" on any AVA response * Content added to your Workspace Notes * Organize and format as needed * Build comprehensive documents **Use Cases:** * Compile research from multiple conversations * Gather insights for reports * Build presentations from chat sessions * Create knowledge documents * Organize findings across sessions **Export Options:** * PowerPoint presentations * Word documents * Copy to clipboard * Save to OneDrive [Learn more about Notes →](/features/workspace#notes-tab) **Collaborate with teammates** **Why Share:** * Speed up work by sharing ongoing analysis * Get input with full context * Transfer knowledge efficiently * Team alignment on complex problems **How to Share:** 1. Click "Share" in conversation header 2. Select individual users or Microsoft 365 groups 3. Set permissions (view or collaborate) 4. Recipients see full conversation history **Access Levels:** * **View**: See conversation but can't modify * **Collaborate**: Continue the conversation **Use Cases:** * Get manager feedback on analysis * Onboard team members quickly * Collaborative problem-solving * Document decision-making process [Explore conversation sharing →](/essentials/core-capabilities#share-conversations) **From conversations to presentations** **Current Capability:** * Export Notes to PowerPoint * H1 headings become slide titles * Bullets formatted automatically * Professional layouts applied **Version 3 (Coming Soon):** * Generate presentations directly from Chat * Intelligent layout selection * Automatic image integration * Brand template support * "Turn this conversation into a presentation" - done! **Use Cases:** * Executive summaries * Sales presentations * Research findings * Meeting preparation * Status reports [Learn about PowerPoint generation →](/essentials/core-capabilities#powerpoint-generation) **Generate Office documents** **Export Options:** * **PowerPoint**: Presentations from notes or conversations * **Word**: Formatted documents and reports * **Excel**: Data tables and analysis * **PDF**: Shareable documents * **Clipboard**: Quick copy and paste **Available From:** * Any chat conversation * Task results * Knowledge Search results * Workspace Notes * Any AVA interaction **Workflow:** 1. Complete work in AVA 2. Choose export format 3. AVA generates document 4. Download or save to OneDrive 5. Edit in Office if needed [Explore export capabilities →](/essentials/core-capabilities#export-capabilities) **AI That Works**: AVA leverages Microsoft Graph API to be your AI super assistant—taking real action across your Microsoft 365 environment, not just answering questions. **Want to learn more?** Visit our comprehensive [Core Capabilities](/essentials/core-capabilities) page to explore all these features in detail. ## Security & Privacy Features Enterprise-grade security built in. AVA only accesses what the authenticated user can access - no elevated privileges All data stays within your Azure tenant - never sent to external SaaS providers Complete audit trail of all user actions via Application Insights TLS 1.3 in transit, Azure-managed encryption at rest Microsoft Entra ID authentication with SAML/OAuth ## Performance & Scale Built for enterprise workloads: * **Auto-Scaling**: Azure Container Apps scale based on demand * **Fast Search**: Sub-second vector search with pgVector * **Large Context**: Support for 128K+ token context windows (model-dependent) * **Concurrent Users**: No hard limits - scales with your Azure resources * **Document Processing**: Parallel embedding generation for fast corpus creation * **Model Load Balancing**: Distribute requests across AI provider regions ## Next Steps See how organizations use these features Start using AVA today Deep dive into AVA Chat Learn about Knowledge Search # Use Cases Source: https://docs.datarm.ai/essentials/use-cases Real-world applications of AVA across teams and industries ## By Team/Role ### Sales Teams **Scenario**: Sales manager needs to review upcoming deals **With AVA**: ``` @Salesforce show me all opportunities closing this quarter with deal value over $50K. Group by stage and highlight any at risk. ``` **Result**: AVA pulls live Salesforce data, analyzes each opportunity, provides risk assessment, and can generate a PowerPoint presentation for the weekly sales meeting **Time Saved**: 2 hours/week → 5 minutes **Scenario**: Preparing for customer meeting tomorrow **With AVA**: ``` @Calendar who am I meeting with tomorrow at 2pm? @Salesforce show me their account history @Mail find all email threads with this contact Draft a meeting agenda based on previous conversations ``` **Result**: AVA connects the dots across calendar, CRM, and email to provide complete account context and drafts a personalized agenda **Time Saved**: 45 minutes → 5 minutes **Scenario**: Need to create custom proposal for prospect **With AVA**: * Upload previous winning proposals to workspace * @Salesforce pull prospect details and requirements * Use AVA Tasks to create multi-step proposal workflow * Export final proposal to Word or PowerPoint **Result**: Consistent, high-quality proposals leveraging past successes **Time Saved**: 4 hours → 45 minutes ### Customer Support Teams **Scenario**: Customer asks complex technical question **With AVA**: Use Knowledge Search with support documentation corpus to find relevant articles instantly, even if exact keywords don't match **Benefit**: Faster resolution, consistent answers **Scenario**: Need full customer context quickly **With AVA**: @Salesforce get case history, @Mail find all correspondence, @Files access shared documents—all in one conversation **Benefit**: Complete context without system-hopping **Scenario**: Complex technical response needed **With AVA**: Query knowledge base, reference similar past cases, draft response using your custom instructions for tone and formatting **Benefit**: Professional, accurate responses quickly **Scenario**: Update multiple systems after resolving issue **With AVA**: @Salesforce update case status, @Mail send customer update, add resolution to knowledge base—all from one conversation **Benefit**: Consistent updates across systems ### Development Teams **Workflow**: 1. @Jira show all issues in next sprint 2. @GitHub check which PRs are still open 3. Ask AVA to analyze capacity and dependencies 4. Generate sprint planning doc with recommendations 5. Export to PowerPoint for sprint kickoff **Value**: Data-driven sprint planning with minimal manual work **Workflow**: 1. @GitHub show me PR #123 2. @Jira find related issue 3. Ask AVA to explain the changes and potential impact 4. Review code with AI-powered analysis 5. Draft review comments **Value**: Faster, more thorough code reviews **Workflow**: 1. @Jira get bug report details 2. @GitHub search for similar past issues 3. @Files access relevant documentation 4. Ask AVA to suggest root cause and fix 5. Update Jira with findings **Value**: Faster bug resolution with context from multiple sources **Workflow**: 1. @GitHub get repository README and code 2. Upload architecture diagrams from OneDrive 3. Use AVA Tasks to generate comprehensive docs 4. Export to Word or Markdown 5. Commit back to repository **Value**: Up-to-date documentation without manual writing ### Executive Leadership **Scenario**: Board meeting tomorrow, need comprehensive update **With AVA**: * @Calendar review this week's meetings * Access recorded Teams meetings for key discussions * @Salesforce get sales metrics * Generate executive summary in notes * Export to PowerPoint for presentation **Scenario**: Need analysis for strategic decision **With AVA**: * Query multiple data sources simultaneously * Analyze trends and patterns * Ask "what-if" questions with different scenarios * Generate decision memo with supporting data **Scenario**: Company-wide announcement needed **With AVA**: * Review previous communications for tone/style * Draft announcement with custom instructions * Get feedback from AVA on clarity and impact * Export final version ### HR Teams **Scenario**: Employee asks about PTO policy **With AVA Knowledge Search**: * Create corpus from HR policy SharePoint site * Employee handbook auto-synced * HR can quickly search: "What's the PTO rollover policy?" * AVA returns exact policy with source citation **Benefit**: Instant, accurate policy answers; Less HR admin time **Scenario**: New hire starting Monday **With AVA**: * @Files access onboarding checklist template * @Calendar schedule orientation meetings * @Mail draft welcome email * Use Tasks to create personalized onboarding plan * Share with manager via Microsoft 365 group **Benefit**: Consistent, comprehensive onboarding **Scenario**: Need to find all documentation for employee review **With AVA**: * @Files search for employee's documents * @Mail find performance review emails * Compile everything in workspace * Generate review summary document **Benefit**: Complete employee context quickly ### Legal & Compliance Teams **Use Case**: Review hundreds of contracts for specific clause **Solution**: Knowledge Search corpus of all contracts; Query: "Find all contracts with indemnification clauses" **Time Saved**: Days → Minutes **Use Case**: Find relevant precedents and case law **Solution**: Corpus of legal documents; Semantic search finds relevant cases even with different terminology **Benefit**: More comprehensive research **Use Case**: Ensure vendor contracts meet requirements **Solution**: Create task template for contract review checklist; Apply to each new vendor contract **Benefit**: Consistent compliance verification **Use Case**: Draft new contract based on templates **Solution**: Upload template contracts; Use AVA to customize for specific situation; Export to Word **Benefit**: Faster drafting with consistency ### Marketing Teams **Workflow**: 1. @Files access brand guidelines and past campaigns 2. @Salesforce get target customer personas 3. Use AVA Tasks for content creation pipeline: * Step 1: Research topic * Step 2: Create outline * Step 3: Draft content * Step 4: Optimize for SEO 4. Export to Word for review **Value**: Data-informed content at scale **Workflow**: 1. Upload campaign performance data 2. @Salesforce get lead conversion data 3. Ask AVA to analyze what's working 4. Generate recommendations 5. Create presentation for team review **Value**: Data-driven marketing decisions **Workflow**: 1. @Calendar review upcoming events 2. @Files access event materials from past years 3. Create updated materials using AVA 4. @Mail draft attendee communications 5. Coordinate via Teams integration **Value**: Efficient event execution ## By Industry ### Financial Services **Scenario**: Wealth advisor preparing for client meeting **With AVA**: * Query client portfolio data from internal systems * Access market research documents via Knowledge Search * @Calendar review past meeting notes * Generate personalized portfolio recommendations * Create presentation deck automatically **Compliance**: All data access logged, respects existing permissions **Scenario**: Quarterly compliance report due **With AVA Tasks**: * Multi-step workflow to gather data from required systems * Automated formatting per regulatory requirements * Source citation for all data points * Export to required format (Excel/PDF) **Benefit**: Faster, more accurate reporting **Scenario**: Analyze risk exposure across portfolio **With AVA**: * Pull data from multiple risk systems * Query historical incident reports via Knowledge Search * AI-powered pattern analysis * Generate risk assessment report **Value**: Comprehensive risk view ### Healthcare **Scenario**: Review patient care guidelines **Solution**: Knowledge Search corpus of clinical protocols; Quickly find relevant treatment guidelines **Benefit**: Evidence-based care **Scenario**: Literature review for research project **Solution**: Corpus of research papers and journals; Semantic search across all publications **Benefit**: Comprehensive literature review **Scenario**: Prepare for accreditation audit **Solution**: AVA Tasks for documentation gathering; @Files access all required documentation **Benefit**: Organized audit preparation **Scenario**: Create next month's staff schedule **Solution**: @Calendar check availability, @Mail coordinate with staff, use AVA to optimize schedule **Benefit**: Efficient scheduling ### Manufacturing **Scenario**: Find all QA procedures for specific product line **Solution**: * Knowledge Search across QA documentation * Query: "Show all quality control procedures for Product X" * Get instant results with source citations **Benefit**: Fast access to critical procedures **Scenario**: Production issue needs root cause analysis **Solution**: * Access incident reports via Knowledge Search * Search for similar past incidents * Use AVA to analyze patterns * Generate investigation report **Benefit**: Learn from past issues **Scenario**: Evaluate supplier performance **Solution**: * @Files access supplier contracts * Query performance data * Use AVA to analyze trends * Create supplier scorecard **Benefit**: Data-driven supplier decisions ### Education * @Files access previous course materials * Knowledge Search educational resources * Use AVA Tasks to create new curriculum * Generate course syllabus and materials * Knowledge Search FAQ and support docs * Quickly answer student questions * Consistent information across support team * Reduce time to resolve student issues * @Calendar coordinate meetings and events * @Mail automate routine communications * Generate reports and documentation * Streamline administrative workflows ## Common Workflow Patterns ### The "Research & Synthesis" Pattern Perfect for: Market research, competitive analysis, literature reviews **Steps**: 1. Create AVA Task with research steps 2. Add relevant documents to workspace 3. Query connected data sources with @ mentions 4. Use Knowledge Search for deep document analysis 5. AVA synthesizes findings across all sources 6. Export final research report **Example**: "Analyze our competitors' pricing strategies" ### The "Data Gathering" Pattern Perfect for: Status reports, dashboards, executive summaries **Steps**: 1. @Salesforce get sales data 2. @Jira get project status 3. @GitHub get deployment stats 4. Ask AVA to combine and analyze 5. Generate summary report 6. Export to PowerPoint or Excel **Example**: "Create weekly executive status report" ### The "Document Processing" Pattern Perfect for: Contract review, resume screening, invoice processing **Steps**: 1. Upload documents to workspace (or use @Files) 2. Create Task template with processing steps 3. Apply template to each document 4. Extract key information consistently 5. Compile results **Example**: "Extract key terms from all vendor contracts" ### The "Communication Drafting" Pattern Perfect for: Emails, announcements, reports, presentations **Steps**: 1. Gather context from connected sources 2. Review similar past communications via @Files or Knowledge Search 3. Apply custom instructions for tone/style 4. Draft with AVA 5. Refine with back-and-forth conversation 6. Export to appropriate format **Example**: "Draft customer email explaining new feature" ### The "Meeting Preparation" Pattern Perfect for: Client meetings, reviews, presentations **Steps**: 1. @Calendar identify upcoming meeting 2. Access previous meeting recordings/notes 3. Gather relevant data from connected sources 4. Use AVA to create agenda and talking points 5. Generate presentation in PowerPoint 6. Add notes in workspace during meeting **Example**: "Prepare for quarterly business review with Acme Corp" ## Success Metrics Organizations using AVA report: On research and data gathering tasks Document search and retrieval Average time saved per knowledge worker Within 30 days of deployment Intuitive interface, familiar patterns Delegated permissions, data in tenant ## Getting Started with Your Use Case What repetitive tasks consume the most time for your team? Begin with AVA Chat and @ mentions for daily tasks Set up Knowledge Search corpora for your document libraries Convert common workflows into reusable task templates Share successful prompts and tasks with your team ## Next Steps Get started with AVA today Learn about conversational AI Set up document search Discuss your use case with us # What is AVA? Source: https://docs.datarm.ai/essentials/what-is-ava Understanding AVA: Your enterprise AI assistant integrated with Microsoft 365 and business systems ## Overview ### Your AI Super Assistant AVA (AI Virtual Assistant) is an enterprise AI productivity platform that runs securely within your Azure tenant. Unlike external SaaS AI tools that just answer questions, **AVA takes action**—it's AI that works, not just talks. Through Microsoft Graph API integration, AVA schedules your meetings, creates your tasks, shares your documents, sends your emails, and generates your presentations. It integrates seamlessly with your Microsoft 365 environment and business systems, using your existing Microsoft Entra identity to be your AI super assistant. Think of AVA as your personal AI assistant that not only knows your calendar and can access your emails, but can also **modify your calendar**, **send those emails**, **share those documents**, and **create those tasks**—all while respecting your permissions and keeping your data secure within your Azure tenant. ## The AVA Difference AVA solves a critical challenge: **How do you give your organization the power of AI while maintaining enterprise security and leveraging your existing Microsoft 365 investment?