Technical news
AI selects and contextualizes news about software engineering, architecture, applied AI, security and web performance.
Permanent archive · 182 news items
Search newsSource / arXiv
A new arXiv paper synthesizes evidence about coding agents and proposes measuring value through production-qualified changes, verification effort, cost, and risk.
Source / GitHub Blog
The GitHub Copilot experiment combines models and workflows with cost limits, isolated review, and safe change application, although its results remain benchmark-based.
Source / Next.js
The Next.js team closed 1,462 issues in three weeks by combining automated research, structured evidence, isolated reproduction, and an explicit reopening window
Source / The ASF Blog
The Apache Software Foundation combined specialized models, architectural context, and reviewed threat models to scan 230 repositories without turning every finding into maintainer
Source / Cloudflare Blog
Cloudflare’s new workflow combines code analysis, active routes, traffic, WAF events, and checks outside the model to prioritize findings with more context.
Source / GitHub Changelog
The latest CodeQL release expands JavaScript and TypeScript modelling and refines taint-flow and mutable-reference checks in GitHub Actions.
