Andrej Karpathy
13 mentions across all digests
Andrej Karpathy is an AI researcher whose observation that backend service setup is too hard for agents inspired a wave of companies — including Stripe — to launch CLI interfaces designed for agent-native infrastructure access.
[AINews] Autoresearch: Sparks of Recursive Self Improvement
Autoresearching Apple's "LLM in a Flash" to run Qwen 397B locally
Apple's LLM in a Flash technique enables a 397B-parameter Qwen model to run on a MacBook M3 Max at 5.5 tokens/sec by streaming 4-bit quantized weights from SSD, leaving only 5.5GB resident in RAM.
What is agentic engineering?
Agentic engineering redefines the developer's role from code author to problem specifier and instruction refiner, establishing a rigorous, production-quality discipline distinct from prototype-quality "vibe coding."
Show HN: Auto-Architecture: Karpathy's Loop, pointed at a CPU
Autonomous agent applies Karpathy's research loop to SystemVerilog CPU design, discovering microarchitecture optimizations 56% better than hand-tuned designs in under 10 hours.
Show HN: How LLMs Work – Interactive visual guide based on Karpathy's lecture
Interactive visual guide breaks down modern LLM architecture from tokenization to inference at production scale (15 trillion tokens, 405 billion parameters), making Karpathy's technical lecture accessible.