What it does
Reviews code against plain-language team rules, using Jev to turn each rule into a probability-backed finding.
Agent tooling
Only you can see your notes.
Screenshot unavailable. Open the experiment ↗
Reviews code against plain-language team rules, using Jev to turn each rule into a probability-backed finding.
Start by writing your review questions in plain English. For example, ask whether error messages get saved. A developer can then set up Hunch to check new code against your questions. It can ask an outside web service or run a private helper model directly on the developer's computer.
The tool flags possible rule issues for humans to inspect, rather than proving a mistake happened. Because sending code off-site shares files with another company, your team can choose the local option to keep private code on their own machines.