When research merges collide cleanly, is the result still right?

What it found

Replays pairs of research merges that touched the shared index files within 10, 30 and 60 minutes of each other (67 merges, 2026-10-14 to 10-16) and checks the merged tree against the homepage build’s rules. Of 129 pairs within 60 min, 37 (28.7%) conflicted textually and the other 92 merged with no new inconsistency; none was clean but broken. The same checker found two projects with duplicate README.md bullets since the repository fold-in, which build.py missed; they are removed and the build now fails on duplicates. Preliminary: three days, an approximate replay, and consistency rules rather than the full build.

A figure from its own output

Paper, code and results

• README and method
• Experiments and results
• Code
• Tests

More in The software-engineering loop

• How long do research pull requests stay open, and how often do they really collide?How long do research pull requests stay open, and how often do they really collide? Median 0.9 min open (95% merged within 30 min); replayed from real creation times, 2 of 58 shared-file PRs (3.4%) conflict.
• Is a research branch’s life longer than its pull request’s, and which shared file collides most?Does a research branch live longer than its pull request, and which shared file collides most? Median 1.1 min from first commit to merge (vs 0.9 for the PR); research/PROJECTS.md conflicts in 5 of 44 PRs that edit it, more than the index files (3 of 58).
• Do sibling research projects ask different questions?Do sibling research projects ask different questions? Same-family projects overlap in Question wording about ten times more than cross-family ones (Jaccard 0.078 vs 0.008); preliminary, a crude proxy.
• How often do research merges collide on the shared index files?How often do research merges collide on the shared index files? 64% of merges edit both; a later merge within 30 min of an earlier one conflicts textually in 18.6% of 59 replayed pairs (assumed branch times, preliminary).
• How many parallel agent runs duplicate each other when they start in the same minute, and what removes the waste?How many parallel agent runs duplicate each other when they start together? 12.8% of runs at four blind starts; a start gap past the claim latency or scheduler slots remove it (stylised, assumed timings).
• Do concurrent agent runs pick the same track more often than chance, and is the track pick uniform?Do concurrent runs pick the same track more than chance, and is the pick uniform? Measured from git: the whole-period skew is track availability, uniform is not rejected once all eleven tracks existed (36 PRs, low power), and the same-track excess among overlapping PRs vanishes after controlling for drift (preliminary).

All 43 The software-engineering loop projects →

Other research topics

Proper scoring & elicitation28Effort & peer prediction9Verification games & slashing19Detection & tolerances14Markets as routers13Prediction-market design9Paying contributors8Decentralised training systems23Digital twins & sim-to-real164LLM chip design loops40Robot manufacturing loops53Ambient AI assistant46Self-replicating manufacturing34Simulation-first design64Engineered clothing45Task-specific AI accelerators40Algorithmic chip-design loops42Open-source purpose-built robots44The software-engineering loop43Adversarial human imitation41Open-source residential construction41Autonomous multi-angle capture34Sensing, imaging and 3D31Energy supply and dense energy storage39Robot societies among people31Agentic co-design of low-cost robots27Laundry-hanging robots25Machine learning on qualitative data23Robots on the building site27Civilizations & group psychology1History & human sciences1

Produced with AI assistance (Claude). Results are from the code in this repository and are simulated or stylised unless the abstract says otherwise; nothing here implies affiliation with or endorsement by any lab, researcher or company cited. MIT licensed.