Validate structured model output before using it
Treat deserialized model output as untrusted data and apply normal business validation before it drives application behavior.
Ai-Engineering
Testing Microsoft Agent Framework Applications
How to test Microsoft Agent Framework applications with fake model clients, tool contract tests, structured output tests, routing tests, workflow tests, and eval-style regression checks.
Use HybridCache for expensive AI work in .NET
Use HybridCache to avoid duplicate AI model calls, with a cached summarization example and deliberate cache-key design.
Keep agent tools narrow and boring
Expose only the small, specific capabilities the agent needs for the current workflow.
Use evals before changing prompts
Treat prompt changes like code changes: measure the behavior before deciding whether the edit helped.
Keep RAG citations tied to retrieved chunks
Citations should point back to the retrieved evidence, not to sources the model guessed after answering.
Observability for Agents with OpenTelemetry, Aspire, and Application Insights
How to make Microsoft Agent Framework behavior visible in production with OpenTelemetry traces, token usage, model calls, tool calls, Aspire local debugging, and Application Insights monitoring.
Use CSnakes when Python is a function call, not a service
Embed a small typed Python capability in .NET when an HTTP service would be more boundary than value.
Expose a Python FastAPI AI tool as an Aspire service
Keep Python AI code behind an HTTP boundary and let Aspire wire it into the local system.
Keep model name and deployment name in config
Do not hard-code model routing decisions into application logic.