Hey everyone! 👋
We've got something exciting lined up, and if you've been building with AI coding agents, this one's going to be right up your alley.
TestSprite is running its Season 3 Hackathon, themed around Loop Engineering — and it's all about trying the part of the AI coding workflow most people skip: the verification loop.
Here's the core idea: your coding agent (Claude Code, Codex, Antigravity, whatever you're working with) writes the code. The open-source TestSprite CLI acts as a real user who is running real tests against your live app and returning verdicts. Then your agent reads the failure bundle, fixes the root cause, and reruns. Passes bank. The loop continues.
Write → Verify → Fix → Verify Again → repeat
