Best books for Machine Learning Platform Engineers

Hands-On Machine Learning with Scikit-Learn, Keras, and TensorFlow

by Aurélien Géron · 2017

~860 pages~22 hrs to read

A code-first introduction to classical ML and deep learning. Read it to understand what your users train and what they need from the platform.

Designing Machine Learning Systems

by Chip Huyen · 2022

~390 pages~10 hrs to read

Covers data pipelines, feature engineering, deployment and monitoring for production ML. A map of the problems an ML platform exists to solve.

Introducing MLOps

by Mark Treveil, Nicolas Omont, Clément Stenac, Kenneth Chen, Mike Del Balso · 2020

Explains the practices around deploying, monitoring and governing models, and who is involved at each stage. A good first overview of MLOps.

Building Machine Learning Pipelines

by Hannes Hapke, Catherine Nelson · 2020

~340 pages~9 hrs to read

Shows how to automate the model lifecycle (data validation, training, serving and orchestration) with TensorFlow Extended. Read it when you build reusable pipelines for other teams.

Machine Learning Design Patterns

by Valliappa Lakshmanan, Sara Robinson, Michael Munn · 2020

~400 pages~10 hrs to read

Catalogues recurring solutions for data, training, serving and reproducibility problems. Useful when you design platform features such as feature stores and model registries.

Reliable Machine Learning

by Cathy Chen, Niall Murphy, Kranti Parisa, D. Sculley, Todd Underwood · 2022

~410 pages~10 hrs to read

Applies SRE practice to ML systems: monitoring, incident response, data quality and organizational ownership. Read it when your platform is on the critical path for production models.

Designing Data-Intensive Applications

by Martin Kleppmann · 2017

~620 pages~16 hrs to read

Explains storage, replication, batch and stream processing and the trade-offs between them. A base for the data systems that sit under feature stores and training pipelines.

Fundamentals of Data Engineering

by Joe Reis, Matt Housley · 2022

Covers the data engineering lifecycle: generation, ingestion, storage, transformation and serving. Useful because ML platforms depend on the same data layer.

Distributed Machine Learning Patterns

by Yuan Tang · 2022

Covers patterns for scaling training and serving across clusters using tools like Kubernetes and Kubeflow. Read it when you support distributed training and inference.

Platform Engineering: A Guide for Technical, Product, and People Leaders

by Camille Fournier, Ian Nowland · 2024

~330 pages~8 hrs to read

Covers how to run an internal platform as a product: scope, team structure, adoption and measurement. Directly applicable to an ML platform with internal users.

Team Topologies

by Matthew Skelton, Manuel Pais · 2019

~240 pages~6 hrs to read

Describes team types and interaction modes for fast flow, including platform and enabling teams. Read it when you decide how the ML platform team works with data science teams.

AI Engineering: Building Applications with Foundation Models

by Chip Huyen · 2024

~350 pages~9 hrs to read

Covers how foundation-model applications are built, evaluated and served, including cost and latency. Read it when your platform must support LLM workloads as well as classical ML.