Govern AI models, tools, and agents with Azure API Management | OD831
AI workloads don’t behave like traditional APIs. They combine models, tools, and agents across providers, making routing, cost control, safety, and observability harder in production. Learn how to use Azure API Management’s AI Gateway to expose endpoints, route requests, and apply policies. Through demos, go from a single request to a production-ready setup with guardrails and telemetry.
To learn more, please check out these resources:
https://aka.ms/build26-next-steps
𝗦𝗽𝗲𝗮𝗸𝗲𝗿𝘀:
Anish Tallapureddy
Mike Budzynski
𝗦𝗲𝘀𝘀𝗶𝗼𝗻 𝗜𝗻𝗳𝗼𝗿𝗺𝗮𝘁𝗶𝗼𝗻:
This is one of many sessions from the Microsoft Build 2026 event. View even more sessions on-demand and learn about Microsoft Build at https://build.microsoft.com
OD831 | English (US) | Agents & apps
Pre-recorded
#MSBuild
Chapters:
0:00 - Introduction to Governing AI with Azure API Management
00:00:45 - Overview of AI Governance Challenges and Enterprise Needs
00:02:42 - Azure API Management Platform and Capabilities
00:05:02 - Introduction to AI Gateway – A Universal Governance Layer
00:07:02 - Detailed Capabilities: Governance, Security, Observability, Scalability, and Velocity
00:10:20 - Model Governance and Demonstration of Policy Enforcement
00:16:03 - New Features – Unified Model API, Token Metrics, and BYO Model Support
00:17:02 - Tool Governance – Managing MCP Servers, APIs, and Tool Security
00:26:00 - Tool Discovery and Agent Integration Demo using AI Gateway
00:28:20 - Enterprise Catalog and Agent Governance Announcements (A2A GA)