Free, comprehensive courses on Python, C, C++, and the mathematical foundations of large language models — taught from first principles to production-ready skills.
What You'll Learn
Every course is self-contained — no fluff, no filler. Clear explanations, worked examples, and hands-on exercises from page one.
Linear algebra, calculus, probability, and optimization — taught through transformers, VAEs, and modern deep learning.
The lingua franca of AI. Data structures, OOP, and scientific computing — built from first principles.
Understand memory, pointers, and how computers actually work. Low-level intuition every serious engineer needs.
Modern C++ for high-performance systems. STL, memory models, and production-grade engineering patterns.
Research & Implementations
Rebuild breakthrough machine learning papers from scratch. There's no better way to build deep intuition than implementing the ideas yourself.
Recursive reasoning with tiny networks, focusing on latent state updates and answer refinement.
An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale - applying Transformers directly to image patches for vision tasks.
Deep G-Network (DQN) combining G-learning with deep neural networks for end-to-end learning of action values from raw pixels.
Adaptive moment estimation optimizer combining benefits of RMSProp and momentum, computing individual adaptive learning rates.
Implementing core LSTM components from scratch: LSTM cells with gates, forward/backward passes, BPTT, initialization, dropout masks, packed sequences, bidirectional LSTMs, and full LSTM blocks.
Framework for estimating generative models via an adversarial process.
The seminal transformer architecture replacing recurrence and convolutions entirely with self-attention mechanisms.
Training generative neural network models of popular reinforcement learning environments to learn a compressed representation of the spatial and temporal aspects of the environment.
Fundamental sequential processing architecture forming the basis of modern recurrent neural architectures.
Devlog
In-depth technical articles on AI safety, systems scaling, and the future of software engineering.
Case studies, model safety evaluations, and codebase scaling audits — written by engineers, for engineers.
Pick a track, open a lesson, and write your first line of code. Everything is free — no sign-up, no credit card.
Browse Courses