Machine Learning
Maths, models, and getting them into production.
advanced · 8–12 months · 3 milestones · 10 topics
Tap a box to open it · right-click to mark done
Tap a box to open it
Full outline
The genuine ML path: the mathematics underneath, classical algorithms, deep learning, and the MLOps work that turns a notebook into a deployed system.
01Mathematical Base
Skippable for a demo. Not skippable for a career.
- Linear algebra
The language every model is written in.
- Calculus & optimisation
How models learn.
- Probability & statistics
Reasoning under uncertainty, which is the entire field.
- Linear algebra
02Classical ML
Often the right answer on real tabular data.
- Supervised learning
Learning a mapping from labelled examples.
- Unsupervised learning
Finding structure without labels.
- Evaluation & validation
The part beginners skip and then wonder why production fails.
- Feature engineering
Usually worth more than a fancier model.
- Supervised learning
03Deep Learning & Production
Neural networks, then getting them to actually serve traffic.
- Neural networks
From a perceptron to a transformer.
- Training in practice
Where theory meets impatience and GPU bills.
- MLOps
A model that is not deployed has produced no value.
- Neural networks