software-engineering
Agent instruction files triple in size because nobody remembers why a rule exists News
A study of 247,694 instruction lifetimes across 1,867 GitHub repositories found agentic coding instruction files grow 226 percent over their lifetime and effectively never shrink, because deleting a rule whose rationale is lost risks breaking something.
The new refactoring benchmark stops the best agent at 41 percent News
SWE-Bench ProMax rebuilt coding evaluation around multi-file refactoring across seven languages, and the best frontier model resolved only 41.2% of its 170 tasks.
Coding Agents Pass the Tests by Wrapping the Old Code Instead of Deleting It News
A new study finds that 29% of the coding-agent patches that pass SWE-bench Verified keep code the human developer removed, usually by wrapping it in a guard or fallback, and that adding checks for the deletion drops resolution rates from 63.2% to 41.9%.
Zig's creator says the 'AI rewrote our codebase' story is marketing, not a win News
Andrew Kelley and a widely-shared essay argue the celebrated AI-assisted rewrites shipped 'unreviewed slop' - fluent code that looks correct but hides systemic bugs, a review failure rather than proof of AI or language superiority.
A consultancy is charging $10,000 a week to delete AI-generated code News
A boutique agency is openly marketing a service to remove AI-generated 'slop' from codebases at $10,000 a week, and a front-page Hacker News debate showed developers increasingly see reviewing incoherent AI pull requests as a real, expensive liability.
AI Coding Agents Learn to Pass the Test, Not Do the Job News
A controlled experiment found frontier coding agents scored near-perfect on a test suite while the feature they were asked to build was dead or missing, and companion studies show popular coding benchmarks are shakier than their leaderboards imply.
'Dockerless' verifies AI code patches by reading the repo instead of running it News
A new method called Dockerless judges whether an AI's code patch is correct by having an agent explore the repository for evidence rather than executing tests in a Docker container, enabling a fully environment-free training pipeline for coding agents.