Package Managers Need to Cool Down
Prompted by a recent supply chain attack, the concept of dependency cooldowns for package managers is gaining renewed attention. This practice suggest...
Simon Willison
How Moda Builds Production-Grade AI Design Agents with Deep Agents
Moda is successfully implementing a multi-agent system, built upon Deep Agents and monitored via LangSmith, to enable non-designers to create and iter...
LangChain
Helping developers build safer AI experiences for teens
OpenAI has introduced new prompt-based teen safety policies for developers utilizing its gpt-oss-safeguard system. This initiative aims to assist in m...
OpenAI
SERHANT. Implements Rapid AI Iteration Playbook on Vercel
SERHANT. has adopted a strategic playbook for rapid AI iteration, leveraging Next.js on Vercel to seamlessly expand its platform to a React Native iOS...
Vercel
Vercel Acquires new.website to Accelerate AI Software Development
Vercel has acquired new.website as part of its strategy to empower users in shipping complete, production-ready software with AI. The new.website team...
Vercel
GitHub Boosts Application Security with AI-Powered Detections
GitHub is expanding its application security coverage by integrating AI-powered detections with its existing CodeQL technology. This enhancement, part...
GitHub Blog
Datasette-files 0.1a2 Released with File Upload Capabilities
The latest alpha version, datasette-files 0.1a2, has been released as a new plugin for Datasette, enabling direct file uploads into a Datasette instan...
Simon Willison
Discussion on ‚AI Slop‘ and the Value of Human Effort
Simon Willison highlights a critical perspective on 'AI slop,' defining it as content that demands more human effort to consume than it took for an AI...
Simon Willison
LangSmith Fleet Enhances Agent Authorization with Two New Types
LangSmith Fleet, a LangChain offering, has introduced two distinct methods for agent authorization: 'Assistants' and 'Claws.' Assistants utilize the e...
LangChain
Exploring ‚Streaming Experts‘ for Efficient MoE Model Execution
Simon Willison discusses the innovative 'streaming experts' technique, which allows larger Mixture-of-Experts (MoE) models to run on hardware with ins...
Simon Willison
Hugging Face Introduces EVA Framework for Voice Agent Evaluation
Hugging Face, in collaboration with ServiceNow AI, has launched EVA (Evaluating Voice Agents), a new framework designed to assess the performance of v...
Hugging Face
LangChain to Showcase Agent Development at Google Cloud Next 2026
LangChain invites attendees of Google Cloud Next 2026 in Las Vegas to visit their booth, specifically targeting those engaged in agent development. Th...
LangChain