A team built a persistent LLM Wiki to stop their agent from re-deriving the same answers from scratch, using a directory of markdown files on a dedicated repo branch. The wiki classifies questions int…
Andrej Karpathy's open-source 'autoresearch' repository, released March 7, 2026, enables AI agents to autonomously run machine learning experiments, achieving an 11% training time reduction for GPT-2 …
Vibe coding, a term coined by former Tesla AI director Andrej Karpathy, allows non-programmers to build software by describing their intent in plain English to an AI model that writes and deploys the …
A vendor-neutral analysis of the vibe coding stack identifies only two paid tools worth buying from day one: the AI editor Cursor ($20/month) and the terminal agent Claude Code ($20/month on Claude Pr…
AI-assisted 'vibe coding' is causing a surge in security vulnerabilities and leaked secrets, with developers producing code at three to four times the rate but introducing security flaws at ten times …
The AI industry is facing a shift from computation to memory as the fundamental primitive for intelligent systems, according to Andrej Karpathy. Just as GPUs replaced CPUs as the dominant compute arch…
Andrej Karpathy has been largely inactive on social media and GitHub since joining Anthropic nearly two months ago, raising questions about whether his reduced public presence is beneficial for the AI…
Andrej Karpathy proposed a 'cognitive core' AI model that prioritizes reasoning over encyclopedic knowledge, suggesting a small, always-on model that relies on external tools for facts. The thesis cha…
A new project turns Ilya Sutskever's famous list of 30 papers for John Carmack into AI-narrated audio episodes, explaining key insights from foundational deep learning works. The list, which Sutskever…
At least 22 professors and researchers from Stanford, Berkeley, and Harvard have left or taken leave in 2026 to join OpenAI, Anthropic, Google DeepMind, and Meta, accelerating a brain drain from acade…
A developer has compiled a free, open-source syllabus for mastering AI and data engineering in 2026, arguing that traditional bootcamps are unnecessary. The curriculum covers data architecture, analyt…
OpenBMB and Tsinghua NLP lab released MiniCPM5-1B, a 1-billion-parameter open-source AI model designed to run on older smartphones. The model achieves top rankings in its class for reasoning and tool …
ContextNest, an open-source CLI tool, launches to provide versioned and governed context for AI agents, addressing the problem of persistent memory loss across chat sessions. The tool tracks changes t…
LangChain released OpenWiki, an open-source tool that uses an AI agent to automatically generate and maintain a structured wiki for codebases, keeping documentation synchronized with code changes via …
Salesforce is promoting Anthropic's Claude Tag inside Slack, a product that competes with its own Slackbot and Agentforce, because Salesforce's equity stake in Anthropic is worth roughly $5 billion, f…
A developer building a product with an AI assistant realized the model wasn't learning between sessions; instead, the compounding knowledge base of project notes, conventions, and logs made the AI app…
A developer has ported Meta's Llama 2 large language model to run locally on vintage DOS machines, including a 1996 Toshiba Satellite with a 200 MHz Pentium MMX processor. The project, called DOS Llam…
Vibe Arcade built every game in its collection using 'vibe coding,' a term coined by OpenAI co-founder Andrej Karpathy that describes generating software by describing desired features in plain langua…
A guide explains how to build a tiny large language model from scratch, covering tokenization, pretraining, supervised fine-tuning, and alignment. It aims to demystify the LLM pipeline for developers …
A developer created a low-tech pattern for providing repository context to AI coding agents using Markdown files and text search. The approach uses short index.md files in each directory for navigatio…