Refactoring Codebases at Scale
Comments on “Refactoring Codebases at Scale”
Claim-checked. Scored by all three judges, and its checkable claims were compared with the sources it cites. How far stories are checked
MerciYanis engineers report rewriting about 250,000 lines across 10 repositories in one day, using roughly 30 parallel agents and 1.6 billion tokens. Their method was to constrain the output rather than review it: contract-style tests through public APIs, a Git check that rejects any modified test file, end-to-end tests, and a handbook of rules. Only two small regressions reached QA. All figures come from the company itself, and the approach relied on unreleased-sounding model versions.
Why it ranked
No. 11 of 124 with a score of 0.70. The panel rated it highest on new information (0.90) and reader interest (0.84), lowest on significance (0.63).
Every answer behind this is in the table under Score.
Score
0.70total under rubric v2.2
- Significance0.63
- Reader interest0.84
- Audience breadth0.69
- Practical relevance0.82
- Substantiated0.66
- Technical depth0.81
- New information0.90
- Counts against: Promotional0.12
- Counts against: Overclaim0.30
Each judge’s answer per question (0–1), their mean and spread, and what the question adds to the score (weight × mean).
| Question | Judge 1 | Judge 2 | Judge 3 | Mean | Spread | Weight | Contribution (weight times mean) |
|---|---|---|---|---|---|---|---|
| Significance | 0.71 | 0.45 | 0.72 | 0.63 | 0.27 | +0.24 | +0.150 |
| Reader interest | 0.81 | 0.80 | 0.90 | 0.84 | 0.10 | +0.26 | +0.218 |
| Audience breadth | 0.67 | 0.70 | 0.70 | 0.69 | 0.03 | +0.13 | +0.090 |
| Practical relevance | 0.86 | 0.75 | 0.86 | 0.82 | 0.11 | +0.22 | +0.181 |
| Substantiated | 0.79 | 0.50 | 0.68 | 0.66 | 0.29 | +0.05 | +0.033 |
| Technical depth | 0.94 | 0.70 | 0.78 | 0.81 | 0.24 | +0.05 | +0.040 |
| New information | 0.88 | 0.85 | 0.96 | 0.90 | 0.11 | +0.05 | +0.045 |
| Promotional | 0.08 | 0.20 | 0.08 | 0.12 | 0.12 | −0.19 | −0.023 |
| Overclaim | 0.22 | 0.35 | 0.32 | 0.30 | 0.13 | −0.11 | −0.033 |
| Sum of contributions | +0.702 | ||||||
| Claim-check penalty | 0.000 | ||||||
| Total, as published | 0.702 | ||||||
Spread is the highest minus the lowest judge answer. marks a spread above 0.40: the judges disagree, and the story is marked contested.
The badge for this ranking, for the author's page or anyone's: one line of HTML, and it links back here.
<a href="https://rubric.hrih.si/editions/2026-10-27/11/"><img src="https://rubric.hrih.si/badge/2026-10-27/11.svg" alt="Ranked #11 on Rubric, 27 Oct 2026" width="245" height="28"></a>Claim check
Claim check: 2 matches source · 6 no source given. Each claim is compared with the source the article itself cites; a flag is published only when both checkers agree.
No source givenNumber
The team rewrote roughly 250,000 lines of code across 10 repositories in a single day, at a cost of about 1.6 billion tokens, one engineer's full-time day and some 30 parallel agents.
The article gives no link or citation for it.
The article says “roughly 250,000 lines of code, spread across 10 repositories, in a single day. The whole initiative cost us about 1.6 billion tokens, one engineer’s full-time day”
The source says No source to compare with.
No source givenNumber
The rewrite also fixed about 50 bugs and edge cases.
The article gives no link or citation for it.
The article says “it also fixed about 50 bugs and edge cases”
The source says No source to compare with.
No source givenNumber
Before the rewrite, test coverage averaged around 60% across repositories, and the team pushed it to 100% on every repository and service.
The article gives no link or citation for it.
The article says “coverage averaged around 60%, and some repositories were already close to 100%”
The source says No source to compare with.
Matches sourceQuote
Martin Fowler calls the style of isolating a function and mocking every collaborator the 'mockist style', as opposed to the classical style that uses real objects where possible.
The cited source says the same thing.
The article says “Martin Fowler calls this the mockist style, as opposed to the classical style where real objects are used wherever possible”
The source says “The classical TDD style is to use real objects if possible and a double if it's awkward to use the real thing”
Source cited by the article: martinfowler.com/articles/mocksArentStubs.html
Matches sourceQuote
Kent Beck lists structure-insensitivity, meaning tests should not change their result if the code's structure changes, as one of his Test Desiderata.
The cited source says the same thing.
The article says “tests should be structure-insensitive, meaning they should not change their result if the structure of the code changes”
The source says “Structure-insensitive — tests should not change their result if the structure of the code changes”
Source cited by the article: kentbeck.github.io/TestDesiderata
No source givenNumber
After the rewrite was deployed to QA, only two small regressions surfaced, and both were fixed within minutes.
The article gives no link or citation for it.
The article says “only two small regressions surfaced, and both were fixed within minutes”
The source says No source to compare with.
No source givenCausal conclusion
Opus 5.5 and GPT 5.6 Sol reached a level of reliability that made the rewrite possible; without it no process would have held up at this scale.
The article gives no link or citation for it.
The article says “Opus 5.5 and GPT 5.6 Sol have reached a level of reliability we wouldn’t have bet on a year ago”
The source says No source to compare with.
No source givenGeneralization
Pull request review has become the new bottleneck, because agents open PRs faster than humans can read them.
The article gives no link or citation for it.
The article says “Pull request review has become the new bottleneck. Agents open PRs faster than we can read them”
The source says No source to compare with.
Also covered by
No other coverage was clustered with this story.
Elsewhere
Elsewhere: HN 2 pts, not on front page
Recorded for context only, checked 27 Oct 2026, 05:46 UTC. It played no part in the ranking.