Microsoft Foundry Multi-Agent Workflows Explained

Multi-agent workflows in Microsoft Foundry, explained with a kitchen, a bowl of Maggi and a new-joiner's first day. Learn the 3 patterns every agentic system uses. This clip is from a live session in my Microsoft Foundry series. One agent can't do everything, just like one person can't run a whole kitchen. Here you'll see how multiple AI agents split the work, and how Foundry Workflows lets you wire them together. What you'll learn: - Why real-world agentic systems use multiple agents plus an orchestrator - Sequential agents: when order matters (onboarding: HR → laptop → manager) - Group chat agents: parallel work with no fixed order - Human in the loop: agents validate, a human approves (leave requests, bill approvals) - How local Python code calls an agent hosted in Foundry (az login + endpoint) - Building a workflow in the Foundry portal, and generating the YAML with Claude or ChatGPT Chapters: 0:00 Why One Agent Isn't Enough 0:31 Sequential Agents (Maggi Example) 1:16 Group Chat Agents: No Fixed Order 1:46 Human in the Loop Approvals 2:36 Run a Foundry Agent from Local Python 3:49 Foundry Workflows: 3 Patterns 4:25 Build a Sequential Workflow 5:16 Create Agents via VS Code & YAML Resources mentioned: 🛠 Build Your Local AI (course, run an LLM locally and call it): https://topmate.io/gauravkhurana/2073657 Watch next: ▶ Microsoft Foundry - AI Platform (full playlist): https://www.youtube.com/playlist?list=PLeGiFBPpRC04 Connect: 🎓 Courses: https://gauravkhurana.com/courses 🌐 Website: https://gauravkhurana.com 📅 1:1 mentoring: https://topmate.io/gauravkhurana