Review economics
karpathy/autoresearch
Read from the 7 most recently merged pull requests ·
Not enough merged pull requests on both sides yet to compare how they land in review.
≥25%
Attributed
a floor
2
Attributed PRs
5
Other PRs
7
Read
Attributed — the latest 2
The rest — the latest 5
#342 fix(analysis): define best_bpb before y-axis scaling
0 reviews · 6.3d to merge
#301 add AMD ROCm fork to notable forks section
0 reviews · 0h to merge
#20 Honor --download-workers instead of hardcoding 8 download workers
0 reviews · 2h to merge
#17 Fix agent crash blindspot by forcing it to read traceback
0 reviews · 7h to merge
#2 add fallback FA3 kernel for non-Hopper GPUs
0 reviews · 4h to merge
How this was measured
25% of merged commits carry agent attribution — a floor, not the share; tools that only complete code inline leave no commit trail, so the unattributed side includes AI-assisted work.
Detected: Claude Code.
“Attributed” means a commit carried an agent’s signature — a co-author trailer, an agent commit identity, or an agent bot account. Tools that only complete code inline leave no such mark, so the other column is “rest”, not “human-written”. Full method and its limits
The badge reads this repo’s current report, so it follows the number.
Tell me when this moves
We re-read karpathy/autoresearch weekly and email only when the number changes materially.
Read your own repos
This one is public. For a private repo, run the same read locally through your own GitHub credentials — nothing is installed and nothing is sent to us.
npx @ambera/review-taxWant this continuously on karpathy/autoresearch — each pull request paired to the task it came from, and the work graded from its review loop rather than its diff size? Claim this repo in Forge
Browse every repo people have read · Read a different repository