Guided learning paths

Build understanding in the right order.

Choose a subject and follow a curated sequence. Each lesson shows where you are and gives you one clear next step.

11 lessons

Start with the foundations

Probability distributions

Learn to recognize the random process behind a problem and choose a distribution that matches it.

1. 1A field guide to probability distributions
2. 2Binomial
3. 3Geometric

Start with A field guide to probability distributions

10 lessons

From measurement to decisions

Experimentation and causal inference

Build an experiment correctly, choose an analysis, and understand what changes when randomization is unavailable.

1. 1P-values and confidence intervals
2. 2z and t statistics
3. 3Designing a trustworthy A/B test

Start with P-values and confidence intervals

7 lessons

Build the workhorses of ML

Linear models, step by step

Go from ordinary least squares to regularization, Bayesian priors, classification, and multiclass models.

1. 1OLS foundations
2. 2Multicollinearity and regularization
3. 3Evaluation and interpretation

Start with OLS foundations

4 lessons

Geometry first, kernels later

Support Vector Machines

Build the maximum-margin classifier, derive its dual, unlock kernels, and extend it to novelty detection.

1. 1Geometry, margins, and hinge loss
2. 2The dual and support vectors
3. 3The kernel trick

Start with Geometry, margins, and hinge loss

5 lessons

From one tree to an ensemble

Trees and boosting

Understand decision trees, Random Forests, gradient boosting, XGBoost, and the practical library trade-offs.

1. 1Decision trees
2. 2Random Forests
3. 3AdaBoost and gradient boosting

Start with Decision trees

7 lessons

Understand how networks learn

Deep learning foundations

Follow the mathematical and architectural choices that keep signals, gradients, and representations useful.

1. 1Jacobians and Hessians
2. 2Activation functions
3. 3Backpropagation and optimizers

Start with Jacobians and Hessians

9 lessons

From generation to efficient models

Practical LLM systems

Learn how decoding, memory, retrieval, quantization, and parameter-efficient fine-tuning work under the hood.

1. 1LLM sampling techniques
2. 2Temperature
3. 3KV cache

Start with LLM sampling techniques

8 lessons

Follow an ad through the system

AdTech from first principles

Trace the auction, infrastructure, machine learning, privacy constraints, and bidding decisions behind an impression.

1. 1The invisible auction
2. 2Who decides which ad gets served?
3. 3The millisecond handshake

Start with The invisible auction