Semantic Kernel: the magical way of integrating LLMs into you
summary
Semantic Kernel is an open-source .NET SDK that turns a three-method console to‑do app into an AI-powered system using Llama 3.2 (4GB) locally and GPT5 Mini in the cloud; the demo in ~40 minutes adds chat history, memory, and plugins calling C# methods so the LLM can query and update a deterministic backing service. Local model gives fluent text but fails at reliable state changes; switching to the cloud model yields correct tool calls and completes list items.
AI-written summary · about
chapters (11)
- 0:00🚀 Build Stuff Conference – Get Your Tickets!
- 0:30Introduction: Why Integrating AI Matters
- 3:00What is Semantic Kernel? (SDK Overview)
- 6:40Why Not Call OpenAI Directly? (Abstraction Benefits)
- 10:20Demo Setup: Simple To-Do App
- 14:10Adding LLM Interaction (Chat + Streaming Responses)
- 18:00Memory & Chat History Explained
- 21:40Integrating C# Logic via Plugins (Tool Calling)
- 26:10Switching Models: Local (Ollama) vs Cloud (GPT)
- 28:30Real Value: Natural Language as UI
- 30:00Challenges: Ecosystem Complexity & Trade-offs
related talks
42mBridge the chasm between your ML and app devs with Semantic Kernel
01h 00mBuilding your own AI Agent using Semantic Kernel
53mUnleash the power of AI: Automate tasks with OpenAI and Semantic Kernel
46mBuilding AI solutions with Semantic Kernel
33mHow to enhance your own AI services using Semantic Kernel
01h 02mDevelop AI agents with Semantic Kernel
45mEverything You Need to Know About Running LLMs Locally
26mSemantic Kernel: Teaching AI to Not Just Repeat, But Complete!