Founding cohort·$160 → $130 for the first cohort.·Full price after that.
Machine Learning Foundations
A first-principles introduction to modern machine learning.
A first-principles introduction to modern ML: linear models through deep networks, with daily problem sets and a capstone project.
For students and mid-career professionals making the jump into ML.
You will be able to
do the work.
- Implement core ML algorithms from scratch — linear models, trees, neural networks
- Train and evaluate neural networks on real datasets, end to end
- Debug a model that isn't learning, and know why
- Read and reproduce results from current ML papers
- Defend a capstone project in front of working researchers
What we cover.
Prerequisites
- Linear algebra and calculus to first-year university level
- Programming experience in any language
- 01
Weeks 1–3 — Foundations
Probability, linear algebra you actually use, gradient descent. Linear and logistic regression from scratch.
- 02
Weeks 4–6 — Classical ML
Trees, ensembles, regularisation, model selection. Building a portfolio of working models.
- 03
Weeks 7–9 — Neural networks
Backpropagation by hand, then in PyTorch. CNNs, sequence models, the modern transformer.
- 04
Weeks 10–12 — Capstone
A real dataset, a real question, a written report. Reviewed by working researchers.
Reserve a place.
We open enrollment a few weeks before each cohort starts. Tell us about your work — we'll write to you first when a place opens.
First cohort forming — instructor-led, with daily problem sets and weekly office hours, delivered in Islamabad and online. Join the waitlist and we'll write to you with dates as the cohort fills. Founding rate for the first cohort — full price after that.