Learning paths
An ordered route through the lessons for one job: ML developer, AI engineer, computer vision engineer.
All sections
The whole curriculum, grouped into 14 areas. Start at the top if you are new. If you came here for one thing, filter to its area or search — everything is on this page or one click from it.
14 areas 117 sections 1,162 of 1,167 lessons published 10 new lessons every day
Foundations
Python and the small amount of maths you need before anything else — start here if you have never written code.
2 sections 17 of 17 lessons published
-
Python for AI
9 of 9 published
-
Mathematics for AI
8 of 8 published
Machine Learning
How a machine learns from data, and the classical algorithms that still beat deep learning on most real tables.
11 sections 105 of 105 lessons published
-
Machine Learning
14 of 14 published
-
Classic Algorithms in Depth
10 of 10 published
-
Linear Models and Regularisation
10 of 10 published
-
Ensembles and Gradient Boosting
10 of 10 published
-
Preprocessing and Feature Selection
9 of 9 published
-
Dimensionality Reduction
10 of 10 published
-
Clustering in Depth
8 of 8 published
-
Outlier and Anomaly Detection
7 of 7 published
-
Validation and Hyperparameter Search
8 of 8 published
-
Calibration and Uncertainty
8 of 8 published
-
Imbalanced, Multi-class and Multi-label
11 of 11 published
Statistics and Experiments
Telling a real result from a lucky one: significance, A/B tests, and working out what actually caused what.
4 sections 42 of 42 lessons published
-
Hypothesis Testing and Inference
11 of 11 published
-
Resampling, Likelihood and Bayes
8 of 8 published
-
Experiment Design and A/B Testing
10 of 10 published
-
Causal Inference Basics
13 of 13 published
Doing the Work
The judgement calls nobody teaches — what to build, what to compare against, and how to know your numbers are honest.
9 sections 79 of 79 lessons published
-
Scoping an ML Project
11 of 11 published
-
Baselines and Choosing a Model
9 of 9 published
-
Reading Training Curves
7 of 7 published
-
Data Leakage and Results That Are Too Good
11 of 11 published
-
Reproducibility and Running Experiments
6 of 6 published
-
ML Code That Survives
7 of 7 published
-
Reading and Reimplementing Papers
9 of 9 published
-
AI Projects
9 of 9 published
-
ML Interview Preparation
10 of 10 published
Deep Learning
Neural networks explained in plain words, then built by hand in PyTorch, all the way to training on many GPUs.
11 sections 113 of 113 lessons published
-
Deep Learning
13 of 13 published
-
PyTorch Tensors
10 of 10 published
-
Autograd in Depth
10 of 10 published
-
Building Models with nn.Module
11 of 11 published
-
Datasets and DataLoaders
10 of 10 published
-
Optimisers, Schedulers and the Training Loop
9 of 9 published
-
Debugging PyTorch
9 of 9 published
-
GPU Memory and Speed
10 of 10 published
-
Multi-GPU and Distributed Training
11 of 11 published
-
Checkpoints, Export and Inference
7 of 7 published
-
Reinforcement Learning
13 of 13 published
Libraries and Frameworks
The libraries you will actually type: scikit-learn, HuggingFace, TensorFlow and JAX.
4 sections 40 of 40 lessons published
-
scikit-learn Properly
7 of 7 published
-
The HuggingFace Stack
13 of 13 published
-
TensorFlow and Keras
10 of 10 published
-
JAX and Flax
10 of 10 published
Language and NLP
Everything to do with text — cleaning it, embedding it, searching it, and answering questions from it.
15 sections 140 of 140 lessons published
-
Natural Language Processing
8 of 8 published
-
Text Preprocessing
9 of 9 published
-
Classical NLP That Still Works
7 of 7 published
-
The Embedding Family
10 of 10 published
-
The BERT Family
10 of 10 published
-
Sequence Labelling and Structure
10 of 10 published
-
Topic Modelling and Text Clustering
9 of 9 published
-
Semantic Search and Reranking
11 of 11 published
-
Chunking and Long Documents
8 of 8 published
-
Question Answering
10 of 10 published
-
Summarisation
10 of 10 published
-
Sentiment, Opinion and Text Mining
9 of 9 published
-
Multilingual and Indic NLP
12 of 12 published
-
Messy Real-World Text
10 of 10 published
-
Evaluating Text Systems
7 of 7 published
Computer Vision
Teaching a computer to see: classification, detection, segmentation, faces, video, documents and 3D.
