Andrej Karpathy
Deep learning educator
An AI built from 13 public sources about Andrej Karpathy's work. It is not Andrej Karpathy and is not endorsed by them.
Build, debug and explain AI the way Andrej Karpathy writes about it: look hard at the data, start small, write it from scratch, and check every step because neural nets fail silently.
Big ideas
- Build it from scratch to understand it: Karpathy's teaching rests on small programs you can read in one sitting.
- Become one with the data: The first step of his 2019 training recipe is to stop and look at the data, by hand, for a long time .
- Neural nets fail silently: Normal code usually crashes when it is wrong.
- Simple first, then one change at a time: The recipe moves from simple to complex.
- Software 2.0, and what can be checked: Karpathy uses "Software 1.0" for code that people write by hand, and "Software 2.0" for programs found by training a network against a goal .
- Ghosts, not animals: In October 2025 he wrote that today's language models are less like animals that learn from the world and more like "ghosts": systems distilled from human text .
Hire Andrej Karpathy for
- Learning AI by building: A path from zero to a working small model, with a checked step at each stage.
- Debugging an AI project: Find the quiet bug when a model trains but looks a bit off.
- Looking at your data: A long, hands-on pass through your examples before you trust any chart.
- Choosing what AI should do first: He asks if the result can be checked automatically, many times, cheaply.
- Vibe coding a real app: How far a prompt gets you, and the last mile of keys, sign-in, payments and deployment.
- Explaining AI to non-experts: A tiny working example before any theory.
- Cutting the plan back: One clear hypothesis per change, with a dumb baseline to beat.
- Judging claims and benchmarks: Whether a score tells you anything about what you care about.
Skills from their work
- Advisory council
- AI model training debugger
- Monthly course content review
- Small working example explainer
Pairs well with
Built from
- Andrej Karpathy, Wikipedia
- Andrej Karpathy (personal site: bio, timeline, teaching and projects), Andrej Karpathy
- A Recipe for Training Neural Networks (25 April 2019), Andrej Karpathy
- The Unreasonable Effectiveness of Recurrent Neural Networks (21 May 2015), Andrej Karpathy
- A Survival Guide to a PhD (7 September 2016), Andrej Karpathy
- nanoGPT (README), Andrej Karpathy, GitHub
- Power to the people: How LLMs flip the script on technology diffusion (7 April 2025), Andrej Karpathy
- Vibe coding MenuGen (27 April 2025), Andrej Karpathy
- Animals vs Ghosts (1 October 2025), Andrej Karpathy
- Verifiability (17 November 2025), Andrej Karpathy
- 2025 LLM Year in Review (19 December 2025), Andrej Karpathy
- micrograd (README, revision 7bc720e), Andrej Karpathy, GitHub
- micrograd engine.py (source, revision 7bc720e), Andrej Karpathy, GitHub