Introducing Duet Autopilot: The self-improving agent for conversational AI | Decagon
Jun 9, 2026
Summary
This post introduces Duet Autopilot, a new capability for the AI Support Agent Platform. Autopilot automates the process of identifying issues in production conversations, tracing them to root causes, and proposing targeted fixes. It incorporates a self-validation loop using simulations against conversation data and curated test sets to ensure proposed updates are effective and do not introduce regressions. The system also enforces enterprise governance through human approval gates and provides detailed visibility into proposed changes and outcomes. The post highlights how Autopilot's own improvement loop leads to compounding gains in update quality over time.
More from the AI Support Agent Platform trail
- DuetBench-2: Measuring agent self-improvement in production | Decagon
- Autopilot in production: Governance of self-improving agents | Decagon
- Proactive customer service strategies for CX leaders | Decagon
- Automated customer service without losing the human touch | Decagon
- Your Decagon agent can now search the live web | Decagon