Full-text search across 71115 articles. Combine with topic and date filters; results sorted by relevance.
Cloudflare has open-sourced Cloudflare OS, an AI productivity environment that provides an agent chat UI, sandboxed application development, and a security framework called Gatekeepers, allowing companies to customize it…
PyTorch, an open-source deep learning framework built in Python, has become the prevailing choice across research and industry with a 63% adoption rate in model training and use in over 70% of AI research implementations…
Ruby on Rails is a strong fit for AI agents because of its token efficiency, predictability, and ecosystem ergonomics, according to an analysis by Martin Alderson, co-founder of CatchMetrics. Alderson found up to a 2.6x …
OpenAI and Anthropic have been privately lobbying Washington regulators to restrict open-weight AI models from China, even as OpenAI CEO Sam Altman publicly supports open-source AI, according to The New York Times. The p…
Codacy warns that AI-generated pull requests create a governance blind spot where malicious intent can emerge across multiple individually acceptable changes, requiring engineering leaders to adopt longitudinal context a…
Chinese AI models from DeepSeek, MiniMax, Tencent, and Xiaomi have seized OpenRouter's top five positions, offering 4–8x cost advantages over comparable US models and triggering real production migrations. Microsoft at B…
OpenAI confirmed GPT-5.6 Sol, Terra, and Luna will launch Thursday, and Claude Fable 5 returned July 1 after a government shutdown. At the AI Engineer World's Fair in San Francisco, over 6,000 attendees focused on agent …
AI agent infrastructure is maturing but facing reliability challenges, as FlashAttention-4 achieves 71% utilization on NVIDIA's Blackwell B200 GPUs while agentic AI systems grapple with architecture questions and securit…
Olin Shivers recounts the history of T, a Lisp implementation developed at Yale in the early 1980s by Jonathan Rees, Dan Weld, and himself, which set a standard for clean design. The project emerged from a small group of…
A MIRI Technical Governance Fellowship project evaluates Trusted Execution Environments (TEEs) for privacy-preserving monitoring in AI governance, finding that while TEEs offer verifiable constraints and user privacy, ha…
Gray Swan co-founders Zico Kolter and Matt Fredrikson discussed the state of AI red teaming, warning that prompt injection and agent vulnerabilities create a new class of security risks that traditional cybersecurity can…
Microsoft released TypeScript 7.0 RC with a Go-rewritten compiler delivering ~10x faster builds, while the Model Context Protocol stabilized its Enterprise-Managed Authorization extension for AI agent connectivity. The E…
Alibaba's 1.7B parameter Qwen3-TTS voice cloning model was fine-tuned using Fully Sharded Data Parallel (FSDP) with PyTorch and Ray, demonstrating memory-efficient distributed training across 4 GPUs. The implementation a…
Microsoft unveiled Majorana 2, its next-generation quantum chip, at the Microsoft Build Live keynote, achieving qubits that are 1,000 times more reliable than the previous generation with a mean lifetime of 20 seconds. T…
The article distinguishes between two distinct security challenges for AI in banking: governing employee use of tools like ChatGPT for daily tasks, and architecting secure production AI agents that can access internal sy…
Engineers who have not yet adopted AI agents for production work face growing pressure as major tech companies restructure around the technology. Shopify, Fiverr, and Block have each cut thousands of jobs while mandating…
This article presents a tutorial on a pure functional C++ library for Software Transactional Memory (STM), which aims to simplify parallel and concurrent programming by allowing developers to build and evaluate concurren…
A benchmark report compares two locally hosted models, qwen3-coder-next (80B total / 3B active MoE) and deepseek-r1:32b (32.8B dense reasoning model), on princeton-nlp/SWE-bench_Verified against published frontier-model …
Ultramarine 1 Flash, an agentic model built by applying large-scale reinforcement learning to the open base model GLM-5.3 Flash, improves on that base across all six knowledge-work benchmarks the company evaluated, by 4.…
Agentic policies that write and iterate on their own Python skills reached a 34% to 65% overall task success rate across ten bimanual tasks, beating end-to-end policies trained on 30 times more demonstrations on 8 of 10 …