You want to understand machine learning, but the math feels like a wall. Terms like "matrix multiplication," "eigenvectors," and "gradient descent" get thrown around without explanation. When something breaks in your ML code, you're stuck.

Here's the truth: Linear algebra is the language of machine learning. Every model—from simple regression to massive transformers—is built on it.

The good news: You don't need to become a mathematician. You need the right linear algebra, taught the right way, for your background.

Why Linear Algebra Matters

• Data is matrices: Rows are samples, columns are features
• Neural networks are matrix operations: Each layer transforms data through matrices
• Everything uses it: PCA, recommender systems, computer vision, NLP, transformers

Without linear algebra, you're following recipes. With it, you understand:

• Why gradient descent works (and when it fails)
• What your model is actually learning
• How to debug shape errors and convergence problems
• When to use which ML technique

And it's learnable. You don't need endless proofs. You need geometric intuition, coding practice, and ML-focused applications.

The Problem: One Size Doesn't Fit All

Most resources assume you're either a complete beginner or a math PhD. But real people are somewhere in between:

• Self-taught developers who can code but never took linear algebra
• Professionals who studied it years ago but forgot most of it
• Researchers from other fields who need the ML-specific perspective

That's why we created three paths—each designed for where you are right now.

Choose Your Path

🧭 Quick Guide:

Choose Alicia if you've never studied linear algebra formally and ML math feels overwhelming.

Choose Beatriz if you took linear algebra in college but need to reconnect it to ML applications.

Choose Carmen if you have graduate-level math and want rigorous ML theory for research.

What Makes These Paths Different?

✅ Curated, not comprehensive - Only what you need, when you need it
✅ Geometric intuition first - See what matrices do before calculating
✅ Code immediately - Implement every concept the same day you learn it
✅ ML-focused - Every topic connects directly to machine learning
✅ Real projects - Build actual ML systems from scratch
✅ 100% free and open source - MIT OpenCourseWare, Khan Academy, 3Blue1Brown

What You'll Achieve

Path 1 (Alicia): Implement algorithms from scratch, use scikit-learn confidently, read ML documentation without fear

Path 2 (Beatriz): Build neural networks in NumPy, read ML papers, debug training failures, transition to ML roles

Path 3 (Carmen): Publish research papers, implement cutting-edge methods, apply ML rigorously to your field

Ready to Start?

Cost: $0 (all the material is free and open-source)
Prerequisites: Willingness to learn and code
Time: 6-14 weeks depending on your path

Choose your path and begin:

→ Path 1: Alicia - Foundation Builder

Perfect for self-taught developers. Start from zero.

→ Path 2: Beatriz - Rapid Learner

Reactivate your math. Connect it to ML fast.

→ Path 3: Carmen - Theory Connector

Bridge your research background to ML.

Linear algebra isn't a barrier—it's a superpower. Let's unlock it together. 🚀