Track · Students and engineers who want depth

ML engineer

The maths and machinery, start to frontier.

A full path from fitting a line to training and aligning large models, with code you can run along the way.

0 of 17 finished

1. 1Intermediate · 6 min · up nextLines and curves: regressionThe humble straight line is the ancestor of every neural network. Linear and logistic regression, explained.
2. 2Intermediate · 6 minGradient descent: rolling downhillHow models find good weights: feel the slope, take a small step down, repeat.
3. 3Intermediate · 5 minOverfitting & the Goldilocks modelToo simple misses the pattern, too complex memorises the noise. How to land just right.
4. 4Intermediate · 7 minDecision trees and random forestsA game of twenty questions that learns from data, and why a whole forest of them still beats neural networks on spreadsheets.
5. 5Intermediate · 6 minFinding groups: clusteringHow a computer can sort things into groups when nobody has told it what the groups are, with k-means step by step.
6. 6Intermediate · 6 minIs my model any good?Accuracy can lie. Meet the confusion matrix, precision, recall and F1.
7. 7Intermediate · 6 minBackpropagation: blame, passed backwardsHow a network with millions of weights figures out which ones to adjust.
8. 8Intermediate · 7 minSeeing and remembering: CNNs & RNNsSpecial network shapes for images and sequences, and why they mattered.
9. 9Intermediate · 5 minEmbeddings: meaning as mapsHow AI turns words, images and songs into points in space where closeness means similarity.
10. 10Advanced · 8 minAttention is all you needThe transformer architecture and the attention mechanism that powers every modern LLM.
11. 11Advanced · 7 minHow LLMs are trainedPre-training, supervised fine-tuning, and learning from feedback: the three-stage recipe.
12. 12Advanced · 8 minMixture of experts: big models, small billsHow some of the largest models only use a small slice of themselves for each token, and what that saves and costs.
13. 13Advanced · 7 minDistillation and the rise of small modelsHow a big model can teach a small one, why small models keep getting better, and when a little model is the right choice.
14. 14Advanced · 7 minReinforcement learning, deeperPolicies, values and rewards: the maths of learning by doing.
15. 15Advanced · 7 minThinking before answering: reasoning modelsWhy letting a model write out its thinking makes it smarter, and how reinforcement learning taught models to think for longer.
16. 16Advanced · 7 minScaling laws, evals & safetyWhy bigger got better, how we measure frontier models, and the open problems in keeping AI safe.
17. 17Advanced · 8 minKeeping AI on our side: alignment and interpretabilityHow developers try to make models helpful, honest and harmless, how they test for danger, and how researchers are learning to read a model's mind.