Building next-level conversational AI apps with .NET and Aspire
summary
Creates a .NET chat app from a template that wires Microsoft.Extensions.AI to a Quadrant vector DB via Aspire, ingests PDFs and GitHub markdown into ~340 vector points, computes embeddings and cosine similarity for RAG, and exposes C# functions as callable tools (weather lookup chained latitude→temperature). Runs local and cloud providers (GitHub models, Olama, Azure OpenAI), spins up and consumes an MCP server (custom build-status tool), returns secret number 1234 and correctly reports the latest ASP.NET Core build failed.
AI-written summary · about
chapters (12)
- 0:00Introduction to AI in .NET
- 4:46Building a basic AI app
- 11:08Using the .NET AI abstractions
- 14:06Retrieval Augmented Generation
- 17:30Advanced tool calling
- 20:11Custom data ingestion
- 22:12GitHub data ingestion demo
- 27:14Understanding embeddings
- 31:25Aspire orchestration and telemetry
- 36:46Model Context Protocol
- 43:22Building a custom MCP server
- 49:37Prompt engineering tools
related talks
59mBuild Next-Gen AI Apps with .NET and Azure
01h 13mBuilding Confident Application Systems with .NET Aspire: From Dev to Deployment
44mAI Building Blocks for .NET: Add intelligence to your C# Apps
58mEasily Add GenAI to .NET Apps using Microsoft.Extensions.AI
46mInfusing your .NET Apps with AI: Practical Tools and Techniques
26mAspire for AI Applications
01h 08mCode to cloud with .NET Aspire
20mBuild Intelligent Apps with .NET and Azure