Developer Forrest Chang turned four observations from AI researcher Andrej Karpathy about how large language models fail at coding into a CLAUDE.md file that Claude Code reads automatically at the sta…
Developer nash_su released LLM Wiki v0.6.11, a GPLv3 open-source desktop application for Windows, macOS and Linux that uses an LLM to build a persistent, structured wiki from ingested documents. The 4…
A developer built SelfContext, a local context layer for AI tools, to stop re-explaining personal history, goals, and preferences in every new AI conversation. The project stores context as ordinary M…
The Tokenizer newsletter #36 reports that a single C file can stream a 744-billion-parameter model from disk, Cloudflare has mapped agentic traffic patterns, and Andrej Karpathy's nanochat trains a GP…
Andrej Karpathy's spec-driven method for AI coding reframes work with models like Claude as a three-layer system: a spec, a verifier, and a persistent environment. The spec layer has Claude interview …
AI startup Inception launched Mercury 2.5, a diffusion-based large language model that delivers a 40% intelligence improvement over its predecessor while processing 1,107 tokens per second on standard…
A developer argues that as AI agents grow more capable, nearly everything — from corporate systems to consumer devices — will need to become "agent-addressable" through code, creating a feedback loop …
An engineer at a financial technology company describes how LLM agents have automated nearly the entire software development lifecycle, from triaging errors and writing code to QA and deployment, leav…
Andrej Karpathy's Autograd, an automatic differentiation engine, forms the foundation of deep learning by calculating derivatives to adjust neural network parameters during training. The engine enable…
Knut Melvær, Principal Developer Marketing Manager, argues that organizations should focus on who owns and maintains agentic context rather than on RAG pipelines, citing Andrej Karpathy's April 2026 p…
Moltbook, a social network for AI agents built entirely by AI, failed within a week due to security flaws and lack of moderation, according to Matthew Schlicht, its creator, and security researchers. …
AutoGPT, an open-source project with 185,000+ GitHub stars, has launched a hosted Platform that lets users build, deploy, and run AI agents that execute complete workflows, with a free self-hosted opt…
Andrej Karpathy's Neural Networks: Zero to Hero, the Full Stack LLM Bootcamp by Full Stack Deep Learning, and Stanford's CS336 are among five free courses curated to take learners from LLM beginner to…
A developer built a wiki-style LLM that maintains structured, cross-linked notes from documents to avoid re-deriving answers from scratch, testing it on The Hound of the Baskervilles. The approach, in…
Neo4j AI Research Engineer Michael Hunger argues that Andrej Karpathy's LLM-maintained wiki recipe, which gained viral attention and inspired Google Cloud's Open Knowledge Format (OKF) in June 2026, c…
A developer reports that prompt engineering is being replaced by harness engineering, as 40% of AI agent projects fail in production. The shift from optimizing single prompts to managing entire contex…
OpenAI CEO Sam Altman said in an interview with Stratechery that he no longer thinks of the harness and the model as entirely separable things, signaling a shift in AI development toward 'harness engi…
The 14th edition of the Dev Opportunity Radar newsletter highlights upcoming hackathons and resources for developers, including the Razorpay AI Buildathon, the AWS Agents for Humans Hackathon, and MLH…
Jerry Tworek, former OpenAI VP and current CEO of Core Automation, estimates human researchers have at least two years left as a meaningful part of AI research discovery, citing the exhausting pace an…
Google Cloud's Data Cloud team published the Open Knowledge Format (OKF) v0.1, a file-system-based standard for packaging organizational knowledge into portable, interlinked concept files, aiming to a…