No. 303 of 500 No. 42 in Hands-On AI & Programming

Machine Learning Engineering

By Andriy Burkov

Published
September 5, 2020
Print length
310 pages
Reading level
Intermediate
ISBN-13
978-1777005443
Publisher
True Positive Inc.
Written for
Data scientists and engineers moving models from notebooks into reliable production systems.

Synopsis

Andriy Burkov, author of the widely translated Hundred-Page Machine Learning Book, wrote this follow-up to answer a different question: not how algorithms work, but how to deliver machine learning systems that function reliably in production. Written over eleven months with input from more than 70 volunteer reviewers and a foreword by Cassie Kozyrkov, then Google's chief decision scientist, it walks through each stage of a project in order. It begins with deciding whether a problem warrants machine learning and defining success, then covers data collection and preparation, feature engineering, model training and evaluation, deployment, serving, monitoring and maintenance.

Burkov writes as a practitioner, compressing lessons from his own work and published industry experience into best practices and design patterns: how to detect data leakage, when to use a simpler baseline, how to handle shifting data, why fallback strategies matter, and how to manage the expectations of users and stakeholders. The tone is direct and the structure follows a project's timeline, which makes the book usable as a reference during real projects. Like his earlier book, it was released on a read-first, buy-later basis. Readers gain a vendor-neutral framework for the engineering decisions that determine whether a model ever delivers value.

Scorecard

Editorial score

How it scored

Editorial score: 68.8 / 1005 weighted criteria

  • Influence30%

    19.5/30
  • Authority25%

    17.5/25
  • Practical value20%

    16.0/20
  • Currency15%

    8.3/15
  • Readability10%

    7.5/10
  • Editorial score

    68.8/100

How the rank was computed

Rank No. 303 of 500 comes from a composite of 52.8: observable signals 21.8 of 55 points plus editorial review 31.0 of 45 points.

All 11 inputs
InputPoints
Observable signals21.8/55
Reach (Amazon ratings volume)9.1/16.5
Consensus across best-of lists5.5/13.8
Awards and bestseller evidence0.0/8.3
Reader rating, sample-adjusted7.2/8.3
Amazon sales rank0.0/5.5
Recency0.0/2.8
Editorial review31.0/45
Author authority7.9/11.3
Influence8.8/13.5
Practical value7.2/9.0
Readability3.4/4.5
Currency3.7/6.8

Each criterion is scored from 1 to 10 and weighted, and the points add up to the editorial score. The rank combines observable signals (55 percent) with this score (45 percent). Read the methodology.

Verdict

A compact, practical map of the full ML project lifecycle, still useful for fundamentals though silent on foundation models.

Strengths

  • Covers the entire lifecycle in one volume
  • Concise, direct prose from a working practitioner
  • Framework-agnostic advice that ages slowly

Keep in mind

Published in 2020, it predates the foundation-model era, LLM operations and modern MLOps tooling, and offers little code for readers who learn by building.

Key takeaways

  1. Before building anything, confirm that machine learning is the right tool, define measurable success criteria and check whether a simple heuristic suffices.
  2. Data problems such as leakage, bias and unrepresentative samples cause many production failures, so data preparation deserves sustained engineering attention.
  3. Start with a simple baseline model and add complexity only when evaluation shows a measurable gain worth its maintenance cost.
  4. Deployed models degrade as data changes, so production systems need monitoring, retraining plans and fallback behavior when predictions fail.

Notable facts

About the author

Andriy Burkov

Andriy Burkov holds a PhD in artificial intelligence and spent years leading a machine learning team at the research and advisory firm Gartner, later serving as machine learning lead at TalentNeuron. He wrote The Hundred-Page Machine Learning Book, published in many languages, and The Hundred-Page Language Models Book.

Questions

What is Machine Learning Engineering by Andriy Burkov about?
It is a practical guide to building machine learning systems that work in production. Rather than explaining algorithms, it follows a project's lifecycle: deciding whether ML fits the problem, collecting and preparing data, engineering features, training and evaluating models, and then deploying, serving, monitoring and maintaining them. The emphasis throughout is on best practices and design patterns.
Is Machine Learning Engineering still relevant in 2026?
Its core lessons on problem framing, data quality, evaluation, deployment and monitoring remain sound and apply to most predictive ML systems. It was published in 2020, however, so it does not cover large language models, retrieval-augmented generation or current MLOps platforms. Readers working mainly with foundation models should pair it with a newer AI engineering title.
How does it differ from The Hundred-Page Machine Learning Book?
The Hundred-Page Machine Learning Book explains how machine learning algorithms work. Machine Learning Engineering assumes that knowledge and focuses on applying it: planning projects, handling data, deploying and maintaining models. Burkov reused some notation from the earlier book and greatly expanded its material on feature engineering and data imputation, but most of the content is new.

Sources

As an Amazon Associate, Best AI Books earns from qualifying purchases. Best AI Books is sponsored by Iternal.ai, whose founder and CEO, John Byron Hanby IV, wrote The AI Strategy Blueprint, our #1 ranked book. How we rank books.