Who This Book Is For

This book was not written for computer scientists. It was written for people who understand a problem deeply — a business, a dataset, a patient, a market — and want to use that understanding to make better predictions.

If that sounds like you, keep reading. This chapter will tell you exactly who this book is for, what you need to bring, and what you don't.

Who This Book Is Written For

Doctors and researchers. You already know how to interpret symptoms, measurements, and outcomes. This book shows you how to let a computer search for patterns across more cases than you could review by hand — and how to stay in charge of what those patterns mean.

HR specialists, marketers, and business owners. You make repeated decisions using data you already collect — who might leave, who might buy, what demand looks like next quarter. This book turns that data into predictions you can act on.

Engineers and analysts. You're comfortable with data, spreadsheets, and maybe even some scripting, but machine learning itself may still feel like a separate discipline. This book closes that gap quickly and practically.

Traders and quants. If you're building automated or rules-based trading systems and want to add ML-driven signals without becoming a full-time data scientist first, this book gives you a practical way to experiment — and an honest look at where those signals can and can't be trusted.

Programmers. You already write code comfortably, but haven't worked with machine learning specifically. This book won't waste your time re-teaching you to program. It gets you straight to running real experiments.

Students and the curious. No professional stake required — just a willingness to learn by doing, on real data, from the very first chapter.

Wherever you fit on this list, one thing connects all of you: you don't need to become a machine learning researcher. You need machine learning to work for you.

You Don't Need to Be a Programmer

You do not need to be a programmer to use AutoML.

If you already write code, that's useful — but it isn't required. This book will teach you enough Python, one step at a time, to run real experiments and understand what's happening. And MLJAR Studio, the tool we'll use throughout the book, is built to lower that barrier even further, so you're never stuck staring at an empty script wondering where to begin.

By the end of this book, you'll be able to go from a spreadsheet to a working prediction — whether or not you've ever written a line of code before.

You also don't need advanced mathematics. As we said in Chapter 1, you don't need to understand an engine to drive a car well — you need to know how to control it, what the warning signs mean, and when to be careful. This book takes the same approach with machine learning. The ideas will be introduced when they're useful, in plain language, with real examples — not as a wall of formulas to get through first.

What You Do Need

A few things will help, though none of them are technical.

A real question. The book works best when you have something you actually want to predict — even a rough one. A dataset you've been sitting on, a decision you keep making by gut feeling, a "what if we could estimate this in advance" that's been in the back of your mind.

A willingness to install and run MLJAR Studio. The examples in this book aren't just theory — you'll run real experiments on real data, on your own computer.

Patience for a bit of trial and error. Machine learning is experimental by nature, as we saw in Chapter 2. Your first result won't always be your best one, and that's normal, not a sign you're doing something wrong.

If This Isn't the Right Book for You

This book won't teach you the mathematical foundations of machine learning in depth — the proofs behind optimization, the theory behind why certain algorithms converge. If you're looking for that, a more academic textbook will serve you better.

It also isn't aimed at experienced machine learning engineers looking for cutting-edge research or novel architectures. If you already build custom models by hand and want to push the state of the art, much of this book will feel too introductory.

This book is for people who want to go from "I have data and a question" to "I have a working prediction I understand and trust" — as directly and honestly as possible.

How This Book Is Organized

You've just finished Part 1, which laid the groundwork: what machine learning is, why AutoML exists, and what kinds of predictions it can make. Everything from here is more hands-on.

Getting set up. Part 2 walks you through Python without any fear involved, the different ways to work with it, installing MLJAR AutoML, and organizing your first project folder. By the end, your workspace will be ready.

Learning the foundations. Part 3 covers how tabular data is organized and introduces the six real datasets you'll use throughout the book. Part 4 walks you through training your first AutoML model and making sense of what it gives you back.

Practicing on real problems. Part 5 is where you'll get real repetition — six full projects, covering both regression and classification, from an insurance dataset all the way through a multi-class wine classification problem. Part 6 then teaches you to properly read what AutoML hands back: leaderboards, metrics, learning curves, feature importance, and more.

Going deeper. Parts 7 and 8 are for once you're comfortable with the basics — choosing the right AutoML mode for a given job, and tuning things like training time, algorithms, and validation strategy when you want more control.

Using models responsibly. Part 9 covers the mistakes and blind spots that matter most in practice — data leakage, fairness, and keeping people in the loop on important decisions.

Putting a model to work. Part 10 shows you how to turn a trained model into something other people can actually use — batch predictions, a simple web app, sharing results with a team.

Wrapping up. Part 11 closes with a checklist, the most common mistakes to avoid, and pointers for where to go next once you've finished the book.

That's eleven parts and a lot of ground — but every chapter is built to stand on the ones before it, so you'll never be asked to know something you haven't been taught yet. And once you're through it, this book works just as well as a reference to come back to, chapter by chapter, whenever a specific question comes up.

Chapter Summary

This book is for domain experts, researchers, analysts, and business users — but also for programmers looking to add ML to their toolkit and traders looking to bring more sophisticated logic to their systems. What connects everyone on that list is not a shared technical background, but a shared goal: turning data you already understand into predictions you can use.

You do not need to be a programmer, and you do not need advanced mathematics. You do need a real question, a willingness to experiment, and some patience along the way.

With that settled, Part 1 is complete. In the next part of the book, we'll get hands-on — starting with how tabular data is organized, and the datasets we'll use throughout the rest of the book.