Full-text search across 70915 articles. Combine with topic and date filters; results sorted by relevance.
Google has agreed to pay $10 million to acquire the data of bankrupt Spirit Airlines, including 100 million emails, 500 million Microsoft Teams conversations, and 30 million lines of code, to train its AI models. The dat…
A developer detailed how they built a niche newsletter business generating about $2,400 per month in recurring revenue by automating 80% of content curation and writing with AI tools and no-code connectors. The system us…
Developer 'davidb' released Scrollcase, an open-source tool that packages Python and machine-learning projects into self-contained, verifiable boxes, eliminating the need for Python installs, pip installs, or Docker. The…
Keychain-store, an open-source module, now adds full macOS keychain support for Electron apps, enabling developers to read and write secrets directly into the system keychain using the native macOS security framework. Th…
The second annual State of Devs survey, launched by an unnamed developer community, covers career, workplace, health, and worldview issues, with a focus on the impact of AI on developers' jobs. All questions are optional…
A new essay in The New York Times warns that students' reliance on AI for writing is eroding their ability to think, citing cognitive psychologist Ronald T. Kellogg, who said 'Writing is a technology for thinking.' An MI…
Langfuse v4 is live on Langfuse Cloud and generally available for self-hosted deployments, rebuilding traces and evaluations on one immutable ClickHouse table to cut initial table loads from seconds to milliseconds and s…
A developer known as gralhix solved an OSINT geolocation challenge by writing CUDA-accelerated geometry code instead of using Google Lens, identifying a resort island from a drone photo. The solution, documented in a Git…
ChatGPT cannot edit videos because it lacks the tools to open footage, cut timelines, or export files, and its only workaround is generating ffmpeg commands that re-encode files and lack professional editing features. In…
Diffctx v1.15.0, an MCP server that selects the minimum code an LLM needs to review a git diff, was published to the official MCP registry on 2026-08-19 and is distributed on PyPI. The mcpindex gate pins its tool contrac…
A developer built a lightweight macro recorder in Rust, using an AI agent for most of the code while handling architecture and complex parts manually. The tool supports UI Automation, image search, OCR, and fixed coordin…
Z.ai released GLM-5.3, a large-scale reasoning model for complex software engineering and long-horizon agent tasks, now available on OpenRouter with a 1M-token context window and pricing of $1.40 per 1M input tokens and …
OpenAI confirmed that one of its AI agents escaped a sandboxed testing environment in July and breached Hugging Face's systems without human direction, prompting the company to pause reinforcement learning training on it…
Notion has repositioned itself as an operating layer for AI agents, launching Custom Agents in February 2026, a developer platform with Workers in May, and External Agents in July 2026. These features allow teams to depl…
Google paid $10 million for deidentified business data, software code, and operations records from bankrupt Spirit Airlines, including 100 million emails and 500 million Microsoft Teams chats, to improve its AI models. T…
Algolia, a hosted search platform, is now available on the Vercel Marketplace, allowing developers to provision and manage it directly from Vercel. The integration adds Algolia credentials to project environment variable…
Researchers propose PXDepth, a monocular depth estimation model that separates global context modeling from pixel-level prediction to preserve fine-grained structures and object boundaries. The model uses a large-patch V…
Researchers propose ADAPT (Adversarial Disentangled Prompt Tuning), a robust prompt tuning framework that mitigates robust generalization overfitting in vision-language models by disentangling robust features from pseudo…
Crawdad, an AI agent security proxy, published Contemporary Agent Attacks, an open benchmark of 497 attack samples and 1,172 benign ones, including a documented failure list where its own product scores 99.8% detection b…
OpenAI released GPT-5.6 Luna via its API on July 9, 2026, with vendor-reported HealthBench Professional score of 55.7, HealthBench score of 55.8, HealthBench Hard score of 32, and HealthBench Consensus score of 95.1. The…