Full-text search across 59572 articles. Combine with topic and date filters; results sorted by relevance.
Cursor released Cursor Router, a request-level classifier that dispatches coding requests to the most suitable model, achieving frontier-quality performance at 30–50% lower cost for early-access enterprise accounts and 6…
Cursor launched an automated model router on July 22nd that selects a model for each coding request, enabling enterprise customers to reduce inference costs without developer intervention. Cursor claims its Intelligence …
Anthropic's official Claude Code skills reveal a gap between documented rules and actual practice. The developer analyzed all 17 official skills and found that the 500-line limit for SKILL.md is a guideline, not a law—th…
A developer argues that the future of AI coding lies not in better prompts but in engineering constraints that enforce best practices. They introduce the Guardian Engineering Framework (GEF), a governance layer that gene…
A developer has created tpu-management, a Claude Code skill and MCP server that automates the full lifecycle of serving Gemma 4 on Google Cloud TPUs, from zone selection and VM creation to benchmarking and teardown. The …
Postgres 19, which reached beta on June 4, adds six features that coding agents without GitHits fail to identify, according to a test by GitHits. The test showed that an agent without GitHits incorrectly stated that core…
A developer created a self-destructing launchd job that automatically patches Claude Code's settings.json to migrate from the deprecated Fable 5 model to Opus on the model's end-of-life date (2026-07-07). The job uses a …
VS Code 1.128, released July 8, introduces multi-chat agent sessions that allow parallel conversations with separate models and histories within a single workspace, and makes Copilot Vision generally available for all su…
Moonshot AI released Kimi K3, a 2.8 trillion parameter open-weight model, on July 16, and within a day it topped the Frontend Code Arena leaderboard with a score of 1679, beating Anthropic's Claude Fable 5. The Beijing-b…
China's Moonshot AI released Kimi K3 on July 16, 2026, and by the next day the model topped LMArena's Frontend Code Arena with a score of 1,679 Elo, beating Anthropic's Claude Fable 5 (1,631) and OpenAI's GPT-5.6 Sol (1,…
A developer built Relay, an open-source Go daemon that implements a signed handoff protocol for AI coding agents, enabling seamless context transfer when an agent hits usage limits. The tool detects quota breaches, reque…
A developer published Part 8 of a 10-part React Native Interview Handbook series, featuring 70 code-output challenges that test JavaScript fundamentals such as closures, scoping, array methods, and type coercion. Each ch…
Moonshot AI's Kimi K3, a 2.8 trillion parameter open-weight model, topped Arena.ai's Frontend Code leaderboard with 1,679 points, beating Claude Opus 4.8 and other frontier models, and is priced at $3 per million input t…
A developer outlines three approaches to programmatically summarize PDFs in 2026, from a custom map-reduce pipeline using PyPDF and OpenAI to a LangChain wrapper and a free no-code option like ChatPDF. The post emphasize…
Quickchat AI published a no-code guide to make AI agent actions reliable, using server-side gates and memory capture to prevent agents from calling APIs incorrectly or losing values between steps. The guide demonstrates …
Anaconda has acquired Kilo Code, an AI development platform used by over 3 million developers that orchestrates nearly 10 trillion tokens per month, to give enterprises visibility and control over AI usage and data. Anac…
OpenAI published a research note on July 8, 2026, detailing reliability concerns in SWE-Bench Pro, emphasizing that coding benchmark scores are only meaningful when task design, grading, and potential training data overl…
The San Francisco Board of Supervisors voted 7-4 to pass a 362-page ordinance making sweeping changes to the city's municipal codes, despite objections that the measure is too broad and concerns over reducing surveillanc…
Wix CEO Avishai Abrahami said the tech industry gives AI 'too much credit' and that it is not smart enough to replace white-collar jobs, despite his company's $200 million investment in the vibe-coding platform Base44. A…
A developer building agentproto argues that coding agents relying solely on frontier model weights produce generic answers because they know only the internet's average. The key to making an agent truly useful for a spec…