Curated video tutorials, playlists, and courses to help you master AI development.
Understanding the Transformer architecture that powers every modern LLM — GPT-4, Claude, Gemini. Covers self-attention, multi-head attention, and positional encoding.
The definitive hands-on tutorial by Andrej Karpathy. Build a GPT language model from scratch using Python and PyTorch, implementing every component — tokenization, embeddings, multi-head attention, and training. 2 hours of pure deep learning.
Andrej Karpathy's complete course on building neural networks from scratch. Starts with backpropagation basics (micrograd), then builds up through MLP, WaveNet, and full transformer language models. The most thorough practical ML course on the internet.
3Blue1Brown's crystal-clear visual explanation of backpropagation — the algorithm that makes neural network training possible. Essential for understanding how LLMs learn from data.
Andrej Karpathy trains GPT-2 from scratch live on screen using modern techniques. Covers FP16 training, Flash Attention, gradient clipping, cosine learning rate schedules, and deploying to cloud GPUs — real-world LLM training from beginning to end.