path IV · Artificial Intelligence

Artificial Intelligence

Build learning from nothing: gradient descent, classical ML by hand, a neural net and an autograd engine you write yourself, up to a tiny transformer. No magic — just math you can see run.

0/45 done45 lessons · 9 weeks
Week 1

Foundations — What Learning Is

0/5

Loss, the optimizer, and linear regression as a learning algorithm you code.

Week 2

Classical ML I

0/5

Logistic regression, cross-entropy, regularization, and honest evaluation.

Week 3

Classical ML II

0/5

Trees, ensembles, boosting, SVMs, and Naive Bayes — the pre-deep toolkit.

Week 4

Unsupervised Learning

0/5

Clustering, PCA (which is just SVD), and representation.

Week 5

Neural Networks I — Build One

0/5

The neuron, the forward pass, and backpropagation derived from the chain rule.

Week 6

Neural Networks II — Autograd

0/5

Computational graphs and reverse-mode autodiff — build your own micrograd.

Week 7

Deep Learning

0/5

Why nets fail to train, convolutions, and sequence models.

Week 8

Modern Deep Learning

0/5

Embeddings, attention, and building a tiny transformer language model.

Week 9

Frontier & Capstone

0/5

How LLMs are trained, generative models, alignment, and an end-to-end capstone.