EpochAI: Codex Engineer Productivity on the Rise
SamuelAlbanie · x · 2026-07-19
This EpochAI study estimates the productivity boost among Codex engineers. The methodology involves using an LLM to reverse-engineer how long a merged PR would take an engineer without AI assistance.
They found that in Q2 2026, 8% of contributor-days are estimated to equate to over 24 hours of manual work, a notable increase from 2% a year ago. The author cautions that this method still makes strict causal attribution difficult.
Related event: Study: AI Significantly Boosts Engineer Productivity(2 posts)→
More from coding & agent
- Dev turns DJI Mavic into local AI security drone built with Opus 5.5 — alexcovo_eth · 2026-10-27
- Microsoft Research: ForkPilot cuts long-horizon agent compute by up to 59% — EliasEskin · 2026-10-27
- Cursor launches Agent SDK: agents as code with Markdown prompts and TypeScript tools — soleio · 2026-10-27
- Vibecheck, the 'usable formal methods' hackathon, sells out two floors in SF — spikedoanz · 2026-10-27
- How do you verify AI agent outputs beyond unit tests and LLM-as-judge? — Practical_Green5775 · 2026-10-27
- AI agents may soon become the only way to run code, not just write it — IgorCarron · 2026-10-27