Full-text search across 56191 articles. Combine with topic and date filters; results sorted by relevance.
A pre-registered study on arXiv (2607.17420v1) found that biographical personas in system prompts produce model-dependent effects on LLM code generation, with the research-librarian persona causing Claude Opus to issue i…
Moonshot AI released Kimi K2.7 Code on June 12, 2026, scoring 75.92% standard and 72.58% strict on code benchmarks. The model is now available via official access, with no benchmark scores or vibe ratings published yet.
AI adoption among software developers has reached 90%, outpacing traditional code review processes and rendering pre-merge human approval obsolete. The growing gap between AI-generated code and manual review demands a sh…
A developer at a company building create-microservices-app deliberately broke their own AI-built booking app to test whether automated contract checks catch plausible-but-wrong code. The experiment showed that a machine-…
Evan launched Ito.ai, a code review tool that runs code in a full environment to find bugs with low false positives, providing screenshots, videos, and run logs. The tool creates a devcontainer, wires up services, and us…
A developer built CodeClipr, a VS Code extension that packages an entire codebase into a single AI-ready context payload with one click, eliminating the need to copy files individually into AI assistants like Claude or C…
Dashboards as code stores a dashboard as a YAML or TSX file in a git repository that is reviewed in a pull request and deployed on merge, rather than as an object clicked together inside a BI tool, according to Bruin, wh…
GitHub has switched Copilot from request-based to token-based billing, metering usage through AI Credits, while Claude Code retains a flat-rate model with throttling. A developer's analysis of published rate sheets shows…
Microsoft's VS Code 1.131, released on July 29, introduces subagent visibility in the Agents window, built-in offline dictation powered by Microsoft's Nemotron model, and Bring Your Own Key (BYOK) support for the Agents …
Anthropic has removed a hidden monitoring feature from its Claude Code tool that secretly checked users' connections to China and AI labs there, transmitting data through encrypted steganography in the system prompt. The…
A two-skill Claude Code plugin designed to analyze and modernize legacy Java 8 codebases to Java 21. The first skill, `/java-modernization-review`, analyzes a Java file and generates a scored report (1–100) across nine d…
The article compares Muse Code, Claude Code, and Codex CLI across seven architecture differences, evaluating their design trade-offs for developers. No specific figures or dates are provided, and the source is a technica…
Meta Platforms Inc. is developing Muse Code, a coding assistant tool similar to Claude Code, according to a report. The tool aims to compete in the AI-powered developer tools market, though specific features and release …
AI has not shifted the bottleneck in software development from coding to code review, according to The New Stack. The article argues that while AI tools assist with code generation, human review remains the critical cons…
The open source game engine Godot has banned AI-authored code contributions, citing an inability to trust heavy AI users to understand and fix their code. The decision reflects growing concerns over AI-generated code qua…
Z.ai disabled the repository-snapshot upload workflow in its ZCode coding assistant and shipped the fix in ZCode v3.14.0 after an independent Chinese blogger, Ferstar, found that a default-enabled setting packaged users'…
A developer trimmed a bloated VS Code extension setup from over 40 down to 10 by disabling everything and re-enabling tools only when actually needed, then published the resulting list. The stated selection criterion is …
A new closed-loop framework for AI coding agents, deployed across a microservices platform with over 35 services, embeds human feedback as behavioral rules to enable persistent improvement without traditional model updat…
A developer building a live automated trading system on Claude Code found that token waste comes not from the model but from bloated context. By implementing six patterns—including a compact memory file, semantic search …
A developer reports that GitHub Copilot has been catching significant issues in code generated by GPT-5.6-Sol-High, suggesting that developers should run multiple AI review cycles to improve code quality.