Automating Low-Risk Code Review at Meta: RADAR, Risk Calibration, and Review Efficiency

• URL: http://arxiv.org/abs/2605.30208v1
• Date: Thu, 28 May 2026 16:44:07 GMT
• Title: Automating Low-Risk Code Review at Meta: RADAR, Risk Calibration, and Review Efficiency
• Abstract summary: At Meta, significant lines of code per human-landed diff grew by 105.9% year over year.<n>Per-developer diff volume rose 51%, with agentic AI responsible for over 80% of that growth.<n>We ask three questions that progress from feasibility through calibration to impact.
• Score: 10.379265985245537
• License: http://creativecommons.org/licenses/by-sa/4.0/
• Abstract: AI-assisted coding tools have altered software production. At Meta, 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 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? 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 increased 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. RADAR reduces median time to close by over 330% and median diff review wall time by 35%. Risk-aware layered automation can materially reduce review bottlenecks created by AI-driven code growth without compromising production safety.

Related papers

• Beyond Lexical Metrics: Sentence-Embedding Detection of Reviewer Habituation in AI Code Review [2.9152186617858304]
  We study 11,429 reviews paired with 10,104 human-authored inline comments from AI.<n>Approval rates rise from 30.5% in reviewers' early periods to 36.6% in late periods.<n>Four hand-crafted linguistic features - lexical diversity, Shannon entropy, technical specificity, and constructive actionability - show no monotonic decline across exposure deciles.
  arXiv  Detail & Related papers  (2026-09-05T18:15:18Z)
• From Code Review to Code Critique: Intent, Drift, and Spotlight for AI-Generated Diffs at Scale [2.8980676513012846]
  ARCTIC is an AI-powered Code Critique system that reframes code review around three capabilities: intent prediction, drift detection, and code spotlight.<n>In the experimental rollout, the drift scores reduces code misalignment by an additional 5.76 points (p = 0.026), intent prediction receives 90.2% approval, and zero defects have been attributed to self-reviewed diffs since launch.
  arXiv  Detail & Related papers  (2026-07-31T15:17:25Z)
• NOVA: A Verification-Aware Agent Harness for Architecture Evolution in Industrial Recommender Systems [24.27056236910489]
  NOVA is a level-aware agent harness for verification-aware architecture evolution.<n>It achieves the highest effective pass rate on L2 ScaleUp and L3 Literature-to-Production tasks.<n>It shortens one literature-to-production cycle by over 13x in human-attended time.
  arXiv  Detail & Related papers  (2026-06-25T16:30:39Z)
• Amplifying, Not Learning: Fine-Tuned AI Text Detectors Amplify a Pretrained Direction [51.56484100374058]
  Text detectors amplify a pretrained typicality axis; they do not construct an AI-vs-human boundary.<n>On raw encoders before any task supervision, projecting onto centroid(AI)-centroid(HC3) achieves NYT-vs-HC3 AUROC 0.806/0.944/0.834 across three architectures.<n>On RoBERTa-base, raw projection exceeds fine-tuning; on RoBERTa-base, full fine-tuning reduces discrimination below raw on both fluent-formal populations tested.
  arXiv  Detail & Related papers  (2026-05-20T19:08:38Z)
• AIRA: AI-Induced Risk Audit: A Structured Inspection Framework for AI-Generated Code [0.0]
  Practitioners have reported a directional pattern in AI-assisted code generation: AI-generated code tends to fail quietly.<n>This paper introduces the Reward-Shaped Failure Hypothesis - the proposal that this pattern may reflect an artifact of optimization through human feedback.<n>We present AIRA, a deterministic 15-check inspection framework designed to detect failure-untruthful patterns in code.
  arXiv  Detail & Related papers  (2026-04-19T19:32:52Z)
• TDAD: Test-Driven Agentic Development - Reducing Code Regressions in AI Coding Agents via Graph-Based Impact Analysis [0.0]
