Courses
Pick the starting point that matches what you already know. The foundation courses are optional refreshers, Harness Engineering stands on its own, and the LLM Systems course builds one cumulative browser project.
Where should I start?
You do not have to finish every course in order. Pick the route that feels closest to you.
Two standalone courses with small NumPy exercises.
Linear Algebra Basics
Learn the small set of array operations that appears throughout machine learning.
Fit, prerequisites, and outcome
- Best for
- People who want the array math behind neural networks to stop feeling abstract.
- Before you start
- Basic Python helps. You do not need prior linear algebra.
- Outcome
- You'll be able to read shapes, combine vectors, and follow the matrix operations inside a neural network.
Machine Learning Basics
Learn how data, predictions, loss, gradients, and simple neural networks fit together.
Fit, prerequisites, and outcome
- Best for
- New ML learners and developers who want a clean training-loop refresher.
- Before you start
- Basic Python helps. NumPy experience is not required.
- Outcome
- You'll be able to read a small training loop and tell what the data, parameters, and evaluation results mean.
A standalone course about the execution layer around a language model.
Harness Engineering
Study the deterministic software that turns model responses into bounded, testable agent behavior.
Fit, prerequisites, and outcome
- Best for
- Engineers building reliable agents around an existing language model.
- Before you start
- You should be comfortable reading Python and thinking in tests and state machines.
- Outcome
- You'll be able to implement and test a small model-agnostic agent harness without depending on the Browser Chat project.
One cumulative course whose lesson files form a working browser chatbot.
Build an LLM System in Your Browser
Build the model, inference runtime, serving layer, and React integration for a complete LLM system that runs in the browser.
Fit, prerequisites, and outcome
- Best for
- Python developers who know NumPy, loss, and gradients and want to see the whole LLM path.
- Before you start
- Know basic linear algebra and ML, or use the two foundation courses as refreshers.
- Outcome
- You'll end up with a tested chatbot that saves data locally, runs a real model, loads the RNN checkpoint you trained, and uses browser adapters checked against the same behavior as your Python code.
Make the ideas stick.
Filter by subject, search any concept, then mark each card Got it or Needs work. Your progress stays on this device.
Work the problem, then check the code.
Practice original interview-style questions in the browser. Drafts and solved status stay on this device.