At the 42 Berlin × Needle Hackathon (21–22 May 2026), my teammate and I entered Midknight Watcher, an autonomous coding agent. Given a hidden spec, it plans, writes code, runs the tests and repairs its own work, all night, with no human writing the code.
My role: Sushi (Shrish Arunesh) wrote the agent. I monitored the overnight run and handled the Friday-morning recovery.
The agent passed 92 of 150 public tests, every line written by the agent itself, and 30 of 100 on the hidden test set. It runs on a custom loop of about 620 lines instead of a framework, so it could be patched live, with a three-tier model fallback: Qwen3-235B, then Llama 3.3 70B, then a local Qwen3.5 9B.
When an API goes down, the agent drops to the next tier. Qwen and Llama fail in different ways, so the backup covers the primary's blind spots instead of repeating them.
Iteration 30 introduced an infinite loop. Because every iteration was committed, one
git checkout restored the 92/150 version.
Some overnight runs scored 0/150, which looked like the agent had broken everything. It hadn't: file timestamps were confusing the test runner. The code was fine; the measurement was wrong.
One test hung and went unnoticed until five Python processes were still running. A timeout on every test run is cheap insurance.
Assume outside services will fail, keep every version recoverable, and check the measurement before trusting the score. Daytonator and Router Rumble both keep working when their AI isn't available, and that habit started here.