Full-text search across 59179 articles. Combine with topic and date filters; results sorted by relevance.
A 2026 journal study found that trillion-parameter large language models (LLMs) such as GPT-4, LLaMA, and Falcon are capable of literate programming, while smaller models excel at recognizing programming languages rather…
Thedotmack has released claude-mem, an open-source memory compression and retrieval engine that gives AI coding assistants persistent long-term memory across terminal sessions. The tool passively monitors developer inter…
A developer argues that AI coding agents should be auditable through execution evidence rather than trusted based on their reports. The post distinguishes between requests and results, proposes a three-rung evidence hier…
A developer has created a take-home test to evaluate AI code reviewers' ability to ignore stale repository context. The test uses a small Flask app with a misleading comment and runs the same prompt under three context r…
A developer built a 169-line Node.js verification script to automatically catch bugs in AI-generated code, addressing the impracticality of manually reviewing every line. The script checks for dead links, missing alt att…
Developer mkaz built Halftrack, a personal trainer app powered by the Pi coding harness, which uses natural language commands to log and query running workouts. The app, available on GitHub, demonstrates Pi's extensibili…
A developer built a scoring script to test the consistency and correctness of free AI code review models on MonkeyCode's free server. The experiment, which ran three prompts three times each on a small Node.js library, f…
A developer has proposed replacing vector-based RAG for coding agents with a Git-backed markdown system, arguing that topic-scoped .md files accessed via MCP provide deterministic context persistence without the overhead…
Researchers found 120 unclaimed package or domain references across 8,265 llms.txt or llms-full.txt files on 6,214 live domains, including sites of defense contractors, Fortune 500 companies, and major tech firms, which …
Microsoft released Agent Lightning v1.0, an MIT-licensed open-source reinforcement learning framework that trains AI agents without modifying agent code, available on GitHub. Using the framework, a Qwen3.5-9B model impro…
A developer built TreasuryForge, an autonomous treasury management agent for the WeMakeDevs × TrueFoundry Agent Harness Hackathon, using TrueForge's agent harness. The agent manages a simulated portfolio across cash, cry…
See-Sol-Lab released DeepSeekGUI v1.0.0, a Windows desktop client that packages DeepSeek's Harness coding agent into a native app with an installer, system tray, built-in browser, and terminal. The tool requires only an …
A week of Hacker News discussions reveals growing developer frustration with AI coding tools, particularly Cursor and GitHub Copilot. Complaints range from Cursor's intrusive UI and model-switching to Copilot's billing q…
A developer explains how to run parallel Claude Code sessions on a single repository using Git worktrees, which allow each session to have its own directory and branch to avoid file conflicts. The approach involves using…
A creator built a working Calendly clone using an AI coding agent on a flat-rate subscription plan, with roughly 5 days and 5 hours of cumulative agent runtime driven by about five prompts, making the marginal compute co…
A developer has built Pixy, a tool that adds a visual editing layer to websites, allowing users to make design changes directly on the page and pass those changes to coding agents. The project aims to streamline communic…
RemObjects released build 71 of CodeBot, adding AI code completion for Delphi 12 and newer, improved network handling, and a desktop avatar. The update also allows creating projects from scratch and smoother background u…
Production AI systems require continuous accountability infrastructure rather than a one-time deployment, according to an analysis of the gap left by 'vibe coding' practices. The piece argues that data pipelines, evaluat…
A physician built a 1,386-card Anki deck for cardiac electrophysiology board review using Claude Code, Python, genanki, and pdftotext in about a day, and discovered that 16 of the first 52 cards were broken because Anki'…
Safer-dependencies, a new security layer for Claude Code, automatically audits and blocks risky dependencies—including CVEs, typosquats, abandoned packages, and version-age issues—across npm, PyPI, RubyGems, Maven, Go, R…