** AVA is deployed as an Azure Enterprise Application within your own Azure environment. This means: * All data processing happens in your tenant * No data is sent to external SaaS providers * You maintain complete control over your information * Compliance requirements are easier to meet AVA leverages Microsoft Entra ID (formerly Azure AD) for authentication: * Users sign in with their work accounts * AVA operates with delegated user permissions * It can only access what the user can access * No elevated privileges or broad access required * Perfect for zero-trust environments Unlike other AI platforms that require creating separate agents for each integration: * AVA intelligently orchestrates all connected data sources * Single conversation can span multiple systems * Context is maintained across data sources * Natural language interface—just ask what you need Built specifically for Microsoft 365 enterprises: * Native integration with Outlook, Teams, SharePoint, OneDrive * Works with your existing security policies * Supports your organizational structure * Familiar Microsoft-style authentication and UX ## Three Core Capabilities ### 1. AVA Chat Your conversational AI interface for everything you need at work. **What you can do:** * Ask about your schedule: "What meetings do I have today?" * Query your email: "Show me emails from Sarah about the project" * Access files: "Find my Q4 budget spreadsheet" * Pull Salesforce data: "What opportunities are closing this month?" * Check Jira: "Show me my open issues" * Query GitHub: "What pull requests are waiting for my review?" **How it works:** Use the `@` mention system to reference any connected data source. AVA understands context and can combine information from multiple sources in a single response. **Example conversation:** ``` You: @Calendar what's on my schedule today? AVA: You have 3 meetings: Team standup at 9am, Client call at 11am... You: @Salesforce check status of opportunities discussed in that client call AVA: Looking at your calendar, the 11am meeting is with Acme Corp. In Salesforce, you have 2 open opportunities with Acme Corp... You: Draft an email to follow up AVA: [Generates email draft using context from both systems] ``` [Learn more about AVA Chat →](/features/ava-chat) ### 2. AVA Tasks Multi-step workflow automation with intelligent orchestration. **What you can do:** * Create multi-step processes with individual instructions * Add documents from your OneDrive or SharePoint for context * Use the SuperPrompt toggle for enhanced AI guidance * Save successful workflows as templates * Share task templates with your team **Use cases:** * Document analysis workflows * Research and synthesis projects * Content creation pipelines * Data transformation tasks * Report generation **Example task:** ``` Task: Quarterly Business Review Preparation Step 1: Review financial data - Pull Q4 financials from [uploaded Excel file] - Summarize key metrics Step 2: Analyze pipeline - @Salesforce get all opportunities closing next quarter - Identify top 10 by value Step 3: Create presentation - Generate PowerPoint with findings - Export to workspace ``` [Learn more about AVA Tasks →](/features/ava-tasks) ### 3. Knowledge Search RAG-powered search across your SharePoint documents and knowledge bases. **What you can do:** * Search multiple document libraries simultaneously * Query specific "corpora" (collections of related documents) * Create custom corpora from SharePoint sites * Configure automatic syncing * Extract specific fields from documents **Perfect for:** * Contract management (search all contracts) * HR policy questions (query employee handbooks) * Technical documentation (find specific procedures) * Customer service (search support documentation) * Legal research (query legal documents) **How it works:** Instead of keyword search, Knowledge Search uses semantic understanding to find relevant information across all your documents, even if they don't contain your exact search terms. **Example:** ``` Using: @Contracts Demo corpus You: Find all contracts with early termination clauses AVA: I found 12 contracts with early termination provisions: 1. Acme Corp - 30 day notice required... 2. TechStart LLC - 60 day notice with fee... [Provides summaries with source citations] ``` [Learn more about Knowledge Search →](/features/knowledge-search) ## The Workspace AVA's right-side panel is like having a second screen for your AI work—similar to artifacts in other AI tools, but integrated with your Microsoft 365 environment. ### Three Workspace Tabs: **Find and add content to your conversation** * Microsoft: Search your OneDrive and SharePoint files * Meetings: Access recorded Teams meetings with transcripts * My Computer: Upload files from your local device **Use it to:** * Add documents for AVA to analyze * Include meeting transcripts for summaries * Reference files in your conversation **Rich text editor for capturing and creating** * Full formatting toolbar * Create structured documents * Export to PowerPoint * Copy to clipboard **Use it to:** * Take notes during conversations * Draft documents collaboratively with AVA * Organize ideas from multiple chats * Generate presentations from your notes **View and edit files from SharePoint/OneDrive** * Toggle between View and Edit modes * Microsoft Office integration * Direct access to your files **Use it to:** * Review documents while chatting about them * Make edits without leaving AVA * Reference multiple files simultaneously [Learn more about Workspace →](/features/workspace) ## Connected Data Sources (MCP Integration) AVA uses the Model Context Protocol (MCP) to integrate with your business systems. Once connected, you can access data using simple `@` mentions. ### Currently Supported: * Outlook (Mail & Calendar) * OneDrive (Files & Folders) * SharePoint (Sites & Lists) * Teams (Messages & Meetings) * To Do (Tasks & Lists) * Opportunities * Leads * Cases * Accounts * Contacts * Custom Objects * Issues * Projects * Boards * Sprints * Comments * Repositories * Pull Requests * Issues * Code Search * Commit History * Business Records * Transactions * Custom Records * Defender Alerts * Security Incidents * Threat Intelligence ## Global Features Available Everywhere ### Prompt Library Create and share custom prompts across your organization: * **Create**: Build prompts with instructions and connected data references * **Filter**: By connected data sources, file types, or teams * **Share**: With individuals, teams, or Microsoft 365 groups * **Organize**: Personal prompts vs team prompts **Example prompt:** ``` Name: "Weekly Pipeline Review" Connected Data: @Salesforce File Types: Excel Prompt: "Review all opportunities closing this week. Create a summary table with: Company, Value, Close Date, and Risk Level. Export to Excel." ``` ### Model Selection Choose the right AI model for your task: * **Azure OpenAI**: GPT-4, GPT-4 Turbo, GPT-3.5 * **AWS Bedrock**: Claude Sonnet, Claude Opus * **Google Vertex**: Gemini Pro, PaLM Each model has different strengths—select based on: * Complexity of the task * Speed requirements * Cost considerations * Context window needs ### Custom Instructions Personalize how AVA works for you: ``` Example custom instructions: "I'm a sales manager at a SaaS company. When analyzing opportunities, always include ARR, sales cycle length, and competitive threats. Format financial data in tables. My working hours are 8am-5pm Pacific Time." ``` AVA will remember these preferences across all conversations. ### Export Capabilities Generate Office documents from your work: From chat conversations or notes Create formatted documents Generate spreadsheets with data Plain text exports ## AI That Works: Microsoft Graph API Integration AVA leverages Microsoft Graph API to be your AI super assistant—taking real action across your Microsoft 365 environment. ### What AVA Can Do For You **Schedule and manage your time** * Schedule meetings with attendees * Find optimal meeting times * Create Teams meeting links * Update or cancel events * Check availability * Set reminders **Example**: "Schedule a 30-minute meeting with Sarah and John tomorrow afternoon to discuss Q4 planning" **Create and track work** * Create tasks in Microsoft To Do * Set due dates and reminders * Assign tasks to team members * Update task status * Track completion * Organize into lists **Example**: "Create a task for John to review the contract by Friday" **Collaborate on documents** * Share files from OneDrive/SharePoint * Set permissions (view/edit) * Create share links * Manage access levels * Revoke access when needed * Track file sharing **Example**: "Share the Q4 budget with the Finance team with edit access" **Compose and send messages** * Draft emails with context * Send emails on your behalf * Reply to messages * Forward with context * Include attachments * Format professionally **Example**: "Draft a follow-up email to the client about yesterday's meeting" **Add and generate images** * Upload images for context * Generate images with DALL-E 3 * Create visuals for presentations * Analyze screenshots * Illustrate concepts **Example**: "Generate an image of a modern office workspace for my presentation" **The possibilities are nearly endless**: As Microsoft expands Graph API capabilities, AVA automatically gains new powers. No updates required—it just works. [Explore all core capabilities →](/essentials/core-capabilities) ## How AVA Maintains Security ### Identity-Based Access * Uses Microsoft Entra ID authentication * Operates with **delegated user permissions** * No service accounts with elevated access * Each user sees only what they're authorized to see ### Data Residency * All processing happens in your Azure tenant * Data never leaves your environment * You control data location and retention * Meets compliance requirements (GDPR, HIPAA, etc.) ### Audit & Compliance * All actions are logged * Integration with Azure Monitor * Telemetry and usage tracking * Support for compliance frameworks ## Who Uses AVA? * Check opportunity pipelines * Update CRM records * Draft customer emails * Prepare presentations * Schedule meetings * Search knowledge bases * Find customer history * Query case information * Access documentation * Create responses * Query GitHub repos * Check Jira tickets * Search code * Review PRs * Generate documentation * Review meeting summaries * Check calendar conflicts * Query business metrics * Draft communications * Access files quickly * Search contract libraries * Query policy documents * Find specific clauses * Generate reports * Track document versions * Search HR policies * Find employee documentation * Query benefits information * Draft communications * Access org charts ## Key Differentiators | Feature | Traditional AI Tools | AVA | | ------------------ | ------------------------- | --------------------------- | | **Deployment** | External SaaS | Your Azure tenant | | **Authentication** | Separate login | Microsoft Entra ID | | **Permissions** | API keys/tokens | Delegated user permissions | | **Data Access** | Data sent externally | Data stays in tenant | | **Agent Creation** | Manual setup required | Automatic orchestration | | **Microsoft 365** | Basic via APIs | Native integration | | **Security Model** | Trust the vendor | Zero-trust compatible | | **Customization** | Limited to vendor options | Full control in your tenant | ## Getting Started Ready to experience AI that works the way your organization does? Use your Microsoft work account—same credentials you use every day Three panels: Session History, Chat Area, Workspace Start typing @ to see your connected data sources Build your first multi-step workflow Try Knowledge Search on a SharePoint site Get up and running in 5 minutes Learn more about AVA ## Next Steps Technical architecture and design Complete feature overview Real-world examples # AVA Chat Source: https://docs.datarm.ai/features/ava-chat Conversational AI with connected data access via @ mentions ## Overview AVA Chat is your conversational AI interface that connects to all your enterprise data sources. Using simple `@` mentions, you can query Microsoft 365, Salesforce, Jira, GitHub, and other business systems—all within a single conversation. Chat naturally with AVA like you would with a colleague Type @ to instantly access any connected data source Combine information from multiple systems in one response Access all your past conversations with full context ## The @ Mention System ### How It Works Start typing `@` in any message and AVA shows you available data sources: Chat showing @Contracts Demo mention with connected data context ``` @Calendar - Microsoft Calendar @Mail - Outlook Email @Files - OneDrive & SharePoint @Salesforce - CRM Data @Jira - Issues & Projects @GitHub - Repositories & Code @Teams - Teams Messages @Security - Microsoft Defender ``` ### Example Conversations **You**: @Calendar what meetings do I have today? **AVA**: You have 4 meetings today: * 9am: Team Standup * 11am: Client Call with Acme Corp (Sarah Johnson) * 2pm: Pipeline Review * 4pm: 1:1 with Manager **You**: @Salesforce get details on the Acme Corp opportunity **AVA**: Acme Corp opportunity: * Value: \$250,000 ARR * Stage: Proposal/Price Quote * Close Date: End of Q1 * Last Activity: Proposal sent 3 days ago * Next Steps: Follow-up call scheduled (today's 11am meeting) **You**: Draft a follow-up email for after our call **AVA**: \[Generates personalized email using context from both systems] **You**: @Jira show my open issues **AVA**: You have 5 open issues: 1. FE-123: Fix login button styling (In Progress) 2. BE-456: API response timeout (To Do) 3. FE-789: Add dark mode toggle (In Review) ... **You**: @GitHub show PR status for FE-789 **AVA**: PR #234 for dark mode toggle: * Status: Open, awaiting review * 2 approvals required, 1 received * All checks passing * Reviewer: @john-smith requested changes yesterday **You**: What changes did John request? **AVA**: \[Shows review comments with context] **You**: @Calendar summarize my week **AVA**: This week you had 18 meetings across 5 days: * 6 customer meetings * 4 internal team meetings * 3 board prep sessions * 5 1:1s with direct reports **You**: Access the recorded board prep meetings and create talking points **AVA**: \[Lists meetings] Would you like me to: 1. Summarize key discussion points 2. Extract action items 3. Create board presentation outline **You**: All three, and export to PowerPoint **AVA**: \[Processes meetings and generates presentation] ## Connected Data Sources ### Microsoft 365 Integration **What you can do:** * Check your schedule * Find meetings by attendee or topic * See availability * Get meeting details and participants **Examples:** * "What's on my calendar tomorrow?" * "When is my next meeting with Sarah?" * "Am I free Thursday afternoon?" * "Show me all meetings this week about the Q4 project" **What you can do:** * Search your inbox and sent items * Find emails by sender, subject, or content * Get email summaries * Draft replies **Examples:** * "Find emails from John about the contract" * "Summarize my unread emails from today" * "Show me the last email thread with Acme Corp" * "Draft a follow-up email to Sarah's question" **What you can