Uber describes AutoClean, an AI system that removes stale feature flags
Overview
Uber's AutoClean has removed more than 16,000 stale feature flags and over 1 million lines of dead code, according to a summary by Meng Shao (@shao__meng) of an Uber Engineering post.
The system splits the work into stages: PolyglotPiranha, a deterministic engine, handles boolean simplification; AI agents make the judgment calls; full monorepo CI checks repairs; and CodeCourt checks four correctness properties before human review.
The figures come from the secondhand summary, not from Uber's post directly.
Written by AI from the articles below · updated Oct 20, 9:59 PM ET
Check the sources:
Article timeline
The articles in this story. Times are ET.
meng shao@shao__mengXUber uses AutoClean agents and validators to remove stale feature flagsAIUber's AutoClean has removed more than 16,000 stale feature flags and over 1 million lines of dead code, according to the author's summary of an Uber Engineering post. It splits work between a deterministic engine, PolyglotPiranha, for boolean simplification, agents for judgment calls, full monorepo CI for repairs, and CodeCourt, which checks four correctness properties before human review.

Heat Trend
- Current comparable heat
- 9
- Peak comparable heat
- 10Oct 20, 10:00 PM ET
- Change in past 24 hours
- –
The trend compares only continuously observed participants, so its scope may be smaller than the current heat total. Hover or tap for hourly heat; use arrow keys with a keyboard.