Foundations of Machine Learning: A Practitioner's Journey — From Mathematical Foundations to Classical Algorithms

The first volume of A Practitioner's Journey. Eighteen chapters take you from linear algebra and probability through every classical machine-learning algorithm worth knowing — regression, trees, ensembles, SVMs, KNN, time series, and recommendation systems — with the math, the intuition, and runnable code, all in one place.

This is the curriculum a working ML practitioner actually needs. Most "intro to ML" books pick a side: pure math with no code, or library-call tutorials that fall apart the moment you try to apply them. Foundations of Machine Learning refuses both. Every chapter is built around a working scenario. Every code example runs. Every concept comes with both the math and the intuition.

You will learn to:

• Reason about linear algebra, calculus, probability, and optimization the way ML uses them
• Derive and implement classical algorithms from first principles, not as library calls
• Choose the right algorithm for the right problem and explain why
• Evaluate models honestly, avoid overfitting, and know when "good enough" is good enough
• Apply the CRISP-DM framework to a real end-to-end case study

Companion volumes: Book 2 Machine Learning in Production covers deep learning, computer vision, and the production engineering stack. Book 3 Artificial Intelligence in Production covers LLMs, RAG, agents, and modern AI infrastructure.