An engineer from TrothByte warns that AI assistants generate assembly code that is confidently wrong in specific, repeatable ways, such as inventing nonexistent mnemonics, flipping operand order, and …
AI development is a rolling sequence of bubbles rather than a single one, according to an analysis that compares the current cycle to the dot-com era. The piece argues that each hype wave—from LLM nov…
Anthropic reported degraded performance affecting claude.ai, platform.claude.com, the Claude API, Claude Code, and Claude Cowork, according to a status update. The outage impacts users of Anthropic's …
Russell Jones's blog now runs a fully automated GitHub Actions workflow, Content Autopilot, that mines project activity, uses Claude Code headless in CI to curate and write Hugo posts, and publishes t…
Multiple developers on Hacker News report abandoning Anthropic's Claude Code for OpenAI's Codex or other providers after experiencing what they describe as a quality decline over the last 2-3 months, …
An engineer measured the effectiveness of a 1,350-line CLAUDE.md file for AI coding agents, running 30 tests across six configurations. The results showed that the model followed the rules 399 out of …
GitKraken released GitLens 19 on August 14, transforming the Commit Graph from a read-only viewer into an interactive workbench with AI-powered features, including AI agent monitoring for Claude Code,…
Mads Thines, a Copenhagen-based designer and product engineer, launched LoreKit, an open-source memory layer that gives AI coding agents a record of earlier mistakes, fixes, and project-specific lesso…
A Hacker News user asked for recommendations on LLM subscriptions or providers to use with the pi harness, noting dissatisfaction with Claude Code over the past 2-3 months and satisfaction with Codex.…
A developer's hands-on comparison found that switching from a standard chat-based AI workflow to a native agentic workflow reduced a legacy Express.js middleware migration from four hours to 12 minute…
Pi, a minimal terminal coding harness, achieved a 66.7% pass rate (20 of 30 tasks) in Composio's benchmark using DeepSeek V4 Flash, the best among four newly tested harnesses, and was the cheapest at …
Andrew Marble of willows.ai introduced three tasks to evaluate AI coding agents as products, measuring efficiency, collaboration, and taste in interactive sessions, and tested five agents including th…
A developer rebuilt his puzzle bulletin board as mostly static HTML to withstand 3 million monthly bot visits, using Claude Code to migrate content from phpBB into a database and regenerate pages week…
Joel Dare, a software developer, announced he is building a minimal 'Dark Software Factory'—an AI-driven software development system that writes code, creates tests, and deploys without developer over…
Anthropic's Claude Code CLI fails on Windows when given multi-line prompts because the npm-installed claude.CMD batch shim truncates everything after the first newline, causing the agent to receive in…
Glider, a new terminal tool, lets users drive multiple AI coding CLIs—Claude Code, Cursor Agent, and Antigravity—from one terminal, relaying permission prompts and maintaining project memory. It suppo…
Unsloth, the open-source fine-tuning library, released Unsloth Desktop, a free beta app for macOS, Windows, and Linux that runs and trains LLMs, diffusion models, and audio models locally, directly co…
A developer discovered that on Windows, a batch-file shim for Claude Code truncated multi-line prompts at the first newline, causing an agent harness to score 3/24 on a web-task eval. By pointing dire…
A developer rebuilt PolyForge, a standalone Python CAD tool, from a ChatGPT skill into an offline package that renders parametric part templates to OpenSCAD, FreeCAD, or Blender without requiring an L…
The TeamPCP campaign in March 2026 used a stolen token to cascade across five ecosystems in eight days, publishing backdoored LiteLLM versions 1.82.7 and 1.82.8 on PyPI that swept LLM API keys, cloud …