Google's FlowAgent fixes failing CI tests at scale, with 28,554 fixes applied
rohanpaul_ai · x · 2026-11-01
A Google paper introduces FlowAgent, an AI agent deployed inside Google's internal tools Critique and Cider that automatically repairs pre-submit test failures in CI.
- Existing program-repair agents mostly work offline post-merge, missing the low-latency needs of developers mid-flow.
- FlowAgent uses a ReAct-style generate-and-validate loop plus pre- and post-execution abstention filters to keep quality under strict latency constraints.
- Manual evaluation on 195 real test failures showed 67.18% accuracy; after Google-wide deployment it suggested fixes on 295,508 changes, with developers previewing 65,069 and applying 28,554.
- Developer interviews indicate the agent is useful for suggesting fixes.
Related event: Google's FlowAgent Auto-Fixes Failing CI Tests(2 posts)→
More from coding & agent
- Dev shows off AI-generated game engine material graphs built from text descriptions — D3VAUX · 2026-11-01
- web-shooter: a local MCP web reader returning answers at ~1k tokens vs ~25k — Dazzling_Dirt_7729 · 2026-11-01
- UIUC, Meta, Amazon & DeepMind release 135-page survey charting agentic reasoning for LLMs — mdancho84 · 2026-11-01
- Dev recalls 2023 function calling as the inflection point for AI coding — dfinke · 2026-11-01
- An escape game where 8 local LLM agents fight for 2 helicopter seats — ViceGodIm · 2026-11-01
- Fake bank chatbot leaked data in 54 of 55 attacks; one gateway cut it to zero — its_vayishu · 2026-11-01