Ai raised task throughput a third and code churn ninefold. faros's 2026 engineering report shows…
ai raised task throughput a third and code churn ninefold. faros's 2026 engineering report shows lines deleted against lines added on merged code up 861%, while per-developer task output grew 33.7%.
the tools made writing faster and deleting faster.
Context
Faros AI's 2026 report takeaways post says task throughput per developer is up 33.7%, epics completed per developer are up 66% and the pull request merge rate per developer is up 16.2%. It defines code churn as the ratio of lines deleted to lines added for merged code in a given quarter, and says it has increased 861% under high AI adoption, at nearly 10 times the prior rate.
The post lists plausible explanations for the churn rise and says all of them are consistent with the data and that the right one likely varies by organization. They include developers accepting AI-generated code quickly and returning to replace it when it proves insufficient. The post says that with line-level provenance data an organization can tell whether deleted lines were recently written, which suggests rework, or were legacy code being refactored.
The 33.7% and 861% figures match the post. The 861% is a change in the deleted-to-added ratio under high AI adoption, not a count of lines, and the note's 'lines deleted against lines added on merged code' reads it correctly. The 'ninefold' wording rounds 'nearly 10 times'.
The note's last sentence, that the tools made writing faster and deleting faster, is the author's reading. The post does not say deletion got faster. It says deletion rose relative to additions and that rework of AI code and productive refactoring of old code both fit the data.
The report is vendor research from a company that sells engineering analytics. The post describes the figures as reported by Faros and the dataset is not shown in what was read here. Whether the 861% comes from high-adoption organizations only, and how many, is unsupported here, not refuted.
Related work
- The AI Engineering Report 2026: The AI Acceleration Whiplash, Ten Takeaways (Faros AI, April 12, 2026) ↗Primary source for the 33.7%, 66%, 16.2% and 861% figures, the churn definition and the three explanations.
- AI Impact on Engineering Productivity: 2026 Report Data (Faros AI) ↗The report's own data page; surfaced in search and not read in full here.
Watch next
- Open the report's data page for the sample size and the definition of high AI adoption. Look for an independent measure of code churn under AI tools.
Sources
Provenance
The note above is reproduced unedited from the original post, first published on Threads on 9 October 2026 at 01:18 IST. Sources are the papers and datasets the note draws on.
View the original post ↗Embed this note
More notes
The air is now being asked to keep its own ledger
the air is now being asked to keep its own ledger: ecmwf’s aifs compo becomes the first ai model to forecast atmospheric composition globally every three hours, cleanair simulates 365 days of pm2.5 over china in ten seconds, and a unified framework maps six pollutants at one kilometer across the whole country. the air now files its own composition report.
read the note →The current is now being asked to draw its own map
the current is now being asked to draw its own map: china’s langya 2.0 predicts six ocean phenomena including internal waves and mesoscale eddies, a deep net called wenhai resolves eddies globally with air sea flux formulas built in, and scripps infers surface currents from the way temperature patterns deform in satellite images. the ocean now files its own circulation report.
read the note →The soil is now being asked to report its own carbon
the soil is now being asked to report its own carbon: a nix color sensor paired with generative data augmentation predicts soil organic carbon without a lab, random forest drives 74 percent of soil health mapping studies, and sentinel 2 tracks five year carbon change across france and italy from 922 samples. the dirt now files its own carbon account.
read the note →