Presentation: Teaching Engineers, Trusting AI: How Education Enabled Autonomous Code Review
Not a pilot. Duolingo deployed an autonomous code-review agent across engineering, paired it with literacy training, and didn't tank quality.
Why it matters
Autonomous code review at scale requires more than the agent itself—it demands developer education and observability. Duolingo's case study shows how cultural adoption, not just tool access, unlocks agent reliability in production.
The key facts
11 to knowDuolingo deployed automated PR risk-assessment bot for code review
Internal AI literacy workshops paired with agent guardrails
Observability dashboards tracked agent performance
Delivery speed increased without increase in defect rates
Focus on cultural AI adoption beyond tooling access
Duolingo deployed autonomous PR risk-assessment bot in code review
Paired agent deployment with targeted developer education workshops
Implemented observability dashboards for agent monitoring
Delivery speed increased without increasing defect rates
Case study on cultural AI adoption beyond tooling access
Safe AI guardrails paired with agent autonomy for risk management
Go to the source
InfoQ AI/MLinfoq.com
Publisher excerpt: Sarah Deitke discusses how Duolingo drives cultural AI adoption beyond tooling access. She explains their internal AI literacy workshops and observability dashboards, then shares a case study on redesigning code review using an automated PR risk-assessment bot. Deitke demonstrates how pairing…