Dataset: 2026-10-27-catalog-ai-v2
Ruleset: b2e8889c66f7
Scanned: 2026-10-27
Window: February–September 15, 2026

Catalog tracking observation. This repository is measured with the same fixed window and ruleset, but it is not part of the published benchmark or its aggregate claims.

Apache Beam is a unified programming model for Batch and Streaming data processing.

01Result

apache/beam ↗

apache/beam AI coding statistics: 4.9%

of 1,329 commits in February–September 15, 2026 were committed by an AI agent — a floor, not an estimate

4.9% agent · top contributor 20%33 contributors

org-wide

Adoption is broad. Many contributors are committing agent-authored work, which is what a real organisational rollout looks like.

agent commits

1,264

other considered commits

agent contributors

20%

largest contributor

02Drivers

What shapes this result

Mode, timing, and detected signatures show whether the result reflects broad adoption or one concentrated workflow.

Assisted or autonomous

65assisted

0autonomous

Assisted — a developer’s identity authored the commit and the agent added itself as co-author. Autonomous — a bot identity authored the commit without a human identity attached.

Agent share by month

Peak month: 7.7%. Bars are scaled to that peak, not to 100%.

Which agents

Agent	Vendor	Commits	Share
Gemini CLI	Google	35	2.6%
Claude Code	Anthropic	26	2%
Cursor Agent	Anysphere	3	0.2%
GitHub Copilot	GitHub	1	0.1%
OpenAI Codex	OpenAI	1	0.1%

03Evidence

Inspect the matches

Public commit samples let you inspect the signatures behind the aggregate result.

Matched commit samples

9c82e053 on GitHubAvoid output inside try-catch in Java IO (#39124)Gemini CLI
5e499388 on GitHubKeyCommitTooLarge logging improvements (#39316)Gemini CLI
55e1ecbb on GitHubAdd metric for Kafka offset commit failures (#38889)Gemini CLI
f5d35728 on GitHub[IcebergIO] Spec aware file reserialization (#39973)Claude Code
e9c37d53 on GitHubRefactor: delete dead SplitAndSizeRestrictions handling from FnApiDoFnRunnClaude Code
34a00c77 on GitHubAddFiles: extract bounded async task plumbing and Parquet footer reads (#3Claude Code
bef9ef19 on GitHub[Python] Reuse Milvus client in RAG vector sink write path (#39378)Cursor Agent
96733599 on GitHubFix JsonToRow swallowing downstream errors when runners fuse transforms. (Cursor Agent
889ac54b on GitHubFix cancelled run handling in flaky test prefetcher (#38579)Cursor Agent
44d4089a on GitHubPotential fix for environment variable built from user-controlled sources GitHub Copilot
023ca6d9 on GitHub[Java] Cache DoFn invoker constructors per instance (#39310)OpenAI Codex

04Method

What entered the denominator

The result begins with commit history, removes merges and routine automation, and reports literal agent signatures as a lower bound.

Exclusion ledger

Commits in window	1,786
Merge commits removed	−116
CI / dependency bots removed	−341
Considered	1,329

Bots are excluded from the denominator entirely. Agent attribution requires a literal supported signature; no code-style inference is used.

Continue exploring

Claude Code commitsBrowse repositories with detected Claude Code signatures →GitHub Copilot commitsBrowse repositories with detected GitHub Copilot signatures →Cursor Agent commitsBrowse repositories with detected Cursor Agent signatures →OpenAI Codex commitsBrowse repositories with detected OpenAI Codex signatures →Gemini CLI commitsBrowse repositories with detected Gemini CLI signatures →Java AI coding statisticsCompare evidence across Java repositories →GitHub AI commit analyzerInspect another repository with the same evidence model →

Reproduce this observation

Run the same detector locally.

Deterministic signature matching; no code-style inference. Squash merges can remove trailers, so this result remains a measured floor.

npx qmmit-cli apache/beam

Scan another repository

Try

Month	Agent share	Considered commits
2026-02	7.7%	143
2026-03	5.5%	182
2026-04	2.4%	170
2026-05	5.1%	175
2026-06	5.6%	197
2026-07	7.1%	196
2026-08	2.2%	180
2026-09	2.3%	86