We’ve been working on Scout QA, a tool that uses AI to catch subtle bugs that usually slip past manual testing and reviews.

Instead of focusing on test automation, Scout analyzes your app’s UI, flows, and logs to surface tiny inconsistencies — things like broken states, unclear error messages, or small regressions that humans might ignore because “it still works.”

The goal isn’t to replace testers but to act like an extra pair of curious eyes — pointing out issues that affect real users’ experience.

We’re exploring questions like: - How can an AI develop an intuition for “something feels off”? - What kind of feedback loop helps it learn from real user bugs?

Would love feedback from developers, QA engineers, and indie makers: what small-but-painful bugs do you wish tools could catch automatically?