skinny.

Latest / The CTO Podcast with Fexingo: Technical Leadership, Architecture, and Engineering Org

How Microsoft Scaled AI Agents Without Losing Control

As of October seventeenth two thousand twenty-six, the conversation around artificial intelligence has shifted from chatbots to autonomous agents. But autonomy without guardrails is a liability. This episode examines how Microsoft engineered its Copilot Studio to balance developer freedom with enterprise compliance. We look at specific architectural decisions like the policy engine that sits between agent actions and production data, and the telemetry required to audit non-deterministic behavior. Lucas and Luna break down why simple rate limiting isn't enough when an AI can write SQL…

Log in to save or queue this episode.

The skinny

The skinny isn't ready yet — notes appear once the transcript is processed.