do:** * Search for documents * Access file content * Find files by name, type, or modified date * Get document summaries **Examples:** * "Find my Q4 budget spreadsheet" * "Show me all PowerPoint presentations from last month" * "What's in the Project Alpha folder?" * "Summarize the contract in the Legal folder" **What you can do:** * Access recorded meetings * Search team conversations * Get meeting transcripts * Find shared files **Examples:** * "Show me recordings from last week's standup" * "What was discussed in the Q4 planning meeting?" * "Find the Teams conversation about the bug fix" ### Business Applications **CRM Data Access:** * Query opportunities, leads, cases, accounts, contacts * Update CRM records * Create new records * Generate reports **Common Queries:** * "Show all opportunities closing this quarter" * "Get details on the Acme Corp account" * "Find all open cases assigned to me" * "Create a new lead for Company X" * "Update opportunity ABC123 stage to Closed Won" **Data You Can Access:** * Opportunities (pipeline, stages, values, dates) * Leads (new prospects, qualification status) * Cases (support tickets, status, resolution) * Accounts (company information, history) * Contacts (people, roles, interactions) * Custom objects (your org-specific data) **Project Management:** * Query issues and projects * Check sprint status * Update issue status * Link issues **Common Queries:** * "Show my open issues" * "What's in the current sprint?" * "Get details on issue ABC-123" * "Update issue XYZ-456 status to In Progress" * "List all bugs in the Mobile project" **Data You Can Access:** * Issues (bugs, tasks, stories) * Projects (project status, members) * Sprints (sprint contents, velocity) * Boards (Kanban, Scrum boards) * Comments and activity **Code Repository Access:** * Search repositories * Check PR status * Review commits * Find issues **Common Queries:** * "Show open PRs in the main repo" * "Get details on PR #123" * "Find commits by author in the last week" * "Show issues labeled 'bug'" * "Search code for function name X" **Data You Can Access:** * Repositories (code, branches) * Pull Requests (status, reviews, checks) * Issues (bugs, feature requests) * Commits (history, changes, authors) * Code search ## Conversation Features ### Context Awareness AVA remembers your conversation history: ``` You: @Salesforce show opportunities over $100K AVA: [Shows 10 opportunities] You: Which of these are closing this month? AVA: [Filters the previous list - maintains context] You: Send me details on the top 3 AVA: [Provides details on highest value opportunities from filtered list] ``` ### Multi-Turn Dialogues Have natural back-and-forth conversations: ``` You: I need to prepare for my client meeting tomorrow AVA: I can help with that. Who are you meeting with? You: Acme Corp AVA: I found your 11am meeting with Sarah Johnson from Acme Corp. Would you like me to: 1. Pull their Salesforce account history 2. Find past email conversations 3. Create a meeting agenda You: All three AVA: [Executes all three tasks and presents comprehensive prep document] ``` ### Cross-Source Intelligence Combine information from multiple sources: ``` You: Analyze my team's productivity this sprint AVA: [Pulling data from multiple sources...] - @Jira: 45 issues completed (velocity: 89 points) - @GitHub: 78 commits, 23 PRs merged - @Calendar: 12 hours in meetings (20% of work week) - @Mail: Average email response time: 2.3 hours Analysis: Strong delivery sprint with good commit velocity. Meeting time is within healthy range. Response time excellent. ``` ## Session Management ### Session History Access all your past conversations: * Organized by recency (newest first) * Search across all sessions * Filter by date or content * Resume any previous conversation ### Sharing Conversations Share sessions with team members: 1. Click "Share" button in session 2. Select recipients (individuals or Microsoft 365 groups) 3. Recipients get full conversation context 4. Useful for handoffs and collaboration ## Export & Integration ### Export Options Select messages to include as slides Generate formatted document from conversation Export data tables and analysis ### Copy & Paste * Copy individual messages * Copy entire conversation * Preserve formatting * Paste into any application ## Tips & Best Practices Instead of: "Check my calendar" Better: "@Calendar what meetings do I have this week?" Explicitly using @ mentions helps AVA route your query to the right source faster. Example: "@Calendar find my meeting with Acme Corp, then @Salesforce show their account history" AVA can connect information across systems for comprehensive answers. Don't start a new conversation for related questions. AVA maintains context. First: "Show me Q4 opportunities" Then: "Which ones are at risk?" Then: "Draft an email to the account owners" Set up custom instructions (in settings) for consistent behavior: * Your role and preferences * How you want data formatted * Your working hours and timezone * Specific terminology your team uses If you ask the same question regularly, save it in the Prompt Library: * "Weekly pipeline review" * "Sprint status summary" * "Email inbox triage" Reuse prompts to save time. ## Common Questions **Q: What data can AVA access?** A: Only data you can access. AVA uses your delegated permissions—if you can't see a file in SharePoint, AVA can't either. **Q: Is my data sent to OpenAI/Anthropic?** A: Your organizational data stays in your Azure tenant. Only the AI-generated responses come from external providers. **Q: Can AVA modify my data?** A: Only if you explicitly ask (e.g., "Update this Salesforce record"). All modifications are logged. **Q: How fast are responses?** A: Typically 2-10 seconds depending on: * Model selected (GPT-5 Nano/Mini, Flash, Haiku models faster than GPT-5 Pro/Opus models) * Data source query complexity * Amount of data to process **Q: Can I use AVA offline?** A: No, AVA requires internet connection to access connected data sources and AI models. **Q: Is there a message limit?** A: No hard limit, but very long conversations may hit model context limits (128K-1M tokens depending on model). Start a new session if needed. **Q: Who can see my conversations?** A: Only you, unless you explicitly share a session. **Q: Are conversations logged?** A: Yes, for audit and compliance purposes via Application Insights. **Q: Can admins see my chats?** A: Admins can access audit logs showing usage patterns, but not conversation content unless required for security/compliance investigation. ## Next Steps Get started with your first conversation Master the @ mention system Create reusable prompts Choose the right AI model # AVA Tasks Source: https://docs.datarm.ai/features/ava-tasks Multi-step workflow automation with intelligent orchestration ## Overview AVA Tasks transforms complex, multi-step workflows into automated processes. Break down any project into sequential steps, add documents for context, and let AVA execute each step intelligently—remembering results from previous steps and pulling live data from connected sources. Create tasks with unlimited sequential steps Save and share successful workflows as templates Enhanced AI guidance for complex tasks Add files from workspace to any step ## Creating Your First Task ### Basic Structure Every AVA Task consists of: Descriptive name for your workflow (e.g., "Weekly Pipeline Analysis") Individual instructions executed in sequence Files added from workspace for context Enable enhanced AI guidance for better results ### Example: Sales Report Generation ``` Task Name: Weekly Sales Report Step 1: Gather Data Instructions: @Salesforce get all opportunities that closed this week. Include company name, deal value, close date, and owner. Step 2: Analyze Trends Instructions: Compare this week's closed deals to last week. Identify: - Total value comparison - Win rate changes - Top performers Step 3: Create Summary Instructions: Generate an executive summary with key insights and recommendations for next week. Step 4: Export Instructions: Create a PowerPoint presentation with: - Slide 1: Executive summary - Slide 2: Closed deals table - Slide 3: Trend analysis - Slide 4: Recommendations Documents: [Last week's report for comparison] SuperPrompt: ON ``` ## Task Capabilities ### Sequential Execution Each step has access to all previous step results: **Step 1 Output**: List of 10 contracts **Step 2**: "Analyze the contracts from Step 1 for payment terms" AVA remembers the contracts and analyzes them **Benefit**: Build complex workflows without manual copying Use @ mentions in any step to pull fresh data: **Step 1**: @Salesforce get Q4 opportunities **Step 2**: @Jira find issues blocking deals from Step 1 **Step 3**: @GitHub check if fixes are deployed **Benefit**: Always work with current information Visual indicators show: * ✅ Completed steps (green checkmark) * ⏳ Current step (spinner animation) * ⏸️ Pending steps (gray) * ❌ Failed steps (red X with error message) **Benefit**: Monitor long-running tasks Each completed step shows: * Full AI response * Execution time * Any data pulled from connected sources * Option to re-run individual steps **Benefit**: Debug and refine workflows easily ### Document Integration Add files from workspace to provide context: 1. Click "Add Documents" button 2. Select "Microsoft Files" tab in workspace 3. Search for your file 4. Click to add to task **Use Cases**: * Add templates for consistent output * Include reference materials * Provide examples for AI to follow * Add data files for analysis 1. Click "Add Documents" button 2. Select "My Computer" tab 3. Drag and drop or browse for files 4. Files uploaded to Azure blob storage **Use Cases**: * Upload CSV/Excel for data processing * Add images for analysis * Include PDFs for extraction * Provide proprietary templates 1. Click "Add Documents" button 2. Select "Meetings" tab 3. Choose recorded Teams meeting 4. Transcript added to task context **Use Cases**: * Analyze meeting discussions * Extract action items * Create follow-up materials * Generate meeting summaries ### SuperPrompt Feature Toggle SuperPrompt ON for enhanced AI guidance: **What It Does**: * Provides more detailed reasoning * Better handles ambiguous instructions * Improves complex analysis * More thorough responses **When to Use**: * Complex analysis tasks * Ambiguous requirements * Creative content generation * Research and synthesis **When to Skip**: * Simple data retrieval * Straightforward formatting * Quick queries (faster without) **Example Difference**: Without SuperPrompt: ``` Step: Analyze the contract Output: [Lists key terms] ``` With SuperPrompt: ``` Step: Analyze the contract Output: [Comprehensive analysis with]: - Key terms with explanations - Potential risks identified - Comparison to standard terms - Recommendations - Referenced sections ``` ## Prompt Library Save and reuse custom prompts to accelerate your team's workflow. AVA Prompt Library showing template cards for Case Deflection, Meeting Summarization, and Resume Analysis ### What is the Prompt Library? The Prompt Library is a centralized repository of: * **Pre-built prompts** for common business tasks * **Team-shared templates** created by your organization * **Personal prompts** you've saved for reuse * **Connected Data integrations** for context-aware responses ### Creating Custom Prompts To create your own reusable prompt: Open the Prompt Library from the navigation Explain the prompt's purpose clearly Include specific instructions and placeholders if needed Add tags for easy discovery (department, function, data source) Select: Personal, Team, or Organization-wide Run the prompt to verify it works as expected Make it available for future use ### Using Connected Data in Prompts Prompts can integrate with your Connected Data sources: ``` Find @[Salesforce] Cases marked "High Priority" from the last week and summarize the key issues ``` This allows AVA to access real-time data from your connected systems while executing the prompt. ### Sharing & Collaboration Private to you, perfect for individual workflows and experimentation. Shared with your Microsoft 365 team, enabling collaboration on common processes. Available to everyone in your company, ideal for company-wide standards and best practices. ## Task Library ### Saving Task Templates After creating a successful task: AVA Tasks interface showing task name, sequential steps, document selection, and SuperPrompt toggle Run the task and verify results Located at top of completed task Descriptive name: "Contract Review Workflow" Explain what the template does and when to use it Choose: Personal, Team, or Organization ### Using Task Templates 1. Click "New Task" button 2. Select "From Template" tab 3. Browse or search templates 4. Click template to load 5. Customize if needed 6. Run task ### Template Management Templates maintain version history; revert to previous versions See how often templates are used across organization Share with Microsoft 365 groups or entire organization Organize by department, use case, or project ## Common Task Patterns ### Research & Analysis Pattern Perfect for: Market research, competitive analysis, literature reviews ``` Task: Competitive Analysis Step 1: Gather Information - @Web search for competitor X product information - @Files access our competitive analysis folder Step 2: Compare Features - Create comparison table: Our product vs Competitor - Highlight differentiators Step 3: Pricing Analysis - @Salesforce get our pricing for similar deals - Compare to competitor's public pricing Step 4: Generate Report - Create executive summary - SWOT analysis - Recommendations - Export to PowerPoint ``` ### Document Processing Pattern Perfect for: Contract review, resume screening, invoice processing ``` Task: Contract Review Step 1: Extract Key Terms - Payment terms and schedule - Termination clauses - Liability limitations - Renewal provisions Step 2: Risk Assessment - Identify unfavorable terms - Compare to standard contract - Flag items requiring legal review Step 3: Create Summary - One-page executive summary - Risk rating (Low/Medium/High) - Recommended actions Documents: [Standard contract template, Company contract guidelines] SuperPrompt: ON ``` ### Content Creation Pattern Perfect for: Blog posts, presentations, reports, proposals ``` Task: Blog Post Creation Step 1: Research - @Files access brand guidelines and past posts - @Web research topic trends - Identify target keywords Step 2: Outline - Create detailed outline - Define key messages - Plan structure (intro, 3 sections, conclusion) Step 3: Draft Content - Write complete draft following outline - Include SEO keywords naturally - Match brand voice Step 4: Optimize - Add compelling headlines - Create meta description - Suggest images - Export to Word ``` ### Data Analysis Pattern Perfect for: Sales reports, performance analysis, trend identification ``` Task: Monthly Sales Analysis Step 1: Gather Data - @Salesforce get all closed opportunities this month - Include: deal value, sales cycle, owner, product Step 2: Calculate Metrics - Total revenue - Average deal size - Win rate by product - Average sales cycle Step 3: Trend Analysis - Compare to last month - Identify top performers - Flag concerning trends Step 4: Visualize - Create charts and graphs - Executive dashboard - Export to Excel Documents: [Last month's report] ``` ## Real-World Examples ### Sales Team: Proposal Generation ``` Task Name: Custom Proposal for [Customer Name] Step 1: Research Customer @Salesforce get account history and open opportunities @Mail find all email correspondence @Calendar review past meeting notes Step 2: Identify Requirements Analyze conversations for: - Pain points mentioned - Budget indicators - Timeline requirements - Decision criteria Step 3: Draft Proposal Use template: [Standard Proposal Template.docx] Customize sections: - Executive summary (customer-specific) - Proposed solution addressing pain points - Pricing based on requirements - Timeline matching their needs Step 4: Create Presentation Generate PowerPoint for proposal presentation: - Title slide - Problem statement (their pain points) - Our solution - Pricing options - Next steps Documents: [Standard Proposal Template, Past Winning Proposals] SuperPrompt: ON Time Saved: 4 hours → 30 minutes ``` ### Legal Team: Contract Analysis ``` Task Name: Vendor Contract Review Step 1: Extract Standard Clauses - Payment terms - Liability and indemnification - Termination provisions - Confidentiality requirements - Dispute resolution Step 2: Compare to Standards Use: [Company Standard Vendor Contract] Identify deviations: - More favorable terms - Less favorable terms - Missing provisions Step 3: Risk Assessment Classify each deviation: - Acceptable - Requires negotiation - Unacceptable (deal breaker) Step 4: Generate Red-Line Create marked-up version with: - Suggested changes in track changes - Comments explaining concerns - Alternative language Documents: [Vendor Contract.pdf, Company Standards.docx] SuperPrompt: ON Time Saved: 2 hours → 15 minutes ``` ### HR Team: Job Description Creation ``` Task Name: Create Job Description for [Role] Step 1: Research Role @Files access similar job descriptions @Web search for market salary data Analyze requirements for this role Step 2: Draft Core Sections - Role summary and purpose - Key responsibilities (8-10) - Required qualifications - Preferred qualifications - Skills and competencies Step 3: Add Company Info Include: - Company overview - Department structure - Growth opportunities - Benefits summary Step 4: Optimize for Recruiting - Add engaging language - Include DEI statement - SEO-friendly title - Clear application process Documents: [Company JD Template, Benefits Guide] Time Saved: 3 hours → 20 minutes ``` ### Development Team: Sprint Planning ``` Task Name: Sprint Planning Document Step 1: Review Backlog @Jira get all issues in backlog Prioritize by: customer impact, effort, dependencies Step 2: Assess Capacity Team velocity: [from previous sprints] Available hours: [team size × sprint length] Planned time off: @Calendar check team calendars Step 3: Select Issues Choose issues fitting capacity Balance: features, bugs, tech debt Check dependencies: @GitHub Step 4: Create Sprint Plan Generate document with: - Sprint goal - Selected issues with estimates - Assignment recommendations - Risk factors - Success criteria Export to PowerPoint for kickoff meeting Time Saved: 2 hours → 15 minutes ``` ## Advanced Features ### Error Handling If a step fails: Error message displayed with details Click step to see what went wrong Edit step instructions and click "Re-run Step" Once fixed, task continues from that step Common errors: * Invalid @ mention (data source not available) * Insufficient permissions * Timeout on long operations * Network connectivity issues ### Task Iteration Improve tasks over time: 1. **Run task first time**: Note what works and what doesn't 2. **Refine steps**: Edit instructions for better results 3. **Adjust SuperPrompt**: Toggle on/off to find optimal setting 4. **Update documents**: Add better templates or examples 5. **Save as template**: Once perfected, save for reuse ### Collaborative Tasks Work with team members: * **Create task** with clear steps * **Share via Microsoft 365 group** * **Team members run** same workflow * **Compare results** and refine * **Update template** with improvements ## Performance Tips Instead of: "Analyze everything and create report" Better: * Step 1: Gather data * Step 2: Calculate metrics * Step 3: Identify trends * Step 4: Create visualizations * Step 5: Write summary * Step 6: Export report Instead of: "Analyze the data" Better: "Calculate mean, median, and standard deviation for sales data. Identify outliers. Create a distribution chart." "Using the data from Step 2, calculate..." "Take the list generated in Step 1 and..." "Based on the analysis in Step 3, recommend..." Upload examples of desired output format. AVA will match the style and structure. Before processing 1000 contracts, test workflow with 2-3. Refine process, then scale up. ## Common Questions **Q: How many steps can a task have?** A: No hard limit, but 5-15 steps is typical. Very long tasks may hit timeout limits. **Q: Can I pause a task and resume later?** A: Not currently. Tasks must complete once started. Save progress by exporting results. **Q: Can tasks run on a schedule?** A: Not yet, but this feature is planned. **Q: Can one task trigger another?** A: Not directly. You can save results and manually start next task with that context. **Q: How many documents can I add?** A: Up to 20 documents per task. Total size limit: 100MB. **Q: What file types are supported?** A: PDF, Word (.docx), Excel (.xlsx), PowerPoint (.pptx), text (.txt), images (.jpg, .png), CSV **Q: Are documents stored permanently?** A: Documents remain attached to the task. Deleted after 90 days of inactivity. **Q: Can I reference specific pages?** A: Yes, in instructions: "Review pages 5-8 of Contract.pdf" **Q: How long does a task take?** A: Depends on complexity and data sources: * Simple 5-step task: 30-60 seconds * Complex 10-step with analysis: 2-5 minutes * Large document processing: 5-15 minutes **Q: Does SuperPrompt slow things down?** A: Slightly (10-20% longer), but often worth it for better quality. **Q: Can I cancel a running task?** A: Yes, click "Cancel Task" button. Completed steps remain visible. ## Next Steps Follow quickstart guide to create your first task Learn to build reusable workflows Tips for effective task creation Browse task templates by use case # Knowledge Search Source: https://docs.datarm.ai/features/knowledge-search RAG-powered semantic search across your SharePoint documents and knowledge bases ## Overview Knowledge Search uses Retrieval Augmented Generation (RAG) to provide intelligent, semantic search across your SharePoint document libraries. Unlike traditional keyword search, Knowledge Search understands the meaning of your questions and finds relevant information even when exact keywords don't match. Understands intent, not just keywords Search multiple document collections simultaneously Every answer includes citations to source documents Automatically updates when documents change ## What is a Corpus? A **corpus** is a collection of documents from a SharePoint site or document library that has been: * Embedded (converted to vector representations) * Indexed for semantic search * Made searchable through AVA Think of it as creating a "knowledge base" from your documents. ### Example Corpora **Contracts Corpus** * Source: Legal/Contracts SharePoint library * 350 vendor and customer contracts * Auto-sync: ON * Field extraction: Company Name, Contract Type, Effective Date, Value **Use Case**: "Find all contracts with auto-renewal clauses" **Employee Handbook Corpus** * Source: HR Policies SharePoint site * 45 policy documents * Auto-sync: ON * Field extraction: Policy Name, Department, Last Updated **Use Case**: "What's the remote work policy?" **Technical Documentation Corpus** * Source: Support/Documentation library * 200+ troubleshooting guides * Auto-sync: ON * Field extraction: Product, Issue Category, Severity **Use Case**: "How do I troubleshoot login timeouts?" **Proposal Library Corpus** * Source: Sales/Proposals SharePoint * Past winning proposals * Auto-sync: OFF (historical reference) * Field extraction: Customer Industry, Deal Size, Product **Use Case**: "Find proposals for healthcare customers" ## Creating a Corpus Click "Knowledge Search" in AVA navigation Opens corpus creation dialog Create New Corpus dialog showing SharePoint site selection and configuration options * Choose from dropdown of available sites * Must have read access to the site * Can preview site contents * Choose specific library within site * See document count preview * Supported types: PDF, Word, Excel, PowerPoint, text Write clear description: "Contains all vendor contracts from 2020-present. Use for contract term searches, renewal dates, and clause analysis." Good descriptions help users know when to use this corpus. **Auto-Sync ON**: AVA periodically checks SharePoint for: * New documents (added automatically) * Modified documents (re-embedded) * Deleted documents (removed from corpus) **Auto-Sync OFF**: Corpus is static, manual updates only **Recommendation**: ON for active document libraries Auto-sync toggle with tooltip explaining automatic document updates Extract custom fields from documents for enhanced filtering: Common fields: * document\_type: Contract, Policy, Guide * department: Sales, Legal, Engineering * date: Effective date, creation date * author: Document creator * company: Customer or vendor name * keywords: Custom tags **Note**: Field extraction cannot be changed after creation. * Click "Create Corpus + Add Files" * AVA begins embedding documents * Processing time: \~1 minute per 10 documents * You can leave and return later ## Using Knowledge Search ### Basic Search Click "@ Connected Data" button, choose corpus from dropdown Example: Select "@Contracts Demo" **Tip**: You can select multiple corpora to search simultaneously Type your question in natural language: Examples: * "Find contracts with termination for convenience clauses" * "What are the indemnification terms in customer contracts?" * "Show me all contracts expiring in Q1 2025" AVA returns: * **Answer**: AI-generated response synthesizing relevant information * **Sources**: Links to specific documents with relevance scores * **Excerpts**: Relevant passages highlighted Each source is clickable to open full document Ask follow-up questions: * "Which of these have the shortest notice period?" * "Compare the payment terms across these contracts" * "Show me the specific contract language" ### Advanced Search Techniques Search across multiple knowledge bases simultaneously: **Select multiple corpora**: * @Contracts Demo * @Vendor Agreements * @MSAs (Master Service Agreements) **Ask**: "Find all payment terms across all contract types" **Result**: AVA searches all three corpora and synthesizes results **Use Case**: Comprehensive research across document types If corpus has field extraction configured: **Query**: "Find contracts where document\_type='MSA' and company contains 'Tech'" **AVA filters** documents matching criteria before searching content **Use Case**: Narrow search to specific document categories **Query**: "Show contracts signed in 2024" AVA understands temporal queries and filters accordingly **More examples**: * "Policies updated in the last 6 months" * "Guides created this year" * "Contracts expiring next quarter" **Query**: "Compare payment terms in Microsoft contract vs Salesforce contract" AVA retrieves both documents and provides side-by-side comparison **More examples**: * "How does our current PTO policy differ from the 2020 version?" * "Compare indemnification clauses across top 5 contracts" **Query**: "Create a table of all contracts with columns: Company, Value, Expiration Date, Auto-Renewal" AVA extracts structured data from multiple documents **Export**: Can export table to Excel **Use Case**: Data extraction and analysis ## Managing Corpora ### Corpus Dashboard View all your corpora and manage embedded files: Knowledge Search showing embedded files from Contracts Demo with file details, descriptions, and pagination * All corpora you own or have access to * Document count per corpus * Last sync timestamp * Storage size * Most searched corpora * Common queries * User adoption metrics * Performance stats * See all documents in corpus * Individual file status * Add or remove specific files * Re-embed modified documents * Who has access * Permission levels * Share with teams or individuals ### Corpus Maintenance **With Auto-Sync ON**: * Add documents to SharePoint library * AVA automatically detects and embeds * Typically within 1 hour **With Auto-Sync OFF**: 1. Navigate to corpus settings 2. Click "Add Files" button 3. Select files from SharePoint 4. Click "Embed Files" **Manual Selection**: * Choose specific documents to add * Useful for selective corpus building 1. Open corpus embedded files view 2. Select documents to remove 3. Click "Remove Files" button 4. Documents removed from search immediately **Note**: Original SharePoint files are not deleted If document content changes: **With Auto-Sync ON**: Automatic re-embedding **With Auto-Sync OFF**: 1. Embedded files view 2. Select modified documents 3. Click "Re-Embed" **When to re-embed**: * Document content significantly changed * Want to pick up recent edits * After field extraction rules change Edit corpus settings: * Description * Auto-sync toggle (can enable/disable anytime) * Sharing permissions * Field extraction (view only, cannot edit) **Storage Management**: * See total storage used * Remove unused documents * Archive old corpora ## How Knowledge Search Works (Technical) ### The RAG Pipeline When you create a corpus: 1. AVA connects to SharePoint using your delegated permissions 2. Downloads documents you have access to 3. Extracts text content from each file 4. Chunks documents into manageable segments (\~500 tokens each) For each chunk: 1. Sent to embedding model (text-embedding-ada-002) 2. Converted to vector representation (1536 dimensions) 3. Vector stored in PostgreSQL with pgVector extension 4. Metadata stored: filename, page number, chunk position When you search: 1. Your question is converted to vector embedding 2. pgVector performs similarity search across all embedded chunks 3. Top K most relevant chunks retrieved (typically 5-10) 4. Relevance scores calculated Retrieved chunks sent to AI model as context: 1. Model receives: your question + relevant document excerpts 2. AI generates response based on actual content 3. Citations added automatically 4. Source documents linked ### Why This Works Better Than Keyword Search | Keyword Search | Knowledge Search (RAG) | | ------------------------------ | ---------------------------------- | | Exact word matches only | Understands meaning and intent | | Misses synonyms and variations | Finds semantically similar content | | No context understanding | Considers document context | | Returns documents, not answers | Generates specific answers | | Manual review of results | AI-synthesized responses | | No source attribution | Automatic citations | **Example**: **Query**: "What's our policy on working from home?" **Keyword Search**: * Might miss documents that say "remote work" or "telecommute" * Returns list of potentially relevant documents * You read through each to find answer **Knowledge Search**: * Finds documents about remote work, telecommuting, work-from-home * Returns: "According to the Remote Work Policy (updated Jan 2024), employees can work from home up to 3 days per week with manager approval..." * Includes link to exact policy document ## Use Cases by Department ### Legal Department **Corpus**: All contracts (vendor, customer, partner) **Common Searches**: * "Find all contracts with limitation of liability caps under \$1M" * "Which contracts allow assignment to affiliates?" * "Show me indemnification obligations in SaaS contracts" * "Create table of all contract renewal dates in next 90 days" **Time Savings**: Days of manual review → Minutes **Corpus**: Internal legal memos, case summaries, precedents **Common Searches**: * "Have we dealt with this issue before?" * "Find similar disputes and their resolutions" * "What was Legal's opinion on X in the past?" **Benefit**: Institutional knowledge accessible instantly **Corpus**: Regulatory documents, compliance policies **Common Searches**: * "What are GDPR requirements for data retention?" * "Show me all data privacy policies" * "What's required for SOC2 compliance?" **Benefit**: Ensure compliance with current regulations ### HR Department **Corpus**: Employee handbook, HR policies **Use**: Answer employee questions instantly "What's the parental leave policy?" "How do I request PTO?" **Corpus**: Benefits guides, provider documents **Use**: Help employees understand benefits "What dental plans are available?" "How does HSA work?" **Corpus**: Onboarding materials, training docs **Use**: New hire questions "What systems do I need access to?" "What's the dress code?" **Corpus**: HR process documentation **Use**: HR team reference "How do I process a termination?" "What's the promotion approval process?" ### IT/Support **Corpus**: Technical support documentation **Searches**: * "How to fix login timeout errors?" * "Steps to reset user password" * "Troubleshoot VPN connection issues" **Benefit**: Faster ticket resolution **Corpus**: User guides, tutorials **Searches**: * "How to set up multi-factor authentication?" * "Steps to configure email on mobile" * "How to