Resources

Understanding Deep Learning

Simon J.D. Prince

If you want to truly understand how deep learning works — not just use it — this is the book. It walks you through the mathematical intuition behind neural networks, starting from function approximation and building up to modern architectures. You'll come away with a clear picture of why things work, not just how to call an API. I found it especially useful for connecting the dots between theory and practice.

Designing Machine Learning Systems

Chip Huyen

This book bridges the gap between training a model in a notebook and running it in production. Chip covers everything from data engineering and feature stores to deployment, monitoring, and iteration. It's the kind of book I wish I had when I first started thinking about ML systems beyond just model accuracy.

Machine Learning System Design Interview

Ali Aminian & Alex Xu

If you're preparing for ML system design interviews, this book gives you a clear framework for structuring your answers. It walks through real-world problems — recommendation systems, ad click prediction, search ranking — and shows you how to break them down step by step. Really helped me think more systematically about designing end-to-end ML systems.

Software Engineering for Data Scientists

Catherine Nelson (O'Reilly)

Most DS/ML folks learn Python but skip the engineering fundamentals. This book fills that gap — it covers Git, Docker, OOP, testing, and how to write production-quality code. It taught me how to structure projects properly and think about code as something that needs to be maintained, not just run once.

Architecture Patterns with Python

Harry Percival & Bob Gregory (O'Reilly)

This one changed how I organize code. It explains repository patterns, service layers, unit of work, and domain-driven design in Python — with practical examples. If you've ever struggled with where to put things in a growing codebase or how to keep your ML pipeline code clean, this book gives you solid patterns to follow.