ML Foundations

The core ideas every AI system is built on

Data, features, training and evaluation — the fundamental mechanics of machine learning, taught visually.

Beginner5 modules · 14 lessons~104 min970 XPAligned with Google ML Crash CourseStart with: AI Literacy Start with: Math for AI

Progress0/14

Module

Data & Features

What models actually see

Seeing Like a Model7 min · 60 XPSeeing Like a Model

The Test Your Model Can't See6 min · 55 XPThe Test Your Model Can't See

Why Data Quality Wins6 min · 55 XPWhy Data Quality Wins

Module

Core Algorithms

Regression, classification, and decision boundaries

Drawing the Line8 min · 70 XPDrawing the Line

Gradient Descent, Properly8 min · 70 XPGradient Descent, Properly

Regression: Fitting Lines That Predict7 min · 65 XPRegression: Fitting Lines That Predict

Module

Working with Data

Taming numbers, categories, and feature combinations

Numerical Data at Scale8 min · 70 XPNumerical Data at Scale

Categorical Data8 min · 75 XPCategorical Data

Feature Engineering & Crosses7 min · 70 XPFeature Engineering & Crosses

Module

Metrics That Matter

Why the number you optimize decides what you ship

Accuracy Is a Trap7 min · 65 XPAccuracy Is a Trap

Precision vs Recall8 min · 80 XPPrecision vs Recall

Thresholds, ROC & AUC8 min · 80 XPThresholds, ROC & AUC

Module

Generalization

Learning patterns that survive contact with new data

Train/Validation/Test Done Right7 min · 70 XPTrain/Validation/Test Done Right

Regularization: Simplicity That Generalizes9 min · 85 XPRegularization: Simplicity That Generalizes

Capstone project

Your first model

Train a classifier on a real tabular dataset (Titanic, Iris, or one you find). Start with the dumbest baseline possible, then beat it. Compare honestly.

Deliverable: A notebook comparing baseline vs. trained model, with a paragraph on what mattered most.

Milestones

Finish every lesson to earn the Machine Learning Foundations Certificate.