Engineering Intelligence : Machine Learning and AI from First Principles

A book for engineers to understand how these systems actually work

When I first tried to learn machine learning, I kept bouncing between three kinds of books: heavy math textbooks, copy-paste tutorials, and big-picture speculation about AI. None of them gave me what I really wanted: a clear way to understand how these systems actually work.

This book is the one I wish I had. It focuses on concepts and understanding rather than proofs or recipes. It aims to explain clearly without unnecessary complexity. And it tries to stay grounded in how real systems work, including where they fail.

I hope it helps you on your learning journey.

– Oliver Nguyen, January 2026. The book is written with the help of AI.

Concepts over implementation. Understanding how attention works, not memorizing function arguments.
Mental models over completeness. Core ideas that transfer, not exhaustive coverage of every variant.
Engineering over philosophy. How systems work and fail, not speculation about consciousness or AGI.

How learning works

See machine learning as prediction and error reduction, not magic or black boxes.

Why models generalize

Learn what ideas like inductive bias and overfitting really mean for real systems.

How modern AI is built

Understand how models, embeddings, retrieval, and tools come together in practice.

The Frontier

Current directions: scaling patterns, multimodal models, self-improvement, and AGI.