Neural Networks: Zero to Hero
Curated by Andrej Karpathy • Added 2026-08-01
Build neural networks, backprop, and GPT from scratch with Andrej Karpathy
Former Tesla AI Director and OpenAI founding member Andrej Karpathy walks you step-by-step from raw micrograd autograd engines to training your own GPT transformer from empty Python files.
- ✓Building a custom micrograd scalar backpropagation engine from first principles
- ✓Implementing multi-layer perceptrons, activations, and loss functions from scratch
- ✓Designing, training, and sampling from character-level and Byte-Pair language models
- ✓Coding a complete Generative Pre-trained Transformer (GPT) architecture line-by-line
- ★Taught by world-renowned AI pioneer
- ★No hidden abstractions or black boxes
- ★100% free with code on GitHub
Add the Learnlists badge to your website
Show visitors that Neural Networks: Zero to Hero is curated and listed on Learnlists. Paste the embed code on your landing page, docs, or footer to drive verified social proof.
Select Style
Pick Standard or Compact in Light or Dark.
Copy Snippet
Includes automatic attribution & canonical link.
Paste into HTML
Compatible with React, Astro, WordPress, Webflow.
Interactive preview · Click to verify target destination
Related in AI & Machine Learning
Other vetted resources in this discipline
TrenTorch
Build machine learning frameworks from scratch by hand
Fanout
Mapped technical learning across AI research, systems, and ML math
Inference Engineering
Deep-dive technical guide to LLM inference optimization & runtime mechanics
