Full-text search across 65571 articles. Combine with topic and date filters; results sorted by relevance.
A developer known as Riven Desk has published a 10-minute checklist for reviewing AI-generated pull requests, emphasizing verification over trusting polished summaries. The approach includes writing a one-sentence purpos…
A developer shares a practical approach to writing coding standards for AI-assisted development, recommending that rule files start with state management. The post provides concrete rules, such as keeping state local unl…
A developer argues that free coding models and remote servers do not eliminate the need for specifications and tests. The post warns that free compute does not purchase a behavioral contract and that retrying agents unti…
Researchers introduced ττ-bench, a benchmark that tests AI systems' ability to construct customer-service agents from realistic business records, client requirements, and production APIs. Across 53 tasks in four domains,…
DietrichGebert released Ponytail, a ruleset for AI coding agents that enforces writing the least code that works, claiming benchmark medians across 12 feature tasks on a FastAPI and React repo. The tool, available for 14…
Anthropic said its Claude agents produced the first complete computer-checked formalization of Fermat's Last Theorem in 11 days, generating 13 million lines of Lean code and about 29,500 intermediate theorems used in the…
The next generation of the free and open source OpenHelper AI Chatbot, v2 Preview, is expected to launch before November 2026 in 64+ countries, featuring improved response quality, reduced failure rates from over 60% to …
A developer has launched Profound Academy, an agentic AI course builder that creates hands-on course materials, interactive tutorials, exercises, and automatically checks student submissions. The tool can generate module…
A developer built Dokpod.io, an AI knowledge vault that uses the Model Context Protocol (MCP) to let AI coding assistants like Cursor and Claude Code query indexed project assets on demand, eliminating the need to re-upl…
FrontierHarness Eval, a new benchmark from Runta, tested nine AI coding harnesses on the same model and found median cost per successful task varies by 17x, with Claude Code v2.1.237 passing 19 tasks but reaching $18.34 …
Help Net Security has released a guide titled 'The Agentic Software Development Guide,' which addresses the trust and accountability challenges that arise when AI accelerates code production. The guide warns that teams o…
Anthropic launched Claude Fable 5.1 and Claude Mythos 5.1, claiming they are the world's most advanced models for coding and knowledge work, with Fable 5.1 available to everyone and Mythos 5.1 limited to U.S. trusted acc…
A developer from Port Harcourt argues that local AI is not dead, citing his own experience where a Windows laptop with integrated graphics successfully ran a 7B coding model for an offline AI coding assistant, while his …
A personal blog roundup for August 2026 highlights the decline of RSS adoption, attributing it to Google's actions, and features commentary on AI coding tools, including a critique of Meta-owned personal superintelligenc…
Vercel's AI SDK harness layer now supports fx, its lightweight open-source coding agent, via the official @ai-sdk/harness-fx adapter, which connects over the Agent Client Protocol using @ai-sdk/harness-acp. fx joins Clau…
A 12-year-old developer has launched DOJO FEED, a social network exclusively for developers aged 5–18, blocking adults at signup with a humorous message. Built with Supabase and Netlify on a POCO C55 phone, the platform …
Academa, founded by PhD students Sina Atalay and Abdullah Geduk, launched a platform that generates long-form STEM lecture videos using LLMs, treating videos as code that can be edited and corrected. Each video includes …
Ration, a native macOS menu-bar app, tracks Claude Code usage across terminals, showing a usage ring, project and branch breakdowns, a 7-day trend, and a shareable recap card with an API-equivalent cost estimate. The app…
Reflexio, a new learning platform for AI agents, claims to reduce task failure by 36% while saving more than 50% on tokens by turning user corrections and successful outcomes into reusable behavior changes. The platform …
A developer's test in Pi, an open-source AI coding tool, found that using subagents doubled token usage and time for the same task: context usage rose 104.7% from 63k to 129k tokens, and elapsed time rose 112.2% from 3m …