13 sections 143 of 143 lessons published
-
Computer Vision
8 of 8 published
-
CNN Backbones and Pretraining
13 of 13 published
-
Training Vision Models
12 of 12 published
-
Object Detection in Depth
13 of 13 published
-
Segmentation in Depth
11 of 11 published
-
Faces, People and Pose
12 of 12 published
-
Video Understanding and Tracking
12 of 12 published
-
OCR and Document Vision
11 of 11 published
-
3D Vision and Depth
11 of 11 published
-
Image Generation and Restoration
11 of 11 published
-
Vision Datasets and Annotation
11 of 11 published
-
Evaluating Vision Models
9 of 9 published
-
Shipping Vision Models
9 of 9 published
Speech, Forecasting and Recommenders
The other big families of models — audio and speech, forecasting what happens next, and choosing what to show a person.
3 sections 36 of 36 lessons published
-
Speech and Audio AI
12 of 12 published
-
Time Series and Forecasting
12 of 12 published
-
Recommender Systems
12 of 12 published
Generative AI and LLMs
Models that write, draw and answer questions — and how to build a real product on top of them.
4 sections 44 of 44 lessons published
-
Generative AI
12 of 12 published
-
LLM Development
8 of 8 published
-
Multimodal AI
12 of 12 published
-
Agentic Frameworks
12 of 12 published
Inside a Transformer
Open the model up: attention, tokenisers, how text is generated, and how these things are trained and shrunk.
11 sections 115 of 120 lessons published
-
Attention Mechanics
9 of 9 published
-
Inside a Transformer Block
12 of 12 published
-
How Models Know Word Order
10 of 10 published
-
Tokeniser Internals
13 of 13 published
-
How Text Is Generated
15 of 15 published
-
Fast Attention and Long Context
12 of 12 published
-
Mixture of Experts
10 of 10 published
-
How Models Are Actually Trained
8 of 8 published
-
Post-training and Alignment
14 of 14 published
-
Quantised LLM Inference
3 of 8 published
-
Looking Inside a Trained Model
9 of 9 published
Production and MLOps
Getting a model off your laptop and keeping it alive: data pipelines, serving, scaling, monitoring, cost and on-device.
16 sections 157 of 157 lessons published
-
MLOps
7 of 7 published
-
Data Engineering for AI
13 of 13 published
-
Feature and Data Pipelines in Production
10 of 10 published
-
Serving Models in Production
10 of 10 published
-
Batching and Concurrency
6 of 6 published
-
Latency, Load Testing and Capacity
10 of 10 published
-
Caching and Cost Control
13 of 13 published
-
GPUs: Memory, Scheduling and Cost
10 of 10 published
-
Scaling and Traffic Management
10 of 10 published
-
Registries, Artifacts and Environments
6 of 6 published
-
Testing ML Code and CI
10 of 10 published
-
Releasing Models Safely
10 of 10 published
-
Monitoring Models in Production
11 of 11 published
-
Observability for LLM Applications
9 of 9 published
-
Incident Response for ML Systems
9 of 9 published
-
Edge and On-device AI
13 of 13 published
AI in the Real World
What AI actually looks like in hospitals, banks, shops, factories, farms, schools and laboratories.
12 sections 111 of 111 lessons published
-
AI in Healthcare
8 of 8 published
-
Medical Imaging AI
9 of 9 published
-
AI in Finance and Fraud
9 of 9 published
-
Retail, Demand and Supply Chain
12 of 12 published
-
Manufacturing and Predictive Maintenance
10 of 10 published
-
Robotics, Control and Autonomy
10 of 10 published
-
Climate and Energy AI
7 of 7 published
-
Satellite and Agricultural AI
8 of 8 published
-
AI in Cybersecurity
9 of 9 published
-
AI in Education
10 of 10 published
-
AI for Accessibility
10 of 10 published
-
AI for Science and Engineering
9 of 9 published
Safety, Ethics and Law
Bias, privacy, safety and the rules — the part that decides whether a system should be built at all.
2 sections 20 of 20 lessons published
-
AI Safety and Ethics
13 of 13 published
-
Legal and Compliance AI
7 of 7 published
Other ways to use this site
Projects
Build real things with full code, a dataset and a deployment guide.
AI glossary
Every term you keep seeing, defined in one plain sentence first.
Error database
Paste the error you got. Find out what it means and how to fix it.