AdvancedAges 15+Small-group online

Introduction to Machine Learning

The real foundations — linear algebra, models by hand, and a project that leads to research.

See the curriculum

Prefer to talk it through first? Book a free consultation — no pressure, and you can enroll right after.

First-class guarantee. Sit the first class — if it’s not the right fit, tell us and we’ll refund you in full.

Max 6 students12 weeks · Sep–Dec 202624 hrs live instruction

The two-part track — and where it leads

You're viewingIntro to ML12 weeks · Sep–Dec 2026
Part 2Intro to ML II12 weeks · live
Fall term 2026 · Sep 6 – Dec 6 · 12 weekly sessions

About this course

A rigorous, math-honest introduction to machine learning for students who intend to go further. This is not a drag-and-drop survey: students build genuine foundations — the linear algebra ML actually depends on, taught from scratch, plus the specific calculus that appears in gradient descent and backpropagation — and implement real models by hand. Over the term they move from what learning-from-data means, through linear and logistic regression, gradient descent, over- and under-fitting and regularization, into the basics of neural networks, and finish with a small end-to-end project. The whole course is built to open the door to a real research project in the sequel — and to give strong students something substantive to show college admissions.

Is this the right fit?

Who it’s for
Motivated students aged 15+ who want a genuine, math-honest foundation in ML — and who intend to do a research project and use it in college applications.
Prerequisites
Basic single-variable calculus (comfort with derivatives) and some Python (loops and functions). No prior linear algebra or ML libraries assumed — the course teaches the linear algebra it needs from scratch.
Tools students use
Python, NumPy, Matplotlib, scikit-learn, and Keras/PyTorch in Google Colab
Where it leads
Most students go on to Introduction to Machine Learning II.

Class schedule

Two weekly time-slot options, so students in Asia, the Gulf, and the Americas can all join live. Pick the one that fits; a coach confirms it when you enroll.

Asia / Gulf
SaturdaysSat 7:00 AM9:00 AM
ETSat 7:00 AMPTSat 4:00 AMGSTSat 3:00 PMISTSat 4:30 PMSGTSat 7:00 PM
Americas
SundaysSun 11:00 AM1:00 PM
ETSun 11:00 AMPTSun 8:00 AMGSTSun 7:00 PMISTSun 8:30 PMSGTSun 11:00 PM

Times convert to your zone automatically and account for daylight saving. A coach confirms your slot when you enroll.

What students walk away with

  • The linear algebra ML depends on — vectors, matrices, dot products, matrix multiplication — taught from scratch
  • Linear and logistic regression, and gradient descent, implemented by hand
  • The conceptual heart of honest ML: bias–variance, over/under-fitting, regularization, and proper evaluation
  • The basics of neural networks and how they learn, via backpropagation
  • A completed, evaluated end-to-end project and an honest write-up — the launch point for research

What a session looks like

  1. 0:00

    Concept block

    The week's idea, developed carefully with the math it needs — no hand-waving.

  2. 0:45

    Coding lab

    Implement the idea in Python and NumPy, often watching a model learn iteration by iteration.

  3. 1:45

    Consolidate

    Tie the code back to the math, and set the between-session exercise.

Curriculum

1

Week 1 — What machine learning is

Supervised vs. unsupervised vs. reinforcement learning; learning from data vs. hand-written rules. Set up Python, Colab, and NumPy.

2

Week 2 — Linear algebra I: vectors

Vectors as data and as geometry; addition, scaling, the dot product and what it means. Why features are vectors.

3

Week 3 — Linear algebra II: matrices

Matrices as data tables and as transformations; matrix multiplication, the transpose, and why vectorized operations matter.

4

Week 4 — Linear regression

The model, the mean-squared-error cost, and visualizing the cost surface. The calculus we need: partial derivatives and gradients, in context.

5

Week 5 — Gradient descent

The optimization idea behind almost all of ML: the learning rate, convergence, and failure modes. Implement it by hand and watch a model learn.

6

Week 6 — Multiple features

Multivariate regression, feature scaling, and feature engineering. Predict a real quantity end to end.

7

Week 7 — Classification & logistic regression

From predicting numbers to predicting classes: the sigmoid, decision boundaries, and cross-entropy loss.

8

Week 8 — Over/under-fitting & regularization

The bias–variance tradeoff, train/validation/test splits, and L2 regularization. The conceptual heart of doing ML honestly.

9

Week 9 — Evaluating models properly

Why accuracy is not enough: precision, recall, F1, confusion matrices, class imbalance, and cross-validation.

10

Week 10 — Neural networks I

From logistic regression to a network of neurons: layers, weights, activations, and forward propagation.

11

Week 11 — Neural networks II

How they learn: backpropagation as the chain rule at scale, and the training loop. Train a network on a classic dataset.

12

Week 12 — Capstone & the road to research

Present a small end-to-end project — a dataset, a question, a trained-and-evaluated model, an honest write-up — and plan the research sequel.

What’s included

  • 12 live small-group classes, capped at 6 students
  • Colab notebooks and datasets for every lab
  • Between-session exercises with feedback
  • A completed end-to-end project with a written report
  • Guidance on how to talk about the work in applications

How to join

  1. Purchase your seat. Choose your time-zone slot at checkout and pay securely. Classes are capped at six, so seats go in the order they’re booked.
  2. We confirm the details. After checkout a coach confirms your slot and asks a few quick questions so the first class is pitched right for your child.
  3. Sit the first class risk-free. If it’s not the right fit, tell us after that class and we’ll refund you in full.

Not sure yet? Book a free consultation or email hello@redwoodacademy.ai and a coach will help you choose.

Frequently asked questions

What math does my student need?

Basic single-variable calculus — comfort with functions, slopes, and derivatives. We teach the specific pieces (partial derivatives, the chain rule as it appears in backprop) as they come up, but a student should not be meeting derivatives for the first time here. No prior linear algebra is required; we build it from scratch.

Do they need to know Python?

Some. A student should be able to write a loop and a function. No NumPy, pandas, or ML libraries are assumed — the course builds those up.

Why is this class two hours?

The material is genuinely mathematical, and each session pairs a concept block with a live coding lab where students implement what they just learned. Two hours gives room to do both properly.

What does this lead to?

Introduction to Machine Learning II, where each student does a guided first research project to a standard they can credibly discuss in a personal statement, interview, or supplemental essay.

Ready when you are.
Your seat is one click away.

Enroll in Introduction to Machine Learning now — and if the first class isn’t the right fit, we’ll refund you in full.

Still deciding? Book a free consultation or email hello@redwoodacademy.ai