Learn AI Fundamentals
See how machines learn - from vectors and lines to neural networks, vision, language and ethics.
About this track
A beginner-friendly tour of artificial intelligence that you can touch. Drag vectors, fit lines by hand, tune a neuron’s weights, roll gradient descent down a loss curve, paint pixels, slide edge-detecting kernels over an image, and generate text one word at a time.
Along the way you’ll learn what AI really is, the bit of math it runs on, how models learn from data and how we check they did, and where AI goes wrong - bias, hallucinations and brittle predictions. Every lesson ends with small Python exercises that run in your browser, and the last one has you build a complete classifier from scratch.
- No AI or math background needed - every formula is explained.
- Basic Python helps for the exercises; the lessons themselves need none.
- Unit 1 · 0/4 lessons
Foundations
What AI is, the math it runs on, how machines learn, and the three kinds of learning. Badge: AI curious
- 0x001What is artificial intelligence?Tell AI from ordinary software, see how AI, machine learning and deep learning nest, and meet the field’s ups and downs. 15 min 0/2 exercises solved 15 min 0/2 exercises solved
- 0x102The math toolkit: vectors and matricesDescribe things as lists of numbers, compare them with the dot product, and multiply matrices - the arithmetic AI runs on. 20 min 0/2 exercises solved 20 min 0/2 exercises solved
- 0x203How machines learn from dataFeatures and labels, the guess-check-adjust loop, train/test splits, overfitting, and why data quality decides everything. 20 min 0/2 exercises solved 20 min 0/2 exercises solved
- 0x304The three types of machine learningSupervised learning from answers, unsupervised learning from structure, and reinforcement learning from rewards. 20 min 0/2 exercises solved 20 min 0/2 exercises solved
- Unit 2 · 0/2 lessons
First models
Fit a line with linear regression and classify with k-nearest neighbors. Badge: Model maker
- 0x405Your first model: linear regressionFit a straight line to data, measure how wrong it is with mean squared error, and grow it to many features. 20 min 0/2 exercises solved 20 min 0/2 exercises solved
- 0x506Teaching a machine to classifyClassify new examples by asking their nearest neighbors, choose k, and see decision boundaries form. 20 min 0/2 exercises solved 20 min 0/2 exercises solved
- Unit 3 · 0/2 lessons
Neural networks
Build neurons and layers, and train them with gradient descent. Badge: Neuron tuner
- 0x607Neural networks: the basicsBuild an artificial neuron from a weighted sum and an activation, then stack neurons into layers. 22 min 0/2 exercises solved 22 min 0/2 exercises solved
- 0x708How networks learn: loss and gradient descentRoll downhill on the loss, pick a learning rate that neither crawls nor explodes, and see how backpropagation trains every weight. 22 min 0/2 exercises solved 22 min 0/2 exercises solved
- Unit 4 · 0/4 lessons
AI in the world
Computer vision, language models, ethics and limitations - and where to go next. Badge: AI explorer
- 0x809Computer vision: how machines seeTreat images as grids of numbers, slide kernels across them to find edges, and stack convolutions into a CNN. 22 min 0/2 exercises solved 22 min 0/2 exercises solved
- 0x9010Language and large language modelsSplit text into tokens, give words meaning with embeddings, and generate text one predicted token at a time. 22 min 0/2 exercises solved 22 min 0/2 exercises solved
- 0xA011AI ethics, bias and limitationsFind where bias sneaks in, recognize what AI can’t do, and check a system’s errors group by group. 20 min 0/2 exercises solved 20 min 0/2 exercises solved
- 0xB012Where to go nextReview the core vocabulary, build a complete tiny classifier end to end, and pick your next track. 15 min 0/1 exercises solved 15 min 0/1 exercises solved