Phase 1 of 6 · Weeks 120

Phase 1 — Mathematical & Programming Foundations

Twenty weeks that turn school algebra into the working mathematics every model is written in — plus the Python to express it.

What you'll be able to do

  • Represent data as vectors and matrices, and transform it with a matrix you wrote yourself
  • Differentiate a function by hand, then check your answer against code
  • Implement gradient descent from scratch and watch it find a minimum
  • Reason about uncertainty with probability distributions and Bayes' theorem
  • Write, test and version real Python — not notebook fragments
  • Decompose a matrix with SVD and explain what each part is doing

What you need before you start

Assumed on day one

  • Comfort with school algebra — solving equations, rearranging formulas
  • Reading a graph: axes, slope, intercept
  • Willingness to write code. No prior programming needed

Taught here — not assumed

  • Set theory and logic — the language datasets and conditions are written in
  • Functions, logarithms, exponentials and basic trigonometry
  • Limits, derivatives and multivariable calculus
  • Vectors, matrices, vector spaces, rank, eigentheory and SVD
  • Probability, random variables, distributions, MLE and confidence intervals
  • Python from installation through functions, data structures, NumPy and Git

Week by week (20 weeks)

  1. W1Algebra Refresher + Python Setup
  2. W2Functions, Logs, Trig + Python Data Structures
  3. W3Limits & Derivatives + Python Functions
  4. W4Chain Rule + Object-Oriented Programming
  5. W5Integration + Modules, Errors, Environments
  6. W6Multivariable Calculus + Files, Testing
  7. W7Linear Algebra I — Vectors + Git Deeply
  8. W8Linear Algebra II — Matrices + NumPy Intro
  9. W9Linear Algebra III — Spaces & Eigen + NumPy Mastery
  10. W10SVD & Decompositions + Pandas Intro
  11. W11Probability I + Matplotlib
  12. W12Probability II + EDA Workflow
  13. W13Probability III + Statistical Functions
  14. W14Statistics + SQL Basics
  15. W15Information Theory + Advanced SQL
  16. W16Optimization Basics + SWE Hygiene
  17. W17Constrained Optimization + Docker
  18. W18Math Consolidation Week
  19. W19Programming Capstone
  20. W20Phase 1 Buffer + Honest Self-Assessment

What you'll build

  • A public `ai-curriculum` repo holding your notes in markdown
  • 20+ small Python exercises — FizzBuzz, prime check, anagram detector, word counter
  • Gradient descent implemented in NumPy on a function you choose
  • A from-scratch linear-algebra toolkit checked against NumPy

Free, no-signup explainers and interactive demos covering this phase's ideas.

Questions about this phase

Do I need to be good at maths to start Phase 1?
No. Phase 1 assumes school algebra and the ability to read a graph. Everything beyond that — calculus, linear algebra, probability — is taught here from first principles, in the order machine learning actually needs it.
Do I need to know how to program?
No. Week 1 starts with installing Python, VS Code and Git. By Week 20 you are writing tested, version-controlled NumPy code.
Can I skip Phase 1 if I already write Python?
Possibly, but check the maths outcomes first. Most developers know the Python and not the linear algebra or probability. The readiness check will point you at the specific weeks worth reviewing rather than making you redo all twenty.
How long does Phase 1 take?
Twenty weeks at the intended pace of one week per week. It is the longest phase because everything after it depends on this material being solid rather than merely familiar.

Start with a real lesson, not a sales page.

Play through a complete lesson — predict, explore the demo, prove you've got it. No account, no card.