access shared drives?" **Benefit**: Self-service for users **Corpus**: Architecture docs, runbooks **Searches**: * "Where is user data stored?" * "What's the backup procedure?" * "How to deploy to production?" **Benefit**: Quick reference for team ### Sales & Marketing **Corpus**: Competitive research, battle cards **Use**: "How do we compare to Competitor X on feature Y?" **Corpus**: Winning proposals and case studies **Use**: "Find proposals for healthcare industry customers" **Corpus**: Product specs, feature descriptions **Use**: "What are the key features of Product X?" **Corpus**: Success stories, testimonials **Use**: "Find case studies showing ROI > 200%" ## Best Practices Create separate corpora for different use cases: ✅ Good: * "Vendor Contracts" corpus * "Customer Contracts" corpus * "NDAs" corpus ❌ Avoid: * Single "All Contracts" corpus with everything **Why**: More focused search, better results Help users understand when to use each corpus: ✅ Good: "Contains all active vendor contracts from 2022-present. Use for: payment terms, renewal dates, SLA requirements. Auto-synced daily." ❌ Avoid: "Vendor stuff" **Why**: Users find the right corpus faster For document libraries that change frequently: * Policy documents * Active contracts * Current product documentation Turn OFF auto-sync for: * Historical/archived documents * Reference libraries that don't change **Why**: Keep corpus current without manual work Plan field extraction before creating corpus: Useful fields: * Document category/type * Date (effective, expiration, creation) * Company/customer name * Department/owner * Status (active, expired, draft) **Why**: Enables filtered searches and better organization After creating corpus, test with common questions: 1. Ask typical user questions 2. Verify relevant documents are returned 3. Check if answers are accurate 4. Refine corpus if needed **Why**: Ensure corpus meets user needs ## Performance & Limits * Max documents per corpus: 10,000 * Max file size: 50MB * Supported types: PDF, Word, Excel, PowerPoint, TXT * Vector search: Less than 1 second * AI response generation: 3-8 seconds * Typical total: 5-10 seconds per query * Vectors stored in PostgreSQL with pgVector * Original files remain in SharePoint * Corpus metadata and embeddings: \~1MB per 100 pages * No hard limit on concurrent searches * Auto-scales with Azure Container Apps * Performance degrades gracefully under load ## Troubleshooting **Problem**: Search returns no relevant results **Solutions**: 1. Try rephrasing question 2. Check if documents are actually embedded (view embedded files) 3. Verify you have access to source SharePoint library 4. Try broader search terms 5. Check if corpus needs re-embedding (if documents updated) **Problem**: Results don't match question **Solutions**: 1. Be more specific in query 2. Use field filters if available 3. Check corpus description - might be wrong corpus 4. Consider splitting corpus into more focused collections 5. Try different phrasing **Problem**: Corpus taking long time to process **Expected Times**: * 10 documents: \~1 minute * 100 documents: \~10 minutes * 1000 documents: \~1-2 hours **If slower**: 1. Check document sizes (large PDFs take longer) 2. Verify SharePoint connection stable 3. Check if multiple corpora processing simultaneously **Problem**: New documents not appearing **Solutions**: 1. Check auto-sync is enabled (corpus settings) 2. Verify you still have SharePoint access 3. Check sync frequency (typically hourly) 4. Manually trigger sync if needed 5. Check embedded files view for sync errors ## Next Steps Step-by-step guide to creating and using a corpus Learn field extraction, multi-corpus search, and more See how teams use Knowledge Search Tips for optimal corpus management # Workspace Source: https://docs.datarm.ai/features/workspace Integrated panel for search, notes, and file viewing—your AI-powered second screen ## Overview The Workspace is AVA's right-side panel—think of it as your AI-powered second screen. It provides three essential tools: Search (find and add files), Notes (capture and create content), and File Viewer (view and edit documents)—all without leaving your conversation. Find files, meetings, and documents to add to conversations Rich text editor for capturing ideas and creating content View and edit SharePoint/OneDrive documents ## Search Tab Your gateway to adding context to AVA conversations. ### Three Search Sources **Search your OneDrive and SharePoint** **What you can find**: * Personal files (OneDrive) * Shared documents (SharePoint sites) * Team files (Microsoft 365 Groups) **Features**: * Full-text search across file contents * Filter by file type (Word, Excel, PowerPoint, PDF, etc.) * Filter by modified date * Filter by location (specific sites or folders) * Preview before adding to conversation **How to use**: 1. Type search term in search box 2. Apply filters if needed 3. Click file to preview 4. Click "Add to Conversation" to include in chat context **Example**: Search for "Q4 budget" → Filter to Excel files → Preview the file → Add to conversation → Ask AVA: "Summarize the key numbers in this budget" **Access recorded Microsoft Teams meetings** **What you can find**: * Your recorded meetings * Meetings you attended that were recorded * Meetings shared with you **Features**: * Filter by date range * Filter by attendees * See meeting duration and participant count * Transcript preview * Direct link to Teams recording **How to use**: 1. Select date range (Last 7 days, Last 30 days, Custom) 2. Search by meeting title or attendees 3. Click meeting to preview transcript 4. Click "Add to Conversation" to include transcript **Example**: Find yesterday's product roadmap meeting → Add to conversation → Ask AVA: "Summarize the key decisions and action items from this meeting" **Use Cases**: * Meeting summaries * Extract action items * Find specific discussions * Compare decisions across meetings * Create follow-up materials **Upload files from your local device** **What you can add**: * Documents from your computer * Screenshots * Downloaded files * Files from external sources **Features**: * Drag and drop upload * Browse to select files * Multiple file upload * Files stored securely in Azure Blob Storage **Supported file types**: * Documents: PDF, Word (.docx), PowerPoint (.pptx), Excel (.xlsx) * Text: TXT, CSV, MD, JSON * Images: JPG, PNG, GIF * Other: Log files, code files **How to use**: 1. Click "My Computer" tab 2. Drag files into upload area OR click "Browse" 3. Select files to upload 4. Wait for upload to complete 5. Files automatically added to current conversation **Example**: Upload contract PDF from email → Ask AVA: "Review this contract and extract key terms" **Privacy Note**: Uploaded files are stored in your tenant's Azure Blob Storage and deleted after 90 days of inactivity ### Search Tips Better search terms = better results ❌ Vague: "document" ✅ Specific: "Q4 sales forecast 2024" ❌ Generic: "meeting" ✅ Specific: "standup meeting November 2024" Narrow results quickly: **File type filter**: * Use when you know format: "budget" + Excel filter * Eliminates irrelevant file types **Date filter**: * "Recent" vs "Modified in last 30 days" vs "Custom range" * Useful for finding latest versions **Location filter**: * Specific SharePoint site * Specific folder path * Your OneDrive only Always preview files first: * Verify it's the right document * Check file size (large files slow conversation) * Ensure it's the latest version * Confirm you have access Click filename to open preview pane You can add multiple files to provide complete context: **Example workflow**: 1. Add current year budget (Excel) 2. Add last year budget for comparison (Excel) 3. Add budget guidance document (Word) 4. Ask: "Compare budgets and check compliance with guidelines" **Limit**: Up to 20 files per conversation (for performance) ## Notes Tab Rich text editor for capturing ideas, taking notes, and creating content collaboratively with AVA. ### Features * Bold, italic, underline * Headings (H1, H2, H3) * Bullet and numbered lists * Blockquotes * Code blocks * Links * Work with AVA to draft content * Copy chat responses to notes * Refine content iteratively * Keep notes across sessions * **Copy to Clipboard**: Paste anywhere * **Export to PowerPoint**: Auto-generate presentation * **Save to OneDrive**: Store for later * Notes persist across sessions * Search your notes * Archive old notes * Share notes with team ### Common Use Cases **Workflow**: 1. Start AVA conversation during or before meeting 2. Open Notes tab 3. Take notes in real-time 4. Use formatting for structure (headings, bullets) 5. Ask AVA to summarize key points 6. Copy AVA's summary to notes 7. After meeting: Export to Word or PowerPoint **Example**: ``` # Product Roadmap Meeting - Nov 15, 2024 ## Attendees - Sarah (PM) - John (Eng) - Lisa (Design) ## Key Decisions - Move Feature X to Q1 2025 - Prioritize mobile app improvements - Defer API v3 to Q2 ## Action Items [ ] Sarah: Update roadmap deck by EOW [ ] John: Size Feature X implementation [ ] Lisa: Share mobile mockups next week [AVA's meeting summary and insights copied here] ``` **Workflow**: 1. Start with outline in notes 2. Ask AVA to draft each section 3. Copy good content from chat to notes 4. Refine and edit in notes 5. Ask AVA for feedback 6. Iterate until satisfied 7. Export to final format **Example - Blog Post**: ``` # Blog Post: 5 Tips for Remote Work Productivity ## Outline 1. Set dedicated workspace 2. Maintain routine 3. Take breaks 4. Over-communicate 5. Use right tools ## Draft (working with AVA) [Introduction - AVA drafted] Working remotely has become the new normal... ## Tip 1: Set Up a Dedicated Workspace [AVA expanded this section with details] [Continue building content section by section] ``` **Export**: When done, export to Word for final polish **Workflow**: 1. Start with seed idea in notes 2. Chat with AVA to explore concepts 3. Copy interesting ideas to notes 4. Organize and structure thoughts 5. Use AVA to elaborate on promising ideas 6. Build comprehensive notes document **Example - Product Feature Ideas**: ``` # New Feature Brainstorm ## Problem Statement Users need faster way to find documents ## Ideas from AVA Discussion - Smart search with AI suggestions - Recent files quick access panel - Tag-based organization - Voice search capability ## Promising: Smart Search [AVA's detailed exploration of this idea] - NLP-powered query understanding - Learn from user behavior - Predictive suggestions - Integration with existing search ## Next Steps - Validate with user research - Technical feasibility assessment - Competitive analysis ``` **Workflow**: 1. Outline project or task in notes 2. Break down into steps 3. Use AVA to think through details 4. Document plan in notes 5. Create checklist format 6. Track progress **Example**: ``` # Q1 Marketing Campaign Planning ## Objectives - Increase brand awareness 25% - Generate 500 qualified leads - Launch in 3 target markets ## Timeline - Week 1-2: Research and strategy - Week 3-4: Creative development - Week 5-6: Asset production - Week 7-8: Launch and promotion ## Tasks [ ] Market research (Sarah) [ ] Define target personas (Team) [ ] Create messaging framework (Lisa) [ ] Design assets (Design team) [ ] Set up tracking (John) [ ] Launch social campaign (Marketing) ## Budget [AVA's budget breakdown based on past campaigns] ## Success Metrics [AVA's suggested KPIs] ``` ### PowerPoint Export Special feature: Generate presentations directly from notes. **How it works**: 1. Write or paste content in notes 2. Use headings to structure (H1 = slide titles) 3. Use bullets for slide content 4. Click "Export to PowerPoint" 5. AVA generates presentation: * H1 headings become slide titles * Content under each H1 becomes slide body * Bullets automatically formatted * Appropriate slide layouts applied **Example Notes**: ```markdown theme={null} # Q4 Sales Results - Revenue: $5.2M (↑ 23% YoY) - New customers: 47 - Expansion: $890K # Top Wins - Acme Corp: $500K ARR - TechStart: $250K ARR - GlobalCo: $180K ARR # Q1 Focus - Expand enterprise segment - Launch new product tier - Increase partner channel ``` **Result**: 3-slide PowerPoint with formatted content **Customization**: * Edit slides in PowerPoint after export * Change themes and layouts * Add images and charts * Refine content ## File Viewer Tab View and edit documents from SharePoint and OneDrive without leaving AVA. ### Features **Read documents without editing** **Features**: * Full document preview * All file types supported (Word, Excel, PowerPoint, PDF) * Page navigation * Search within document * Zoom controls * Print option **Use Cases**: * Reference documents while chatting * Review contracts during analysis * Check data in spreadsheets * View presentations **Example**: Open Q4 budget Excel → View data → Ask AVA in chat: "Analyze the marketing spend in this budget" **Make changes using Microsoft Office Online** **Features**: * Full Office Online editing capabilities * Auto-save to SharePoint/OneDrive * Track changes support * Comments and collaboration * Version history **Supported for editing**: * Word documents (.docx) * Excel spreadsheets (.xlsx) * PowerPoint presentations (.pptx) **Not editable** (view only): * PDF files * Legacy Office formats (.doc, .xls, .ppt) * Image files * Text files **How to edit**: 1. Open file in viewer 2. Click "Edit" button (top right) 3. Make changes using Office editor 4. Changes save automatically 5. Close editor to return to view mode **Example**: Open contract Word doc → Switch to Edit mode → Make revisions → Save → Ask AVA: "Check if my changes address the concerns we discussed" **Work with multiple documents simultaneously** **Features**: * Open multiple files in separate tabs * Switch between files easily * Each file independent (can edit one, view another) * Tab management (close, reorder) **Use Cases**: * Compare documents side-by-side (mentally) * Reference one doc while editing another * Review multiple contracts * Cross-check data across spreadsheets **Example**: Open 3 tabs: * Tab 1: Contract template (view) * Tab 2: New contract draft (edit) * Tab 3: Vendor's proposal (view) Work across all three to draft contract ### File Viewer Benefits **Problem**: Constantly switching between AVA and SharePoint breaks flow **Solution**: Keep reference documents open in File Viewer while chatting **Example**: * Chat: "Review this contract for payment terms" * Viewer: Contract open for reference * Chat: "The payment schedule on page 5 seems unusual..." * No need to leave AVA **Problem**: Need to make quick edits but don't want to lose chat context **Solution**: Edit directly in File Viewer, changes save automatically **Example**: * Discussing document with AVA * AVA suggests improvements * Switch to Edit mode * Make changes * Continue discussion about revised version **Benefit**: Files open directly from SharePoint/OneDrive **Features**: * Always viewing latest version * Changes sync immediately * Respects SharePoint permissions * Maintains version history * Supports co-authoring **Security**: Uses your delegated permissions—can only open files you have access to **Benefit**: Open files remain available across sessions **Features**: * Files stay open when you close AVA * Return later and files are still there * Quick access to frequently referenced docs * Can close individual tabs when done **Note**: Files aren't downloaded—always pulling from SharePoint ## Workspace Tips & Tricks Before asking complex questions: 1. Add relevant files via Search tab 2. Add meeting transcripts if applicable 3. Upload any external documents 4. Then ask your question with full context **AVA has better context** = **Better answers** Use Notes tab throughout your workflow: * Capture ideas during brainstorming * Copy useful responses from chat * Structure thoughts before asking complex questions * Build documents iteratively * Create presentations from collected insights When AVA references a document: * Open it in File Viewer * Verify AVA's interpretation * Check specific sections * Make edits if needed * Continue informed conversation **Example workflow**: 1. **Search**: Add last month's sales report 2. **File Viewer**: Open current month's data 3. **Chat**: "Compare these two reports" 4. **Notes**: Capture key findings 5. **Notes**: Export summary to PowerPoint **Using all workspace features together** = **Maximum productivity** Speed up your workflow: **Navigation**: * `Ctrl+1`: Switch to Search tab * `Ctrl+2`: Switch to Notes tab * `Ctrl+3`: Switch to File Viewer tab **Notes**: * `Ctrl+B`: Bold * `Ctrl+I`: Italic * `Ctrl+K`: Insert link **General**: * `Ctrl+F`: Find in document (File Viewer) * `Ctrl+S`: Save (Notes/File Editor) ## Common Workflows ### Research & Analysis **Search Tab**: Find all relevant documents * Market research reports * Competitor analyses * Internal strategy docs **File Viewer**: Open key documents for reference * Keep 2-3 most important docs open in tabs * Can quickly check specific information **Chat**: Ask analytical questions * "Compare these companies' approaches" * "What trends do you see across reports?" * "Identify gaps in our strategy" **Notes**: Document findings * Copy key insights from chat * Structure into clear sections * Add your own observations **Notes → Export**: Generate presentation or report * Export notes to PowerPoint * Or copy to Word for detailed report ### Document Creation **Search Tab**: Locate template or past examples * Search for similar past documents * Find company templates **File Viewer**: Keep template/example open * View in one workspace tab * Reference while creating new content **Notes**: Create new document * Start with outline * Use AVA to draft sections * Build content iteratively **Chat + Notes**: Improve content * Ask AVA for feedback * Request revisions * Copy improved versions to notes **Export**: Generate final document * Export to Word or PowerPoint * Make final edits in Office * Save to SharePoint ### Meeting Preparation **Search Tab → Meetings**: Locate related past meetings * Previous meetings with same attendees * Meetings on same topic **Search Tab → Microsoft Files**: Find relevant materials * Project documents * Data and reports * Previous presentations **File Viewer + Chat**: Review and analyze * Open key files * Ask AVA to summarize * Identify key points **Notes**: Prepare for meeting * Draft agenda * Create talking points * Prepare questions * Export to PowerPoint if presenting ## Workspace FAQs **Q: How many files can I have open?