  TDAD builds a dependency map between source code and tests so that before committing a patch, the agent knows which tests to verify and can self-correct.<n>This paper presents TDAD, an open-source tool that performs pre-change impact analysis for AI coding agents.
  arXiv  Detail & Related papers  (2026-03-18T17:38:22Z)
• Will It Survive? Deciphering the Fate of AI-Generated Code in Open Source [3.6525095710982924]
  A prevailing hypothesis suggests that code is "disposable", meaning it is merged quickly but discarded shortly thereafter.<n>We investigate this hypothesis through survival analysis of 201 open-source projects, tracking over 200,000 code remediation units authored by AI agents versus humans.
  arXiv  Detail & Related papers  (2026-01-23T15:00:46Z)
• Unraveling the "Anomaly" in Time Series Anomaly Detection: A Self-supervised Tri-domain Solution [89.16750999704969]
  Anomaly labels hinder traditional supervised models in time series anomaly detection. Various SOTA deep learning techniques, such as self-supervised learning, have been introduced to tackle this issue. We propose a novel self-supervised learning based Tri-domain Anomaly Detector (TriAD)
  arXiv  Detail & Related papers  (2023-11-19T05:37:18Z)
• LiDAR View Synthesis for Robust Vehicle Navigation Without Expert Labels [50.40632021583213]
  We propose synthesizing additional LiDAR point clouds from novel viewpoints without physically driving at dangerous positions. We train a deep learning model, which takes a LiDAR scan as input and predicts the future trajectory as output. A waypoint controller is then applied to this predicted trajectory to determine the throttle and steering labels of the ego-vehicle.
  arXiv  Detail & Related papers  (2023-08-02T20:46:43Z)
• RADAR: Robust AI-Text Detection via Adversarial Learning [69.5883095262619]
  RADAR is based on adversarial training of a paraphraser and a detector. The paraphraser's goal is to generate realistic content to evade AI-text detection. RADAR uses the feedback from the detector to update the paraphraser, and vice versa.
  arXiv  Detail & Related papers  (2023-07-07T21:13:27Z)
• AAVAE: Augmentation-Augmented Variational Autoencoders [43.73699420145321]
  We introduce augmentation-augmented variational autoencoders (AAVAE), a third approach to self-supervised learning based on autoencoding. We empirically evaluate the proposed AAVAE on image classification, similar to how recent contrastive and non-contrastive learning algorithms have been evaluated.
  arXiv  Detail & Related papers  (2021-07-26T17:04:30Z)
• ScoreGAN: A Fraud Review Detector based on Multi Task Learning of Regulated GAN with Data Augmentation [50.779498955162644]
  We propose ScoreGAN for fraud review detection that makes use of both review text and review rating scores in the generation and detection process. Results show that the proposed framework outperformed the existing state-of-the-art framework, namely FakeGAN, in terms of AP by 7%, and 5% on the Yelp and TripAdvisor datasets.
  arXiv  Detail & Related papers  (2020-06-11T16:15:06Z)
• DADA: Differentiable Automatic Data Augmentation [58.560309490774976]
  We propose Differentiable Automatic Data Augmentation (DADA) which dramatically reduces the cost. We conduct extensive experiments on CIFAR-10, CIFAR-100, SVHN, and ImageNet datasets. Results show our DADA is at least one order of magnitude faster than the state-of-the-art while achieving very comparable accuracy.
  arXiv  Detail & Related papers  (2020-03-08T13:23:14Z)
• Reanalysis of Variance Reduced Temporal Difference Learning [57.150444843282]
  A variance reduced TD (VRTD) algorithm was proposed by Korda and La, which applies the variance reduction technique directly to the online TD learning with Markovian samples. We show that VRTD is guaranteed to converge to a neighborhood of the fixed-point solution of TD at a linear convergence rate.
  arXiv  Detail & Related papers  (2020-01-07T05:32:43Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.