Full-text search across 69339 articles. Combine with topic and date filters; results sorted by relevance.
A cost-based semantic routing demo combining vLLM Semantic Router (vSR) with agentgateway reduced LLM spend by 35.4% on a 50-prompt Go and Rust coding dataset, according to a developer evaluation published July 16, 2026.…
On 2026-07-20, Simon Willison reported that Claude Code is now running on Bun, a Rust-based JavaScript runtime, while Alibaba's Qwen 3.8 reached number two on Hacker News with 737 points, signaling a shift in AI competit…
Seven years of LLM releases from GPT-2 (1.5B parameters, February 2019) to Kimi K3 (2.8T parameters, July 2026) show that the decoder-only transformer architecture barely changed while almost everything around it was rep…
A developer ran a cross-vendor audit on ten queued blog posts, using a different model (Gemini 3.1 Pro High) to review content written and previously reviewed by a single model family. The audit found seven issues, six o…
Moonshot AI, the Beijing-based developer of the Kimi series of AI models, is targeting an IPO filing in Hong Kong in the second half of 2026 after its valuation surged from $4 billion to between $20 billion and $30 billi…
AI-forward engineering teams face a growing security risk as coding agents with skipped permission checks run next to production-adjacent credentials on laptops, exposing the entire product to potential compromise, accor…
The Salamanca City Central School District in New York will deploy a humanoid robot named Sally, created by Realbotix, to teach high school AI, coding and robotics courses this fall at a cost of more than $50,000. Superi…
A developer argues that accepting AI-assisted code contributions based solely on output quality misses critical issues like maintenance burden, shifting quality definitions, and long-term ownership. The bar for review sh…
A developer running a fleet of Claude Code agents discovered that an agent broke prompt submission by replacing the Enter key with backslash-Enter, which created a newline instead of submitting. The fix was locally sound…
A developer explains that the first prompt in a new repo is critical for agentic coding, as it sets the trajectory for all subsequent prompts. They demonstrate how AI config kits like CLAUDE.md and tested prompts transfo…
A developer built AIpine, an iPhone and iPad app that serves as a local file library for AI-generated content. The app supports formats like JSX, HTML, Mermaid, Markdown, CSV, JSON, and Python, offering both preview and …
PDNob Pro PDF Editor, an AI-powered tool for editing, converting, summarizing, and translating PDFs, is offering a lifetime subscription for $34.99 (MSRP $349.95) with code SAVE15 through July 26. The deal includes 20GB …
Retriever has launched an in-browser agentic dataset enrichment tool that lets users bulk-enrich any webpage, such as a conference attendee list, with LinkedIn profiles and work emails, then score and contact the top lea…
Developer Greg Corriston built revüe, an MIT-licensed Agent Skill for Claude Code, Codex, and Cowork that enforces a deterministic review workflow before shipping AI-generated work. The tool combines local validators, sc…
A developer applied a 1913 inventory formula to determine the optimal time to restart LLM coding sessions, finding that context behaves like inventory. By instrumenting ~150 real sessions across 1,016 transcripts, they d…
Google is renaming its AI-powered research assistant NotebookLM to Gemini Notebook, folding it into the Gemini portfolio, and expanding cloud-based computing access to Google AI Pro subscribers. The rebrand preserves all…
A new MIT Press book, 'Inventing ELIZA,' provides the first comprehensive critical analysis of Joseph Weizenbaum's pioneering chatbot through rediscovered original source code and previously unseen scripts, revealing a f…
Nano Collective, a developer community, has released a suite of open-source, local-first AI tools including Nanocoder (a coding agent), Nanotune (for fine-tuning LLMs locally), and packages like prompt-scrub and get-md, …
Anthropic's automated chatbot dismissed a security vulnerability report from Wiz researchers, claiming the issue fell outside Claude Code's threat model, even though Anthropic had already patched the flaw nine days earli…
An unnamed U.S. company spent $500 million on Anthropic's Claude in a single month because no spending limit was set, according to an AI consultant cited by Axios. Uber burned through its entire 2026 AI budget by April a…