** A: Up to 20 files in File Viewer tabs simultaneously **Q: Are uploaded files stored permanently?** A: Files uploaded via "My Computer" are deleted after 90 days of inactivity **Q: How much can I write in Notes?** A: No practical limit; notes support very long documents **Q: Do notes sync across devices?** A: Yes, notes are stored in your AVA account and available on any device **Q: Can others see my workspace?** A: No, your workspace is private unless you explicitly share **Q: Can I open any SharePoint file?** A: Only files you have permission to access in SharePoint **Q: Can I edit shared files?** A: Yes, if you have edit permissions in SharePoint **Q: What happens to files I upload?** A: Stored in your tenant's Azure Blob Storage with your security context **Q: Can I print from File Viewer?** A: Yes, use browser print function (Ctrl+P) **Q: Does Notes support images?** A: Currently text only; export to Word/PowerPoint for images **Q: Can I share my notes?** A: Yes, via Share button or by exporting to OneDrive **Q: Do File Viewer edits sync to SharePoint immediately?** A: Yes, Office Online auto-saves every few seconds ## Next Steps Get started and explore workspace features Advanced techniques and workflows Maximize productivity with Notes Organize and work with files effectively # GitHub Integration Source: https://docs.datarm.ai/integrations/github Connecting AVA to GitHub repositories
GitHub
# GitHub Integration **Coming Soon**: This section is under development. Check back soon for GitHub integration instructions. ## What You'll Learn * Creating a GitHub App for AVA * Configuring repository access * Setting up delegated permissions * Accessing Repositories, Pull Requests, Issues, Code * Organization vs personal account setup * Troubleshooting GitHub connectivity ## Use Cases * Query pull request status * Search code across repositories * Review PR comments and discussions * Track issue status ## In the Meantime Learn how development teams use AVA with GitHub in our [Use Cases](/essentials/use-cases#development-teams) documentation. ## Related Topics * [Admin Guide - MCP Servers](/admin-guide/mcp-servers) - Integration configuration * [Development Workflows](/essentials/use-cases#development-teams) - Code review examples *** *Last Updated: January 2025* # Jira Integration Source: https://docs.datarm.ai/integrations/jira Connecting AVA to Jira project management
Jira
# Jira Integration **Coming Soon**: This section is under development. Check back soon for Jira integration setup instructions. ## What You'll Learn * Creating a Jira OAuth application * Configuring AVA MCP server for Jira * Accessing Issues, Projects, Boards, Sprints * User permission mapping * Troubleshooting Jira connectivity ## Use Cases * Query sprint status and backlog * Create and update issues via AVA * Generate sprint reports * Search across Jira projects ## In the Meantime See how development teams use AVA with Jira in our [Use Cases](/essentials/use-cases#development-teams) documentation. ## Related Topics * [Admin Guide - MCP Servers](/admin-guide/mcp-servers) - Integration setup * [Development Use Cases](/essentials/use-cases#development-teams) - Sprint planning examples *** *Last Updated: January 2025* # Microsoft 365 Integration Source: https://docs.datarm.ai/integrations/microsoft-365 Connecting AVA to Calendar, Mail, Files, and Teams
Microsoft 365
# Microsoft 365 Integration **Coming Soon**: This section is under development. Check back soon for detailed Microsoft 365 setup instructions. ## What You'll Learn * Automatic Microsoft 365 integration via Entra ID * Configuring delegated permissions for Graph API * Accessing Calendar, Mail, Files, OneDrive, Teams, To Do * Troubleshooting Microsoft 365 connectivity * Permission scopes and user consent ## Available Services * **Calendar**: Schedule queries, meeting summaries * **Mail**: Email search, drafting, summarization * **Files**: OneDrive and SharePoint document access * **Teams**: Meeting transcripts, channel messages * **To Do**: Task lists and reminders ## In the Meantime Microsoft 365 integration is automatically configured during AVA deployment through Microsoft Entra ID. See our [Deployment Architecture](/deployment/architecture#mcp-server-layer) for details. ## Related Topics * [Entra ID Setup](/admin-guide/entra-id-setup) - Configuring delegated permissions * [Security Model](/deployment/security#microsoft-entra-id-integration) - Understanding delegated access *** *Last Updated: January 2025* # NetSuite Integration Source: https://docs.datarm.ai/integrations/netsuite Connecting AVA to NetSuite ERP
NetSuite
# NetSuite Integration **Coming Soon**: This section is under development. Check back soon for NetSuite integration setup instructions. ## What You'll Learn * NetSuite OAuth configuration for AVA * Setting up role-based access * Accessing business records and transactions * Custom record types * Troubleshooting NetSuite connectivity ## Use Cases * Query customer and vendor records * Access transaction history * Generate financial reports * Search across NetSuite modules ## In the Meantime Contact [avasupport@datarm.ai](mailto:avasupport@datarm.ai) for NetSuite integration assistance. ## Related Topics * [Admin Guide - MCP Servers](/admin-guide/mcp-servers) - Setting up integrations * [Deployment Architecture](/deployment/architecture#mcp-server-layer) - MCP technical details *** *Last Updated: January 2025* # Integrations Overview Source: https://docs.datarm.ai/integrations/overview Connecting AVA to your enterprise applications # Integrations Overview AVA uses the Model Context Protocol (MCP) to connect with your enterprise applications. Each integration runs as a dedicated Container App in your Azure tenant, using your existing SSO configuration. **Note**: Some integration guides are still under development. Check back soon for detailed setup instructions. ## Available Integrations Connected Data panel showing GitHub, Jira, Salesforce, Microsoft 365, and other integrations
Microsoft 365
Microsoft 365
Salesforce
Salesforce
Jira
Jira
GitHub
GitHub
NetSuite
NetSuite
Common connection issues and solutions Need another integration? Email [avasupport@datarm.ai](mailto:avasupport@datarm.ai) ## How MCP Integrations Work Each integration runs as a Container App in your Azure tenant Uses your existing SSO setup for each application (Salesforce OAuth, Jira OAuth, etc.) Operates with user permissions—no elevated access Users access data via simple @ mentions in conversations ## Beyond Standard Integrations **Any Enterprise Application, One AI Assistant**: AVA can connect to virtually any application already configured in your Microsoft Entra ID with OAuth or SAML SSO. ### Leverage Your Existing SSO Infrastructure Your organization has likely already set up dozens of enterprise applications in Microsoft Entra ID with single sign-on. AVA can extend to these applications through MCP, turning your existing SSO ecosystem into a connected AI platform. **Examples of Additional Integrations:** Workday, BambooHR, ADP, SAP SuccessFactors Monday.com, Asana, ClickUp, Smartsheet Zendesk, Freshdesk, ServiceNow, Intercom HubSpot, Marketo, Mailchimp, Pardot QuickBooks, Xero, Bill.com, Concur Your internal tools with OAuth/SAML **Simple Requirements:** * ✅ Registered as Enterprise Application in Entra ID * ✅ OAuth 2.0 or SAML 2.0 authentication * ✅ SSO already working for your users * ✅ API access available As you add Enterprise Applications to your Entra ID, AVA's integration possibilities grow—creating a unified AI assistant across your entire technology stack. [Learn more about MCP extensibility →](/essentials/how-it-works#extend-ava-to-any-enterprise-application) ## Technical Details Learn more about MCP server architecture in our [Deployment Architecture](/deployment/architecture#mcp-server-layer) documentation. *** *Last Updated: January 2025* # Salesforce Integration Source: https://docs.datarm.ai/integrations/salesforce Connecting AVA to Salesforce CRM
Salesforce
# Salesforce Integration **Coming Soon**: This section is under development. Check back soon for step-by-step Salesforce integration instructions. ## What You'll Learn * Creating a Salesforce Connected App * Configuring OAuth 2.0 for AVA * Setting up delegated user permissions * Accessing Opportunities, Leads, Cases, Accounts, Contacts * Custom object access * Troubleshooting Salesforce connectivity ## Use Cases * Query pipeline and opportunities in conversations * Update CRM records via AVA Tasks * Generate reports from Salesforce data * Search across Salesforce objects ## In the Meantime Review the [Use Cases](/essentials/use-cases#sales-teams) documentation to see how sales teams use AVA with Salesforce CRM. ## Related Topics * [Admin Guide - MCP Servers](/admin-guide/mcp-servers) - Setting up integrations * [Sales Use Cases](/essentials/use-cases#sales-teams) - Real-world examples *** *Last Updated: January 2025* # Troubleshooting Integrations Source: https://docs.datarm.ai/integrations/troubleshooting Common integration issues and solutions # Troubleshooting Integrations **Coming Soon**: This section is under development. Check back soon for comprehensive troubleshooting guidance. ## What You'll Find * Connection timeout issues * Authentication failures * Permission errors * @ mention not working * MCP server health checks * Azure Container Apps logs * Common error codes and solutions ## In the Meantime For immediate assistance with integration issues: Email: [avasupport@datarm.ai](mailto:avasupport@datarm.ai) Review MCP server architecture ## Related Topics * [Admin Monitoring](/admin-guide/monitoring) - Checking MCP server health * [Security Model](/deployment/security) - Understanding delegated permissions * [FAQ](/support/faq) - Common questions *** *Last Updated: January 2025* # Welcome to AVA by DataRM Source: https://docs.datarm.ai/introduction Enterprise AI assistant that integrates seamlessly with your Microsoft 365 environment and business systems ## What is AVA? ### AI That Works™ **AVA (AI Virtual Assistant)** is an enterprise-grade AI assistant that runs securely within your Azure tenant. Unlike traditional SaaS AI tools that just answer questions, AVA *takes action*—scheduling your meetings, sharing your documents, creating tasks, and generating presentations. It integrates directly with your Microsoft 365 environment and business systems through Microsoft Graph API, using your existing identity and permissions to be your AI super assistant. **AI That Works, Not Just Talks**: AVA doesn't just chat—it completes real work in your Microsoft 365 environment. Schedule meetings, create tasks, share files, generate presentations, and more. Conversational AI with access to your calendar, email, files, Salesforce, Jira, and more Multi-step workflow automation with intelligent task orchestration RAG-powered search across your SharePoint documents and knowledge bases ## Core Capabilities Available Everywhere These powerful features work across all of AVA—Chat, Tasks, and Knowledge Search—making every interaction more productive. **Real-time web searches** for current news, research, and information beyond your data **Add images** to provide visual context or **generate images** with DALL-E 3 and GPT-image-1 **Schedule meetings**, create tasks, share documents—AVA takes action via Microsoft Graph API **Copy any response** to your Workspace Notes to build documents and organize insights **Share ongoing conversations** with teammates to collaborate and speed up work **Full presentation creation** from your conversations—Version 3 coming soon with enhanced automation **Export to Word, Excel, PowerPoint** from any conversation, task, or search result **The possibilities are nearly endless**: AVA leverages Microsoft Graph API to act as your AI super assistant, completing real work across your Microsoft 365 environment. ## Core Capabilities ### Integrated Workspace AVA features a powerful three-panel interface: AVA main interface showing session history, chat area, and file viewer workspace Access all your past conversations, tasks, and knowledge searches organized by recency Central interaction space where you converse with AVA and use @ mentions to pull data from connected sources Tabbed interface for Search, Notes, and File Viewer—similar to artifacts in other AI tools but integrated with your Microsoft 365 environment ### Connected Data Sources Pull information from multiple systems using simple @ mentions: Mail, Calendar, Files, OneDrive, Teams, To Do—all via @Microsoft 365 Apps Real-time web searches with @Web for current information and research Opportunities, Leads, Cases, Accounts, Contacts Issues, Projects, Boards, Sprints Repositories, Code, Pull Requests, Issues Business records and transactions Extend to any app with OAuth/SAML SSO in your Entra ID **Beyond Standard Integrations**: AVA can connect to virtually any Enterprise Application already configured in your Microsoft Entra ID with OAuth or SAML. Your existing SSO infrastructure becomes AVA's integration platform. [Learn more →](/integrations/overview#beyond-standard-integrations) ### Powerful Productivity Features Generate Word documents, PowerPoint presentations, and Excel spreadsheets directly from your conversations or notes Access and summarize recorded Microsoft Teams meetings, add them to conversations for context Create, save, and share custom prompts with your team. Filter by connected data sources and file types Choose from multiple AI models across Azure OpenAI, AWS Bedrock, and Google Vertex AI ## Explore the Documentation Find the information you need—whether you're an end user, administrator, or decision maker. Enterprise security, seamless integration, and proven results Learn how to use AVA Chat, Tasks, and Knowledge Search effectively Deploy and manage AVA in your Azure tenant Connect Microsoft 365, Salesforce, Jira, GitHub, and more Rollout strategies and training resources for your organization Videos, templates, and learning materials ## Quick Links Coming soon! Step-by-step video guides Ready-to-use prompts by role and use case Productivity shortcuts for power users Common issues and solutions ## Get Started Authenticate with your Microsoft work account Try [AVA Chat](/features/ava-chat), [Tasks](/features/ava-tasks), and [Knowledge Search](/features/knowledge-search) Use @ mentions to access your [connected data sources](/integrations/microsoft-365) Build custom prompts in the [Prompt Library](/guides/prompt-library) and share with your team ## Popular Features Pull data from any connected source with simple @ syntax Create corpora from your SharePoint sites for intelligent document search Master the Notes, Search, and File Viewer capabilities Generate PowerPoint, Word, and Excel files from your work ## Need Help? Email us at [avasupport@datarm.ai](mailto:avasupport@datarm.ai) - we typically respond within 4 business hours For new customers and product demonstrations # Quickstart Guide Source: https://docs.datarm.ai/quickstart Get started with AVA in under 5 minutes ## Welcome to AVA! This quickstart guide will walk you through your first AVA experience—from signing in to having your first conversation with connected data. You'll be productive in minutes. **Prerequisites:** You need a Microsoft work account with access to AVA. If you don't have access yet, [contact your IT administrator](mailto:avasupport@datarm.ai) or [get in touch with us](https://datarm.ai/contact). ## Step 1: Sign In AVA uses your Microsoft work account (same credentials you use for Office 365): 1. Navigate to your AVA URL (provided by your administrator) 2. Click **"Sign in with Microsoft"** 3. Enter your work email and password 4. Complete multi-factor authentication if prompted 5. Grant AVA delegated permissions when asked Once signed in, you'll see AVA's three-panel interface: Session History (left), Chat Area (center), and Workspace (right). **First time?