#ci

Hacker News highlighted SWE-CI, an arXiv benchmark that evaluates whether LLM agents can sustain repository quality across CI-driven iterations, not just land a single passing patch.

A front-page Hacker News thread drew attention to SWE-CI, an arXiv benchmark that evaluates coding agents on 100 real repository evolution tasks rather than one-shot bug fixes. The paper frames software maintainability as a CI-loop problem and reports that even strong models still struggle to avoid regressions over long development arcs.