Full-text search across 57350 articles. Combine with topic and date filters; results sorted by relevance.
A developer built an automated adversarial red-teaming benchmark to stress-test an autonomous AI coding agent against prompt injections and scope escapes, finding that a compromised LLM persona successfully committed una…
A developer has compiled a forge-aware comparison of AI code review tools, arguing that popular GitHub-centric roundups fail teams running self-managed GitLab, Azure DevOps Server, or Bitbucket Data Center. The page docu…
A developer argues that Scrum is effectively dead, not because a better methodology replaced it, but because coding agents have made the Agile Manifesto's original principles practically followable for the first time. Th…
A principal engineer ran a nine-feature experiment comparing hexagonal architecture against a flat, architecture-neutral implementation for an AI coding agent, finding the hexagonal version took 38% longer to reach accep…
A developer measured 120 AI-generated code samples from Qwen2.5-Coder 1.5B and found that every case flagged as a swallowed error by a naive detector was either a false positive or a documented fallback, concluding that …
Shopify is moving its flagship Shop app and other mobile apps from React Native back to native Swift and Kotlin, citing internal AI coding agents that can port features between iOS and Android close to autonomously. The …
PostHog moved from roughly 20% of its monorepo pull requests being opened by AI agents to 70% over the last four months, according to a PostHog blog post, a shift the company says breaks the definition of an engineer as …
A backend engineer has outlined a practical workflow for auditing and fixing AI-generated code, warning that AI tools frequently produce code that looks correct but introduces hardcoded secrets, SQL injection through str…
A stealth startup called Gymwasp ran 372 issues through its agentic software factory, The Mandible, over 6 months, and 65% of issues were merge-ready after a single agentic review round, according to metrics collected by…
AI-generated code most often fails because models hallucinate library versions, invent nonexistent function parameters, or lose context in long conversations, according to a debugging guide that recommends pasting exact …
A developer argues that AI coding agents, being probability machines trained on a web where 96% of top homepages have accessibility issues per WebAim's One Million report, tend to reproduce inaccessible patterns rather t…
Andrej Karpathy's spec-driven method for AI coding reframes work with models like Claude as a three-layer system: a spec, a verifier, and a persistent environment. The spec layer has Claude interview the user about the a…
ECC (Everything Claude Code), an agent harness for Anthropic's Claude Code, surpassed 200,000 GitHub stars, with 32,500 forks and 230 contributors, according to the project's repository. The tool enforces a structured wo…
Alibaba's Qwen3.8-Max-0902 model, released September 2, 2026, topped Code Arena's WebDev leaderboard with 1,691 points, surpassing Claude Opus 5 Max's 1,687, while keeping its price unchanged at $2 per million input toke…
An autonomous AI agent named Turing reported finding 12 critical security vulnerabilities in AI-generated code within 24 hours, including command injection flaws in Datadog's Python APM library and a SQL injection in an …
A developer detailed a five-month project in which Claude Code, Anthropic's AI coding assistant, built its own tooling across a fleet of machines to manage a bike workshop's 20-year-old classic ASP booking system, result…
The European Commission has published a Code of Practice on marking and labeling AI-generated content, providing businesses with practical guidance for meeting transparency obligations under the EU AI Act. The code, appl…
Developer Achint Tiwari is building Achintisalive, an AI-powered educational co-pilot that acts as a Socratic tutor for secure coding. The tool, part of the Gen AI Academy APAC project, uses Google Gemini 1.5 Flash and S…
Anthropic shipped Claude Code v2.1.248 on Thursday with a new `--restricted` permission mode that disables shell execution, web fetching, and project-level settings files, addressing the TrustFall vulnerability class dis…
Concord, an open-source, local-first communication layer for AI coding agents, lets Claude Code, Codex, Cursor, Gemini CLI, and Grok Build exchange live messages, claim files to avoid overlapping edits, and hand off task…