Multi-Agent AI Systems
Building Multi-Agent Systems with Azure AI Foundry
Notes and synthesis from BRK197: Build multi-agent systems the right way with Azure AI Foundry.
Session: BRK197 Date: Wednesday, Nov 19, 2025 Time: 10:15 AM - 11:00 AM PST Location: Marriott Marquis, Yerba Buena Ballroom BO2
Coming Soon
This article will be published during/after Microsoft Ignite 2025 (Nov 18-21). Check back soon for full coverage of multi-agent orchestration, the Microsoft Agent Framework, and production deployment patterns.
Session Overview
Topics Covered:
- Microsoft Agent Framework (Semantic Kernel + AutoGen)
- Shared state management
- Human-in-the-loop patterns
- OpenTel observability
- MCP toolchains
- A2A (Agent-to-Agent) protocol
- Activity Protocol
- Deploying as containerized agents on Azure AI Foundry
Why This Matters: Moving from single-agent prototypes to fleet-level orchestration requires proper architecture, observability, and deployment patterns. This session covers the production-ready approach.
Key Questions
Going into this session, I'm looking for answers to:
- Orchestration - How does shared state work across multiple agents?
- Observability - What does proper OpenTel instrumentation look like for multi-agent systems?
- MCP Integration - How do agents share tools and context?
- Deployment - What's the container architecture for agent fleets?
- Human-in-the-loop - Where do humans intervene in agent workflows?
Stay Tuned
Full notes, synthesis, and implementation insights coming soon.
Session: BRK197 | Nov 19, 2025 | Marriott Marquis, Yerba Buena Ballroom BO2