** AVA may show you a quick tour of the interface. We recommend taking it! ## Step 2: Your First Chat Let's start with a simple conversation to get familiar with AVA. Remember: **AVA is AI that works**—it can take action, not just answer questions! ### Try These Examples **Type in chat**: ``` @Microsoft 365 Apps what meetings do I have today? ``` **What AVA does**: * Connects to your Microsoft Calendar * Retrieves your meetings * Displays them in an organized list **You'll see**: All your meetings with times, attendees, and locations **Try follow-up**: "Send me details on the 2pm meeting" **Type in chat**: ``` @Microsoft 365 Apps show me unread emails from this week ``` **What AVA does**: * Queries your Outlook inbox * Filters to unread emails from last 7 days * Summarizes the messages **You'll see**: List of unread emails with senders, subjects, and snippets **Try follow-up**: "Summarize the most important ones" **Type in chat**: ``` @Microsoft 365 Apps find my Q4 budget spreadsheet ``` **What AVA does**: * Searches your OneDrive and SharePoint * Finds files matching "Q4 budget" * Returns relevant files **You'll see**: Matching files with locations and modified dates **Try follow-up**: "Open it in the workspace" **Type in chat**: ``` @Web what are the latest news on AI developments? ``` **What AVA does**: * Searches the web in real-time * Finds current information * Summarizes results with sources **You'll see**: Up-to-date information with source links **Try follow-up**: "Find recent articles about GPT-5" **If your organization has Salesforce connected**: ``` @Salesforce show me my open opportunities ``` **What AVA does**: * Queries Salesforce with your permissions * Retrieves opportunities assigned to you * Formats the results **You'll see**: Your opportunities with stages, values, and close dates **Try follow-up**: "Which ones are closing this month?" **AVA doesn't just answer—it acts!** **Schedule a meeting**: ``` Schedule a 30-minute meeting with [colleague name] tomorrow at 2pm to discuss Q4 planning ``` **What AVA does**: * Checks availability * Creates calendar event * Sends invitation * Sets up Teams link **Create a task**: ``` Create a task to review the budget by Friday ``` **What AVA does**: * Creates task in Microsoft To Do * Sets due date * Adds it to your list **Upload and analyze an image**: ``` [Upload a screenshot] What's in this image? ``` **What AVA does**: * Analyzes the image * Describes contents * Answers questions about it ### Understanding @ Mentions The `@` symbol is how you access connected data sources: * **@Microsoft 365 Apps** - Calendar, Mail, Files, OneDrive, SharePoint, Teams messages and meetings * **@Web** - Real-time web searches for current information * **@Salesforce** - CRM data (if configured) * **@Jira** - Issues and projects (if configured) * **@GitHub** - Repositories and code (if configured) * **@NetSuite** - Business records (if configured) Start typing `@` and AVA will show you all available data sources! ## Step 3: Try the Workspace The Workspace panel on the right has three tabs. Let's explore them: ### Search Tab 1. Click **"Search"** tab in workspace 2. Select **"Microsoft Files"** 3. Search for a document (try "meeting notes") 4. Click a file to preview it 5. Click **"Add to Conversation"** button Now ask AVA: **"Summarize the key points from this document"** AVA reads the document and provides a summary! ### Notes Tab 1. Click **"Notes"** tab 2. Type some notes: ``` # My First AVA Session Today I learned: - How to use @ mentions - How to search for files - How to add files to conversations ``` 3. Use the formatting toolbar to style your notes 4. Notes are automatically saved You can export notes to PowerPoint! Try adding more content and clicking **"Export to PowerPoint"** ### File Viewer Tab 1. Ask AVA: **"@Microsoft 365 Apps find the employee handbook"** 2. When AVA returns results, click a file 3. File opens in **File Viewer** tab 4. Toggle between **View** and **Edit** modes 5. Make changes if needed (saves automatically to SharePoint) ## Step 4: Create Your First Task Tasks are perfect for multi-step workflows. Let's create a simple one: Click **"Tasks"** in the main navigation Click **"New Task"** button Enter: "Meeting Summary and Follow-Up" Create three steps: **Step 1: Get Meeting Info** ``` @Microsoft 365 Apps show details of my last team meeting ``` **Step 2: Summarize** ``` Create a bullet-point summary of key discussion points and decisions from the meeting ``` **Step 3: Generate Follow-Up** ``` Draft a follow-up email to send to attendees with: - Summary of key decisions - Action items - Next meeting date ``` Click **"Run Task"** and watch AVA execute each step sequentially! After the task completes, you'll have a meeting summary and a drafted follow-up email! Save successful tasks as templates by clicking **"Save as Template"** after completion ## Step 5: Try Knowledge Search If your organization has set up Knowledge Search corpora, try this: Click **"Knowledge Search"** in main navigation Click **"@ Connected Data"** and choose a corpus (e.g., "HR Policies") Try: "What's the remote work policy?" Or: "How much PTO do employees get?" AVA searches documents and returns: * AI-generated answer * Source document citations * Relevant excerpts If you don't see any corpora available, Knowledge Search hasn't been set up yet. Ask your administrator to create corpora from your SharePoint document libraries. ## Step 6: Customize AVA Make AVA work better for you with custom instructions: Click your profile icon (top right) → **"Settings"** Click **"Custom Instructions"** tab Example: ``` I'm a sales manager in the enterprise segment. When showing opportunities: - Always include ARR (Annual Recurring Revenue) - Highlight deals over $100K - Show sales cycle length My working hours: 8am-6pm Pacific Time Format currency in USD with commas ``` Click **"Save"** - AVA will remember these preferences! ## Common Questions **Only what you can access.** AVA uses your Microsoft credentials and operates with your permissions. If you can't see a file in SharePoint, AVA can't either. **Yes!** AVA runs entirely within your organization's Azure tenant. Your data never leaves your environment. All processing happens in your tenant. **No**, unless you explicitly share them. Your chats are private by default. Data source connections (Salesforce, Jira, GitHub, etc.) are configured by your administrator. Contact your IT team to request additional integrations. Try: * Being more specific * Using @ mentions explicitly * Breaking complex requests into simpler questions * Checking that you have access to the data source AVA is a web application that works in any modern browser, including mobile browsers. There's no dedicated mobile app yet. ## Tips for Success **AI That Works™**: Remember, AVA doesn't just answer questions—it takes action! Schedule meetings, create tasks, share files, generate images, and export to Office formats. The possibilities are nearly endless with Microsoft Graph API. Explicitly reference data sources with @ for faster, more accurate results AVA remembers context—refine answers with follow-up questions Schedule meetings, create tasks, share documents—let AVA do the work for you Create reusable prompts in the Prompt Library for frequent tasks Try Chat, Tasks, and Knowledge Search to find what works best for your workflow Upload documents and images to add context to your conversations Copy AVA responses to Notes to build comprehensive documents Share your conversations with teammates for faster collaboration ## Example Workflows to Try Now that you're familiar with basics, try these real-world scenarios: 1. **@Microsoft 365 Apps** find tomorrow's client meeting 2. **@Salesforce** get account history 3. **@Microsoft 365 Apps** find recent email threads 4. **@Web** search for latest company news 5. Ask AVA to create a meeting prep document 6. Export to Notes → PowerPoint 1. **@Salesforce** show opportunities closing this week 2. **@Jira** show completed issues 3. **@GitHub** show merged PRs 4. Ask AVA to create status report 5. Export to Word document 1. Go to Knowledge Search 2. Select relevant corpus 3. Ask: "Find all mentions of \[topic] in contracts" 4. Review results with source citations 5. Ask follow-up questions to refine 1. Open Notes tab 2. Create outline 3. Ask AVA to draft each section 4. Copy good content from chat to notes 5. Iterate until satisfied 6. Export to desired format ## Next Steps Learn about all AVA capabilities Master conversational AI See how your team can use AVA Understand the technology ## Get Help **[avasupport@datarm.ai](mailto:avasupport@datarm.ai)** Our team typically responds within 4 business hours **[Contact Sales](https://datarm.ai/contact)** For new customers and product demonstrations *** **Congratulations!** 🎉 You've completed the AVA quickstart. You now know how to: * Have conversations with connected data using @ mentions * Use the Workspace (Search, Notes, File Viewer) * Create multi-step Tasks * Search documents with Knowledge Search * Customize AVA with custom instructions Start exploring and make AVA work for you! # Glossary Source: https://docs.datarm.ai/resources/glossary AVA terminology and concepts explained # Glossary Quick reference for AVA terminology, concepts, and features. ## A ### @ Mentions Syntax for accessing connected data sources in AVA. Type `@` followed by the source name (e.g., `@Microsoft 365 Apps`, `@Web`, `@Salesforce`) to pull data from that system into your conversation. ### @Microsoft 365 Apps Unified @ mention for accessing all Microsoft 365 services including Calendar, Mail, Files, OneDrive, SharePoint, Teams messages, and meetings. Replaces individual mentions like @Calendar, @Mail, @Files, etc. ### @Web Real-time web search capability allowing AVA to find current information, news, and data beyond its training or your organizational data. Perfect for research, market intelligence, and up-to-date information. ### AI Super Assistant AVA's positioning as more than a chatbot—an AI assistant that takes action across your Microsoft 365 environment through Microsoft Graph API integration. ### AI That Works™ AVA's core philosophy and tagline, emphasizing that AVA takes real action (scheduling meetings, creating tasks, sharing files) rather than just answering questions. ### AVA Chat Conversational AI interface with @ mention access to connected data sources. Main feature for interacting with AVA and your business systems. ### AVA Tasks Multi-step workflow automation feature allowing sequential execution of complex processes with intelligent orchestration. ### Azure AI Foundry Microsoft's AI deployment platform, primary provider for AVA's AI models (GPT-5, DALL-E 3, etc.). ### Azure Container Apps Serverless container service hosting AVA's backend Python services and MCP servers. Provides auto-scaling and high availability. ## C ### Conversation Sharing Capability to share ongoing AVA conversations with teammates, including full context and history. Supports view-only and collaborative access levels. ### Copy to Notes One-click feature to copy any AVA response to Workspace Notes for document building and organization. ### Corpus (plural: Corpora) Collection of documents indexed for Knowledge Search. Created from SharePoint sites or document libraries for semantic search. ### Core Capabilities Features available across all of AVA (Chat, Tasks, Knowledge Search): images, Microsoft actions, copy to notes, conversation sharing, exports, and PowerPoint generation. ### Custom Instructions Personalized preferences and behavior settings that AVA remembers across all conversations and tasks. ## D ### DALL-E 3 OpenAI's image generation model available in AVA for creating visuals from text descriptions. Offers natural and vivid style options. ### Delegated Permissions Security model where AVA operates with the authenticated user's permissions only—no elevated access or service accounts. ## E ### Embeddings Vector representations of text used for semantic search in Knowledge Search. Generated by AI models and stored in pgVector. ### Entra ID Microsoft's identity and access management platform (formerly Azure AD). Provides authentication and authorization for AVA. ### Export Capabilities Feature set allowing generation of Office documents (PowerPoint, Word, Excel) from AVA conversations, tasks, and notes. ## G ### GPT-5 Series Latest OpenAI models available in Azure including GPT-5, GPT-5 Pro, GPT-5 Codex, GPT-5 Chat, and GPT-5 Mini/Nano variants. ### GPT-image-1 Latest image generation model from OpenAI with improved quality and detail compared to DALL-E 3. ## K ### Knowledge Search RAG-powered semantic search feature for querying SharePoint documents across multiple corpora simultaneously. ## M ### MCP (Model Context Protocol) Standardized integration layer connecting AVA to enterprise applications like Microsoft 365, Salesforce, Jira, and GitHub. ### MCP Servers Individual container apps running integration logic for each connected data source. Deployed separately for independent scaling. ### Microsoft Graph API Unified API for Microsoft 365 services. Enables AVA to take actions like scheduling meetings, creating tasks, sharing files, and sending emails. ### Microsoft To Do Task management application integrated with AVA. Tasks created by AVA appear in Microsoft To Do with due dates and assignments. ## P ### pgVector PostgreSQL extension for efficient vector storage and similarity search. Used for Knowledge Search embeddings. ### PowerPoint Generation Capability to create presentations from AVA conversations or notes. Version 3 coming with enhanced automation and intelligent layouts. ### Prompt Library Repository of saved, reusable prompts that can be shared with individuals, teams, or the entire organization. ## R ### RAG (Retrieval Augmented Generation) AI technique combining semantic search with language generation. Used in Knowledge Search to provide accurate answers with source citations. ## S ### SuperPrompt Enhanced AI guidance toggle in AVA Tasks that provides more detailed instructions and context for complex workflows. ## W ### Workspace Right-side panel in AVA with three tabs: Search (find files/meetings), Notes (rich text editor), and File Viewer (view/edit Office documents). ## Related Resources Detailed explanation of cross-app features Core concepts and architecture Technical architecture details Comprehensive feature overview *** *Last Updated: January 2025* # Contact Support Source: https://docs.datarm.ai/support/contact Get help from the AVA support team ## Get Help with AVA **Need technical support?