When AI writes the code, who tests it?
Dec 4, 2026 | 3:00 PM - 3:10 PMStartup Stage
Dec 4, 2026 | 3:00 PM - 3:10 PM
Startup Stage
Description
Most developers now use an AI coding agent every day. The productivity gain is clear. The friction that follows is almost always about quality: code that ignores conventions, requirements that get missed, end-to-end journeys left uncovered.
Writing a feature now takes a few hours. Verifying it still takes as long as it did before. The bottleneck has moved from the code to the test. Generating tests with AI does not solve it. Someone still has to run them, maintain them and prove what they checked, every day.
In 10 minutes, with no jargon: what mass adoption of coding agents changes for software quality, what agentic verification looks like, and how one team doubled its release frequency with zero production incidents.




