Understand and fix Agent Framework apps with observability and evals | DEM361
Microsoft agent framework Observability Llm evaluation Ai testing Open telemetry Open inference protocol Agentic ai Debugging ai Ai monitoring Agent performance Msbuild 2026
This advanced demo session teaches developers how to instrument Microsoft Agent Framework applications for observability using open standards (OpenTelemetry, Open Inference Protocol) to visualize LLM decision-making in complex multi-agent systems. It also demonstrates using LLMs as judges to evaluate AI app performance at scale, targeted at developers building production agentic applications.