Reach 320M+ users in flow of work building your own copilot for Teams
summary
Build a custom Copilot that runs inside Teams using Teams Toolkit and the AI Library (now GA for .NET, JavaScript and Python), deploy a model from Azure OpenAI Studio, and ground it with RAG via Azure AI Search; the planner schema maps LLM-detected intents to implemented actions that return adaptive cards, charts or say/do plans, with inline citations, thumbs up/down feedback and streaming responses. Copilot handoffs link Copilot extensions to full apps; partners like Meltwater, Hexagon and ArcGIS already ship these integrated experiences.
AI-written summary · about
chapters (10)
- 0:00Building custom Copilots in Teams
- 1:39Northwind Traders Copilot demo
- 4:28Copilot components and tech stack
- 7:56Teams Toolkit setup
- 11:28Teams AI Library architecture
- 26:22Business logic and adaptive cards
- 30:12New Copilot UX features
- 34:25Extending Copilots with plugins
- 37:15Partner Copilot examples
- 42:34Recap and resources
related talks
43mHarness the power of AI: Extend Copilot and beyond
23mKEYNOTE: From Copilot to Flight Captain
49mBuild your custom Copilot with Microsoft Copilot Studio
45mThe era of the AI Copilot
45mExploring Copilot Capabilities
14mon Collaboration, Copilot & the Future of Work
15mAI Studio - Creating and scaling your custom copilots
26mThe Future of SharePoint, Teams & Copilot