Andrej Karpathy
Founder · ex-Tesla / OpenAI
Eureka Labs
Former Director of AI at Tesla (Autopilot) and co-founder of OpenAI. Now building AI-native education at Eureka Labs. Known for deeply technical yet accessible explanations of how modern AI systems work.
Core Positions & Ideas
Software 2.0
Neural networks are not just a tool but a new programming paradigm. Instead of writing explicit logic, we specify goals and optimize over a vast space of programs via gradient descent. This shifts software engineering from writing rules to curating training data.
LLMs as the New Operating System
Large language models are becoming the OS of AI applications — orchestrating tools, managing memory, routing tasks. Building on LLMs is analogous to building apps on Unix. The LLM is the kernel; everything else is userspace.
Vibe Coding as a Legitimate Paradigm
Coined the term 'vibe coding': fully giving in to AI for code generation without reading every line. Argues this is a legitimate new mode of programming that massively boosts throughput for exploration, prototyping, and one-off scripts.
Vision-Only Autonomy is the Right Bet
Led Tesla Autopilot's controversial decision to remove radar and rely purely on cameras, arguing that if humans navigate with eyes, a vision-only system has the highest ceiling. Once considered reckless — now widely validated by the industry.
Education Must Be Rebuilt for the AI Era
Founded Eureka Labs to reimagine education with AI tutors available to everyone. Believes the scarcest resource in AI adoption is not compute or data, but understanding — and open education is the highest-leverage intervention.
Essential Reading & Watching
Software 2.0
The landmark essay that reframed neural networks as a new mode of programming, predicting much of what has since come true about AI-driven software development.
Read / Watch →
The Unreasonable Effectiveness of Recurrent Neural Networks
A classic demonstration of character-level RNNs generating text, code, and mathematics — made deep learning accessible to a generation of practitioners.
Read / Watch →
Neural Networks: Zero to Hero
A free, ground-up video course rebuilding GPT from scratch in Python. Widely considered the best hands-on deep learning resource available anywhere online.
Read / Watch →
Let's build GPT: from scratch, in code, spelled out
A 2-hour deep dive implementing a GPT-class language model from scratch, demystifying transformer architecture for a broad technical audience.
Read / Watch →
Intro to Large Language Models
A 1-hour accessible overview of how LLMs work, their capabilities, limitations, and security challenges — one of the most-watched AI explainers ever made.
Read / Watch →
Recent Writing (8)
microgpt
.post-header h1 { font-size: 35px; } .post pre, .post code { background-color: #fcfcfc; font-size: 13px; /* make code smaller for this post... */ } This is a brief guide to my new art project microgpt, a single file of 200 lines of pure Python with no dependencies that trains and inferences a GPT. This file contains…
February 12, 2026Read original →
Deep Neural Nets: 33 years ago and 33 years from now
.post-header h1 { font-size: 35px; } .post pre, .post code { background-color: #fcfcfc; font-size: 13px; /* make code smaller for this post... */ } The Yann LeCun et al. (1989) paper Backpropagation Applied to Handwritten Zip Code Recognition is I believe of some historical significance because it is, to my knowledge,…
March 14, 2022Read original →
A from-scratch tour of Bitcoin in Python
.wrap { max-width: 900px; } p { font-family: sans-serif; font-size: 15px; font-weight: 300; overflow-wrap: break-word; /* allow wrapping of very very long strings, like txids */ } .post pre, .post code { background-color: #fafafa; font-size: 13px; /* make code smaller for this post... */ } pre { white-space: pre-wrap;…
June 21, 2021Read original →
Short Story on AI: Forward Pass
p { text-align: justify; } .post pre, .post code { border: none; background-color: #eee; } The inspiration for this short story came to me while reading Kevin Lacker’s Giving GPT-3 a Turing Test. It is probably worth it (though not required) to skim this post to get a bit of a background on some of this story. It was…
March 27, 2021Read original →
Biohacking Lite
Throughout my life I never paid too much attention to health, exercise, diet or nutrition. I knew that you’re supposed to get some exercise and eat vegetables or something, but it stopped at that (“mom said”-) level of abstraction. I also knew that I can probably get away with some ignorance while I am young, but at…
June 11, 2020Read original →
A Recipe for Training Neural Networks
Some few weeks ago I posted a tweet on “the most common neural net mistakes”, listing a few common gotchas related to training neural nets. The tweet got quite a bit more engagement than I anticipated (including a webinar :)). Clearly, a lot of people have personally encountered the large gap between “here is how a…
April 25, 2019Read original →
(started posting on Medium instead)
The current state of this blog (with the last post 2 years ago) makes it look like I’ve disappeared. I’ve certainly become less active on blogs since I’ve joined Tesla, but whenever I do get a chance to post something I have recently been defaulting to doing it on Medium because it is much faster and easier. I still…
January 20, 2018Read original →
A Survival Guide to a PhD
This guide is patterned after my “Doing well in your courses”, a post I wrote a long time ago on some of the tips/tricks I’ve developed during my undergrad. I’ve received nice comments about that guide, so in the same spirit, now that my PhD has come to an end I wanted to compile a similar retrospective document in…
September 7, 2016Read original →
