Full-text search across 68348 articles. Combine with topic and date filters; results sorted by relevance.
A developer published a seven-step guide for protecting production databases from AI coding agents, covering controlled backups, restore verification, a separate development database, and settings that keep agents away f…
Moonshot AI's Kimi K3.1 Pro and a faster Kimi K3.1 Pro Highspeed variant are now available on Vercel's AI Gateway, callable via the AI SDK, Chat Completions, Responses and Anthropic Messages APIs under the model name moo…
LangChain's State of Agent Engineering survey of 1,340 practitioners found that 89 percent have observability on their agents but only 52.4 percent run offline evaluations against a test set, with a third naming quality …
A developer built Claude Task Master (CLI `claudetm`), an open tool on the Claude Agent SDK that autonomously runs the full pull-request lifecycle: it reads a high-level goal, writes a plan with success criteria, makes c…
Amazon Web Services made training sessions in Amazon SageMaker AI serverless model customization generally available, letting customers write and coordinate their own post-training loop — including custom loss functions …
Recursive SI released Flint, a 40-billion-parameter mixture-of-experts coding model, on October 21 under an Apache 2.0 license with no revenue gate, no OEM clause and no service-operator trigger, publishing weights on Hu…
A developer published a step-by-step n8n workflow that polls an RSS feed every 30 minutes, summarizes new items with an Anthropic Claude model via the Basic LLM Chain node, validates the model's JSON output in a Code nod…
A developer built TidySloth CLI, a local-first Python command-line tool that scans a target directory, sorts files into Code, Documents, Images, Archives and Others subfolders, and generates a Markdown manifest (TIDY_IND…
A benchmark asking models to write a Zstd compressor and decompressor pair in Rust from RFC 8878 found all tested Anthropic models functionally capable, with only 4 of 81 runs failing on edge cases, and that property-bas…
A TCL "Smart TV Pro" running Google TV was stripped of manufacturer ads and telemetry in about two hours using the open-source CleanYourTV package list for the TCL G03 board, driven by Claude Code, which disabled 25 pack…
Arm introduced Arm AGI CPU, a new class of Arm Neoverse-based processors built for agentic AI workloads, and at OCP Global Summit unveiled a new CPU evaluation methodology backed by Fujitsu, Meta and OpenAI. Arm said ind…
OpenAI's "Patch the Planet" initiative, working with security firms Wiz and Calif, uncovered two critical PHP memory-corruption vulnerabilities — CVE-2026-109618 and CVE-2026-109619 — rated 9.2 out of 10 on the CVSS v4.0…
Security researchers at Calif discovered two PHP memory-management flaws, dubbed PHP2Shell, that potentially affect 70% of the world's computer systems and are rated 9.2 out of 10 on the CVSS v4.0 severity scale, accordi…
PHP maintainers announced fixes roughly 12 hours out for two remotely triggerable PHP runtime vulnerabilities, CVE-2026-109618 and CVE-2026-109619, both rated 9.2 out of 10 and dubbed php2shell, disclosed by Calif.io, Wi…
A developer built GrammarBar, a free, open-source macOS menu bar app that fixes, polishes and translates text in place without leaving the current application. The app shells out to Claude Code's headless mode (defaultin…
A developer outlined a practical approach to building safe, auditable agentic AI systems on the JVM using Spring Boot and Spring AI, where a language model plans steps, calls tools, observes results, and iterates toward …
Vũ Xuân Toàn, a developer and indie maker from Bắc Ninh, Vietnam, built a living system map for Mầm Văn, his React/TypeScript/Supabase/Gemini AI literature-learning platform for Vietnamese middle-school students. The map…
Anthropic made fast mode generally available for Claude Opus 5.5 on October 15, replacing the waitlist and the `anthropic-beta: fast-mode-2026-02-01` header with a single `"speed": "fast"` payload field that delivers up …
MiniMax launched M3.1 Flash on Monday, saying agent tasks cost 70% less to run than on its previous, unnamed model and that inference runs nearly three times faster, at a flat $0.15 per million input tokens and $0.60 per…
Google announced at PyTorch Conference 2026 that TorchTPU, its native PyTorch stack for Google TPUs, is now open source on GitHub, making TPUs a native option for PyTorch alongside GPUs and CPUs. Co-developed with Meta's…