AI Building Blocks for .NET: Add intelligence to your C# Apps

Bruno Capuano gives a practical, opinionated walkthrough of the core building blocks for adding AI features to .NET apps, from the Microsoft.Extensions.AI abstractions to vector data ingestion, MCP tooling with the C# SDK, and wrapping it all up with the Microsoft Agent Framework.

Overview

The video is structured around a set of “building blocks” for creating intelligent applications in .NET, with short demos for each block.

Foundations: Microsoft.Extensions.AI

• Introduces Microsoft.Extensions.AI as the foundation layer for AI-enabled .NET apps.
• Positions it as a starting point for wiring AI capabilities into C# applications.

Intelligence: VectorData + DataIngestion

• Covers using VectorData and DataIngestion as the “intelligence” layer.
• Focuses on working with vector data as part of building intelligent app experiences.

Tools: MCP with the C# SDK

• Introduces MCP (Model Context Protocol) and using it via the C# SDK.
• Demonstrates how MCP fits into the toolchain for intelligent .NET apps.

The whole package: Microsoft Agent Framework

• Presents the Microsoft Agent Framework as the way to bring the pieces together.
• Shows how the framework can be used to package the overall approach into an agent-style application.

Links

• .NET AI docs: https://learn.microsoft.com/en-us/dotnet/ai/
• Gen AI .NET for Beginners: https://aka.ms/genainet

Video chapters

• 00:00 Welcome and Intro
• 00:28 What we're talking about today
• 02:58 Today's building blocks
• 03:40 Foundations: Microsoft.Extensions.AI
• 05:19 Demo
• 10:50 Intelligence: VectorData + DataIngestion
• 13:49 Demo
• 22:55 Tools: MCP with the C# SDK
• 24:39 Demo
• 31:15 The whole package: Microsoft Agent Framework
• 32:15 Demo
• 43:30 In Summary
• 43:39 Wrap