Deep Dive into Microsoft Agent Framework for AutoGen Users

From the creator

On this video we are going to deep dive into Microsoft Agent Framework for AutoGen users. Eric will show a practical migration path. It starts by covering what stays the same and what changes at a glance. Then, it covers model client setup, single‑agent features, MCP Support, Agent-as-a-tool, finally multi‑agent orchestration with concrete code side‑by‑side and so on. Microsoft Agent Framework, a new open-source project from Microsoft that converges Semantic Kernel and AutoGen into a single pro-code SDK for building multi-agent systems. AutoGen is a framework for building AI agents and multi-agent systems using large language models (LLMs). It started as a research project at Microsoft Research and pioneered several concepts in multi-agent orchestration, such as GroupChat and event-driven agent runtime. The project has been a fruitful collaboration of the open-source community and many important features came from external contributors. ✅ Chapters: 00:00 Introduction 02:10 What's Microsoft Agent Framework Recap 03:50 Model Clients 05:16 Single Agent 10:47 Single Agent (Tool Use) 11:14 Single Agent (Streaming) 13:01 MCP Support 13:30 Agent-as-a-Tool 18:17 Key Difference - Conversation State 23:05 Thread & Memory 24:41 Key Difference - Message Types 25:59 User Approval Content type 30:09 Key Difference - Middleware (NEW) 34:38 Multi-Agent Orchestration Patterns 36:18 Key Difference - Programming Model 43:48 Built-in Orchestrations - Sequential 47:42 Built-in Orchestrations

Published 2025-10-30 · Watch on YouTube ↗
Availability and playback restrictions are controlled by the original channel.