Working notes · Azure & AWS
Grounding, agents and evaluation - with the numbers attached.
Notes from production deployments on Copilot Studio, Azure AI Search, Azure AI Foundry and Amazon Bedrock. Retrieval quality, agent routing, eval gates, and the invoices that follow - measured before claimed.
- 1
- Posts
- 4
- Platforms
- 2026
- Since
How to effectively use Application Insights for Copilot Studio agents
Two telemetry scopes, one conversation ID, and the queries that turn "the agent gave me a wrong answer yesterday" into a root cause you can point at.
Read the piece →16 min · 3,173 words