Build your own CoPilot
summary
Shows how to build a custom CoPilot using Azure OpenAI Service and Semantic Kernel with .NET, wired into an Umbraco CMS plugin architecture. Demonstrates function-calling plugins, chat-history memory, audio input and image generation, and fine-tuning in Azure AI Studio. Live demos query CMS entities, navigate to a found homepage, and even created user via a generated API call (it actually created the user), while noting token costs and model choices like GPT‑4o Mini.
AI-written summary · about
chapters (6)
- 0:00Introduction to copilots
- 17:31Building a custom copilot
- 25:59Implementing custom actions
- 33:49Multi-modal AI features
- 38:17Demo: Natural interface
- 47:15Learning resources
related talks
45mThe era of the AI Copilot
36mWhat's new with Copilot Studio
49mBuild your custom Copilot with Microsoft Copilot Studio
48mBuild your own copilot with Copilot Studio
45mBuilding a copilot: Azure AI Studio or Copilot Studio
12mA conversation with Charles Lamanna on building custom copilots with Microsoft
12mThe mechanics of building and extending Microsoft Copilot
45mExploring Copilot Capabilities
