一句话：在 Meta 部署 RADAR 多级漏斗自动审查并放行低风险 diff，量化风险阈值与安全、效率的权衡。
AI 辅助编码使 Meta 人均 diff 量同比增长 51%、落地代码行增长 105.9%，agentic AI 贡献逾八成，而及时获审 diff 比例下降，审查带宽成为瓶颈。作者部署 RADAR 多阶段漏斗：按作者身份与来源分类并施加资格门控，再经静态启发式、机器学习 Diff Risk Score、LLM 自动审查与确定性校验，合格 diff 自动落地。评估用 535K+ 条遥测、政策前后对比与双重差分分析。结果：331K+ diff 自动落地；风险阈值由 25 分位放宽至 50 分位使批准率达 60.31%；回滚率为非 RADAR 的 1/3、生产事故率为 1/50；相比人工审查，中位关闭时间缩短逾 4.3 倍，审查墙钟时间减少 35%。
AI-assisted coding tools have altered software production. At , significant lines of code per human-landed diff grew by 105.9% year over year and per-developer diff volume rose 51%, with agentic AI responsible for over 80% of that growth. Meanwhile, the share of diffs receiving timely review has declined, exposing a widening gap between code supply and reviewer bandwidth. We therefore ask three questions that progress from feasibility through calibration to impact: (1) can risk-stratified automation operate at scale across diverse organizations, (2) how does tuning the risk threshold affect the trade-off between automation yield and safety, and (3) to what extent does automated review reduce end-to-end latency for AI-generated changes? To answer them, we deployed RADAR (Risk Aware Diff Auto Review), a multi-stage funnel that classifies each diff by authorship and source type, applies eligibility gates, static heuristics, a machine-learned Diff Risk Score, LLM-based Automated Code Review, and deterministic validation before landing qualifying changes. We evaluate RADAR through telemetry covering 535K+ RADAR-reviewed diffs, observational before–after comparisons for policy changes, and difference-in-differences analysis of efficiency outcomes. RADAR has reviewed 535K+ diffs and landed 331K+. Relaxing the Diff Risk Score threshold from the 25th to the 50th percentile raised the approve rate to 60.31%. The revert rate for RADAR-reviewed diffs is 1/3 that of non-RADAR diffs, and the Production Incident () rate is 1/50 that of non-RADAR diffs. Compared to human-reviewed diffs, RADAR reduces median time to close by more than 4.3× and median diff review wall time by 35%. We conclude that risk-aware layered automation can materially reduce review bottlenecks created by AI-driven code growth without compromising production safety during incremental rollout.
@inproceedings{abs-2605-30208,
title = {Automating Low-Risk Code Review at Meta: RADAR, Risk Calibration, and Review Efficiency},
author = {Chris Adams and Arjun Singh Banga and Parveen Bansal and Souvik Bhattacharya and Rujin Cao and Pedro Canahuati and Nate Cook and Brian Ellis and Prabhakar Goyal and Gurinder Grewal and Tianyu He and Matt Labunka and Alex Manners and David Molnar and Ging Cee Ng and Vishal Parekh and Jiefu Pei and Frederic Sagnes and James Saindon and Will Shackleton and Sid Sidhu and Gursharan Singh and Karthik Chengayan Sridhar and Matt Steiner and Pratibha Udmalpet and Sean Xia and Stacey Yan and Audris Mockus and Peter C. Rigby and Nachiappan Nagappan},
booktitle = {ASE},
year = {2026},
doi = {10.1145/3832783.3834493}
}