Playground · Learn
Learn
A shelf of talks and tutorials we come back to — on AI, machine learning and building software well. Filter by topic, or follow a learning pathway tailored to your role.
No videos match that combination — try a different topic or pathway.
AI Foundations
AI vs Machine Learning
A quick, clear map of how AI, machine learning and deep learning actually relate.
But what is a neural network?
The clearest visual introduction to what a neural network actually is.
Gradient descent — how neural networks learn
The intuition behind how neural networks learn from data.
Backpropagation, intuitively
What backpropagation is really doing when a network learns.
How AI Image Generators Work (Stable Diffusion / DALL·E)
How diffusion models turn random noise into images.
The Strange Math That Predicts (Almost) Anything
Markov chains — the deceptively simple math behind language models, PageRank and much of modern ML.
LLMs & Transformers
How Large Language Models Work
A short, accessible explainer on what LLMs are and how they generate text.
Word Embedding and Word2Vec
How words become vectors — the foundation of how language models represent meaning.
Transformer Neural Networks, ChatGPT's foundation
A clear, step-by-step walk through the transformer architecture behind ChatGPT.
Transformers, the tech behind LLMs
A visual walkthrough of the transformer architecture behind modern AI.
Attention in transformers, step-by-step
The attention mechanism — the key idea that makes transformers work — explained visually.
How might LLMs store facts
A visual look at where factual knowledge actually lives inside a transformer's weights.
Intro to Large Language Models
A one-hour, non-hype primer on how LLMs work and where they're going.
Let's build GPT: from scratch, in code
Build a transformer language model line by line, in code.
Building & Using AI
What is Retrieval-Augmented Generation (RAG)?
How RAG grounds LLMs in your own data — the backbone of most enterprise AI.
How I use LLMs
Practical, real-world patterns for getting the most out of LLM tools.
Let's build the GPT Tokenizer
Build the tokenizer behind GPT from scratch — the crucial, often-overlooked first step in every LLM.
Deep Dive into LLMs like ChatGPT
A thorough tour of how modern chat LLMs are actually built and trained.
Stanford CS229 · Building Large Language Models
A full Stanford lecture on how modern LLMs are built — pretraining, data and post-training.