In this episode, Willem Meints explains how you can use Semantic Kernel to leverage AI in the software you build.

00:00:10 - Introduction
00:00:38 - Semantic Kernel introduction
00:01:23 - Integrating Semantic Kernel into your software
00:05:32 - Semantic Kernel architecture
00:09:30 - Use-case demo: HR assistant "Eva"
00:12:41 - "Eva" code walkthrough
00:21:40 - Adding context specific information to the requests (RAG)
00:27:46 - Language awareness
00:29:00 - Context data(-size) restrictions
00:31:44 - Calling multiple tools (reasoning)
00:32:32 - Learning resources
00:35:42 - Wrap up

learning Resources:

Workshop - Building Effective LLM-Based Applications with Semantic Kernel: 
https://github.com/wmeints/effective-llm-applications-workshop

Context Engineering:
https://github.com/wmeints/context-engineering

Book:
https://leanpub.com/effective-llm-applications-with-semantic-kernel/

This video was made possible by Info Support (https://www.infosupport.com).

--

dotnetFlix

A video transcript walkthrough of Microsoft Semantic Kernel, explaining how developers can build LLM-powered applications beyond simple chatbots. Covers the framework's architecture including the kernel object, plugins, tool calling, and filters. Demonstrates a real-world RAG-based HR chatbot (EVA) built with ASP.NET Core, Azure OpenAI, SignalR for streaming, and PostgreSQL. Explains how tool/function calling works, how MCP fits into the picture, context window limitations, and multi-step reasoning. Also touches on using AI coding agents like GitHub Copilot to build such features, with a reminder that human code review remains essential.

How To Use AI In Your Software with Semantic Kernel