** Email us at [avasupport@datarm.ai](mailto:avasupport@datarm.ai) - our team typically responds within 4 business hours. ## Support Channels **[avasupport@datarm.ai](mailto:avasupport@datarm.ai)** Response within 4 business hours **[Contact Sales](https://datarm.ai/contact)** For new customers and demos ## Before Contacting Support Please have the following information ready: Your organization name and AVA instance URL Clear description of the problem What actions led to the issue Any error messages or screenshots Browser version, OS, etc. ## Issue Severity Levels * **P0 - Critical**: System down, data loss, security breach * **P1 - High**: Major functionality broken, significant impact * **P2 - Medium**: Feature not working as expected, moderate impact * **P3 - Low**: Minor issue, cosmetic problem, question ## Additional Resources Common questions and answers Current system status Complete documentation API documentation ## What to Include in Your Support Request To help us assist you quickly, please include: * Organization name and AVA instance URL * Clear description of the issue * Steps to reproduce the problem * Any error messages or screenshots * Browser and OS information (if relevant) ## Looking for Something Else? For new customers and product demonstrations Common questions and answers # Creating Tasks Source: https://docs.datarm.ai/user-guide/creating-tasks Building multi-step workflows with AVA Tasks # Creating Tasks **Coming Soon**: This section is under development. Check back soon for a complete guide to creating and managing tasks in AVA. ## What You'll Learn * Understanding multi-step task execution * Adding documents to task context * Using the SuperPrompt toggle * Saving tasks as templates * Sharing task templates with your team * Real-world task examples by role ## In the Meantime Visit our [AVA Tasks Features](/features/ava-tasks) page for comprehensive documentation on task capabilities and examples. ## Related Topics * [Task Examples](/features/ava-tasks#real-world-examples) - Sales, support, dev workflows * [Task Library](/features/ava-tasks#task-library) - Reusable templates * [Use Cases by Role](/essentials/use-cases) - Department-specific examples *** *Last Updated: January 2025* # First-Time User Guide Source: https://docs.datarm.ai/user-guide/first-time-user Your first session with AVA - everything you need to know # First-Time User Guide **Coming Soon**: This section is under development. Check back soon for a comprehensive guide for first-time AVA users. ## What You'll Learn * How to sign in with your Microsoft work account * Understanding the three-panel AVA interface * Your first @ mention conversation * Exploring connected data sources * Creating your first task * Using the workspace panel ## Get Started Now While we complete this guide, you can start with our [Quickstart Guide](/quickstart) which covers the essentials of getting started with AVA. ## Key Resources * [What is AVA?](/essentials/what-is-ava) - Understanding AVA's capabilities * [AVA Chat](/features/ava-chat) - Conversational AI features * [Quickstart Guide](/quickstart) - Get up and running quickly *** *Last Updated: January 2025* # Keyboard Shortcuts Source: https://docs.datarm.ai/user-guide/keyboard-shortcuts Boost your productivity with keyboard shortcuts # Keyboard Shortcuts **Coming Soon**: This section is under development. Check back soon for a comprehensive list of keyboard shortcuts to speed up your workflow. ## What You'll Learn * Chat navigation shortcuts * Task management shortcuts * Workspace panel shortcuts * Search shortcuts * Cross-platform shortcuts (Windows/Mac) ## In the Meantime Most standard keyboard shortcuts work in AVA: * **Ctrl/Cmd + Enter**: Submit chat message * **Ctrl/Cmd + K**: Focus search bar * **Ctrl/Cmd + /** - Keyboard shortcut reference (coming soon) For detailed feature usage, see: * [AVA Chat](/features/ava-chat) * [AVA Tasks](/features/ava-tasks) * [Workspace](/features/workspace) *** *Last Updated: January 2025* # User Guide Overview Source: https://docs.datarm.ai/user-guide/overview Learn how to use AVA effectively in your daily work # User Guide Overview Welcome to the AVA User Guide! This section is designed to help end users get the most out of AVA's AI-powered capabilities. **Note**: Some pages in this section are still under development. Check back soon for additional training materials and guides. ## What You'll Find Here Getting started with your first AVA session Tips and tricks for conversational AI Building multi-step workflows Finding information across documents Boost your productivity with shortcuts Learn from our quickstart guide ## Quick Links * [AVA Chat Features](/features/ava-chat) - Detailed feature documentation * [AVA Tasks Features](/features/ava-tasks) - Workflow automation guide * [Knowledge Search](/features/knowledge-search) - RAG-powered search * [Workspace Panel](/features/workspace) - Notes, Search, and File Viewer ## Need Help? Get help from our support team Frequently asked questions *** *Last Updated: January 2025* # Searching Knowledge Source: https://docs.datarm.ai/user-guide/searching-knowledge Finding information across your documents with semantic search # Searching Knowledge **Coming Soon**: This section is under development. Check back soon for best practices on using Knowledge Search effectively. ## What You'll Learn * Understanding semantic vs keyword search * Creating effective search queries * Working with multiple corpora * Interpreting search results and relevance scores * Using extracted fields * Tips for better search accuracy ## In the Meantime Explore our [Knowledge Search Features](/features/knowledge-search) documentation for detailed information on RAG-powered search capabilities. ## Related Topics * [How Knowledge Search Works](/features/knowledge-search#how-knowledge-search-works-technical) - RAG pipeline explained * [Creating a Corpus](/features/knowledge-search#creating-a-corpus) - Admin setup * [Search Examples](/features/knowledge-search#real-world-examples) - Use cases *** *Last Updated: January 2025* # Using AVA Chat Source: https://docs.datarm.ai/user-guide/using-ava-chat Tips and best practices for conversational AI # Using AVA Chat **Coming Soon**: This section is under development. Check back soon for advanced tips and tricks for using AVA Chat effectively. ## What You'll Learn * Mastering @ mentions for data access * Writing effective prompts * Using context from previous messages * Leveraging model selection * Creating reusable prompts in the Prompt Library * Sharing conversations with your team ## In the Meantime Explore our comprehensive [AVA Chat Features](/features/ava-chat) documentation for detailed information on all chat capabilities. ## Related Topics * [Prompt Library](/features/ava-chat#prompt-library) - Save and share prompts * [Model Selection](/features/ava-chat#model-selection) - Choose the right AI model * [Connected Data Sources](/essentials/what-is-ava#connected-data-sources) - Using @ mentions *** *Last Updated: January 2025* # Why Choose AVA? Source: https://docs.datarm.ai/why-ava Enterprise AI that works—secure, integrated, and purpose-built for Microsoft 365 environments # Why Choose AVA? ## AI That Works™, Not Just Talks Most AI assistants just answer questions. **AVA takes action.** AVA completes real work in your Microsoft 365 environment—scheduling meetings, creating tasks, sharing files, generating presentations, and orchestrating complex workflows across your business systems. It's the difference between an AI chatbot and an AI super assistant. Deployed in your Azure tenant with identity-based access control No need to create separate agents—AVA intelligently orchestrates all integrations Built specifically for Microsoft 365 enterprise environments AVA understands context and automatically pulls relevant data from connected sources Operates with your permissions—no elevated access required Choose the right model for each task across multiple providers ## The AVA Advantage **Your data never leaves your environment** AVA is deployed as an Enterprise Application within your own Azure tenant. Unlike SaaS AI tools that send your data to external clouds, AVA processes everything within your infrastructure. **Benefits:** * Complete data sovereignty and control * Meet strict compliance requirements (GDPR, HIPAA, SOC2) * No third-party data exposure * Audit trail stays in your tenant * Aligns with existing Azure governance Your IT team controls deployment, updates, and access—not an external vendor. **Delegated permissions—AVA accesses only what you can access** AVA uses Microsoft Entra ID (formerly Azure AD) for authentication. It operates with **your delegated permissions**, not elevated admin privileges. **What this means:** * If you can't access a file in SharePoint, neither can AVA on your behalf * No "shadow RBAC" or separate permission systems to manage * Every action is logged with your identity * Conditional Access policies apply to AVA * MFA and security policies are enforced **Unlike other AI tools:** * Many AI platforms require admin-level API keys that can access *everything* * AVA never has broader access than the user—it's fundamentally more secure **One conversation, all your data sources** Unlike other AI platforms where you must create and configure separate agents for each integration (one for Salesforce, another for Jira, another for email), AVA seamlessly orchestrates across all your connected data sources in a single conversation. **Example:** ``` You: @Salesforce show opportunities closing this quarter AVA: [Returns Salesforce data] You: For each opportunity, check @Jira for blocking issues AVA: [Automatically combines Salesforce + Jira data] You: Email the report to my team AVA: [Generates report and drafts email via Outlook] ``` **One AI assistant, zero configuration required.** Just use @ mentions and AVA handles the complexity. **Native integration with your entire Microsoft ecosystem** AVA leverages Microsoft Graph API for deep integration: **Outlook:** * Read and send emails * Schedule meetings with availability checking * Access calendar events * Manage inbox with AI assistance **Teams:** * Access meeting transcripts * Search channel messages * Create Teams meetings * Summarize conversations **OneDrive & SharePoint:** * Search across all your files * View and edit documents in AVA * Share files with proper permissions * Create folders and organize content **To Do:** * Create and manage tasks * Set due dates and reminders * Organize with lists * Track completion **And more:** * Planner integration * Power BI data access * Microsoft Security (Defender) * All via unified @Microsoft 365 Apps mention ## Enterprise-Grade Technical Foundation AVA is built on modern, proven Azure infrastructure designed for scale and reliability. ### Architecture **Scalable, serverless Python backend** * Auto-scales based on demand * Handles concurrent users seamlessly * Containerized for consistency * Rapid deployment updates **High-performance vector storage for RAG** * Sub-second semantic search * Stores document embeddings * Efficient similarity matching * Scalable to millions of documents **Standardized integration layer** * Connects to any enterprise system * Extensible architecture * Community-driven server ecosystem * Custom connectors supported **Enterprise identity & access** * SAML/OAuth authentication * Delegated permissions model * Conditional Access support * MFA enforcement **Secure secrets management** * API keys encrypted at rest * No hard-coded credentials * Managed identities for services * Audit logging for access **Monitoring & telemetry** * Performance tracking * Error detection * Usage analytics * Custom dashboards ### Security Features **Your data stays secure** * **Encryption in Transit**: TLS 1.3 for all communications * **Encryption at Rest**: Azure-managed encryption for databases and storage * **Secrets Management**: Azure Key Vault for API keys and credentials * **Network Security**: Container Apps in isolated virtual network * **Private Link**: Optional Azure Private Link for VPN-only access **Identity-based security model** * **Microsoft Entra ID**: Enterprise SSO authentication * **Delegated Permissions**: AVA uses user's token for all API calls * **No Privilege Escalation**: Can't access data user can't access * **Conditional Access**: Respects your org's access policies * **MFA Support**: Multi-factor authentication enforced **Meet regulatory requirements** * **Data Residency**: Processes data in your Azure region * **GDPR Compliant**: Data stays in your tenant * **HIPAA Compatible**: Suitable for healthcare deployments * **SOC 2**: Azure infrastructure compliance * **Audit Trails**: Complete logging in Application Insights **You maintain control** * **Tenant Isolation**: Single-tenant deployment * **Admin Controls**: IT manages users, access, and policies * **Azure Governance**: Integrates with existing Azure controls * **Update Control**: You control when to update AVA * **Data Retention**: You define retention policies ## Flexible AI Model Selection Choose the right model for each task—AVA supports multiple AI providers for maximum flexibility. **Primary AI provider** * GPT-5 family (including Pro, Codex, Chat) * GPT-4o for balanced performance * o1 and o3-mini for reasoning * GPT-image-1 and DALL-E 3 for images * Native Azure integration **Alternative provider** * Claude Opus 4.1 (best for coding) * Claude Sonnet 4.5 (1M token context) * Claude Haiku 4.5 (fast, cost-effective) * Latest Anthropic models **Google AI models** * Gemini 2.5 Pro with Deep Think * Gemini 2.5 Flash (best price-performance) * Gemini Pro for general tasks * Flexible deployment options **Why this matters:** * **Best tool for the job**: Use Claude for coding, GPT-5 for reasoning, Gemini for speed * **No vendor lock-in**: Switch providers as new models emerge * **Cost optimization**: Choose economical models for routine tasks * **Future-proof**: New models added as they become available ## Proven Results **Up and running in hours, not weeks** Unlike AI platforms requiring agent configuration, data connectors, and complex setup, AVA deploys quickly: * Azure infrastructure: 2-4 hours * User provisioning: Minutes via Entra ID * Integrations: Zero configuration required **Employees actually use it** AVA's @ mention system and natural language interface mean minimal training: * Intuitive for end users * Works with existing workflows * No specialized knowledge required * Immediate productivity gains **Measurable business impact** Organizations report: * 40% faster research and analysis * 60% reduction in report generation time * 30% improvement in data-driven decisions * Hours saved per employee per week **Admins love the control** IT teams appreciate: * Runs in their Azure tenant * Standard Azure monitoring * Delegated permissions model * No shadow IT risks ## How AVA Compares | Feature | AVA | Generic AI Chatbots | Other Enterprise AI | | ---------------------- | --------------------------- | ------------------------- | ------------------- | | **Deployment** | Your Azure tenant | External SaaS | Usually SaaS | | **Data Location** | Never leaves tenant | External clouds | Varies | | **Permissions** | Delegated (user-scoped) | Often admin-level | Mixed | | **M365 Integration** | Native via Graph API | Limited or via connectors | Varies | | **Agent Setup** | Zero—seamless orchestration | Manual per integration | Often complex | | **AI Models** | Multi-provider choice | Single provider | Usually locked in | | **Real Actions** | Yes (via Graph API) | Usually chat-only | Varies | | **Enterprise Support** | Full Azure infrastructure | SaaS support | Varies | ## Get Started with AVA [Contact our team](https://datarm.ai/contact) to discuss your requirements and use cases We'll work with your IT team to deploy AVA in your Azure tenant User provisioning via Entra ID and optional training for your team Your team begins using AVA immediately—no complex setup required ## Next Steps Learn what AVA can do for your team Understand AVA's technical architecture Try AVA with our quickstart guide Request a demo or talk to our team ## Need More Information? See how organizations use AVA across departments Deep dive into AVA's security architecture Technical details for IT administrators Get in touch: [avasupport@datarm.ai](mailto:avasupport@datarm.ai)