I hunt the bugs that survive code review and only detonate in production — the ones a second replica exposes, the ones a policy engine quietly hides, the ones that spin forever without ever throwing. Then I prove them.
+ Read the codebase until the failure mode is obvious
+ Write the test that FAILS on main
+ Fix it in the smallest diff the maintainers would have written themselves
+ Watch that same test PASS
- No padding. No drive-by nitpicks. No "looks good to me" reviews.
| Project | Landed | The bug |
|---|---|---|
| CopilotKit/OpenBot AI coworker platform |
 |
A snapshot cache living in process memory meant element refs silently stopped resolving the moment a second replica existed — moved it into Postgres. A deny rule written for one action surface refused every other surface too. A policy dry-run recorded a refused connector call as a success. |
| edwardkim/rhwp Rust document engine |
 |
Rendering fidelity against ground-truth documents, the CLI & agent surface, and a long campaign of static-analysis panic & DoS fixes. |
| leookun/cursor-byok Local Cursor backend |
 |
A tool-call id reused across rounds wedged the run forever, with no timeout on that path. A codec alias gap aborted the turn on every Bash call, mid-stream. Empty tool arguments killed the entire run with EOF while parsing a value. |
| shy3130/tick-stock-panel A-share quant workbench |
 |
A forced-exit signal evaporated under pandas copy-on-write — chained assignment, dropped without a warning. An upload handler read whole files into memory with no cap at all. |
| lightningpixel/modly Image to 3D desktop app |
 |
Headless runs wrote their results into an unindexed folder the app could never see. |
| akitaonrails/ai-memory Long-term memory for agent CLIs |
 |
A packaging test that could never pass on Windows — found by running the suite where CI does not. |
|
Ask patent questions in plain language. Get answers with the measurement shown, not asserted. |
A live treemap of exactly what your AI coding agent is doing to your disk. |
|
After the agent says it is done — prove the work actually happened. |
Ask out loud which public funding you can actually apply for. Self-hosted, real data. |
A maintainer closed one of my pull requests with: "Closing this as already landed, but you found a real bug and got there independently."
The fix had shipped inside someone else's PR — after I left a review pointing at the exact failure mode.
That is the job. Not the badge on the pull request — the bug being gone.
The reviews I leave are the ones I would stake my name on. Silence otherwise.