Skip to content
Trending storyMonitoring

Google's FlowAgent repairs failing CI tests before merge

1 article1 sourcesince Nov 1Last article

Overview

AISummary of 1 article

Google researchers describe FlowAgent, an AI agent that catches test failures during the pre-submit phase of continuous integration and runs an agent that edits, builds, and reruns tests until they pass.

Experts judged 67.18% of its fixes correct across 195 real failures, and developers applied 28,554 fixes in production, according to a report by Rohan Paul on X.

The median FlowAgent run took 9.85 minutes, compared with a 27.48-minute median before developers start editing. The report says many fixes still arrived too late to be used, so speed limits adoption.

Written by AI from the articles below · updated Nov 1, 6:32 AM ET

Check the sources:

Latest developmentNov 1, 6:31 AM ET
Google's FlowAgent fixes failing tests before merge, but speed decides adoption

Article timeline

The articles in this story. Times are ET.

Nov 1
  1. Rohan PaulX
    Google's FlowAgent fixes failing tests before merge, but speed decides adoption

    AIGoogle researchers introduce FlowAgent, an AI agent that repairs failing tests during the pre-submit phase of continuous integration. Experts judged 67.18% of its fixes on 195 real failures correct, and developers applied 28,554 fixes in production. Its median run took 9.85 minutes, compared with a 27.48-minute median before developers start editing, yet many fixes still arrived too late to be used.

    Image from @rohanpaul_ai's post

Heat Trend

Current comparable heat
4
Peak comparable heat
10Nov 1, 7:00 AM ET
Change in past 24 hours
-51%

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.