Observe and control agents across any framework with open source tools | BRK250
Ai agents Agent governance Ai safety Microsoft agent framework Adversarial testing Enterprise ai Responsible ai Open source ai tools Agent evaluation Human In The Loop Ai reliability Msbuild 2026
This session teaches developers how to implement comprehensive AI agent governance across Microsoft Agent Framework and open-source stacks. It covers turning requirements into evaluations, stress-testing against adversarial risks, defining agent risks in YAML format, and implementing open controls that work across frameworks. Designed for enterprise developers building production AI agents who need safety, reliability, and oversight mechanisms.
Related videos
Architecture for running agents in production
What's new in Google Cloud's agent platform
Build context-aware agents: From data to decisions | BRK240
Build work-ready agents with Foundry + Work IQ, govern with Agent 365 | DEM340
From prototype to production build and run agents at scale | BRK241
Enable agents for enterprises using Agent 365 SDK | OD840
Beyond the hype: Orchestrating end-to-end developer workflows with agents
Overview of AI Evaluation (The Context Window #05)
Making AI Reliable in the SDLC: The Role of Deterministic Context at Scale
The Trust Lifecycle for Safe, Secure and Reliable AI Agents
The MCP Megalith: Surviving the 1,000-Tool Explosion and Auth Fragmentation
