What this book covers

Chapter 1, Introduction to ML Engineering, explains what we mean by ML engineering and how this relates to the disciplines of data science and data engineering. It covers what you need to do to build an effective ML engineering team, as well as what real software solutions containing ML can look like.

Chapter 2, The Machine Learning Development Process, explores a development process that will be applicable to almost any ML engineering project. It discusses how you can set your development tooling up for success for later chapters as well.

Chapter 3, From Model to Model Factory, teaches you how to build solutions that train multiple ML models during the product life cycle. It also covers drift detection and pipelining to help you start to build out your MLOps practices.

Chapter 4, Packaging Up, discusses best practices for coding in Python and how this relates to building your own packages and libraries for reuse in multiple projects.

Chapter 5, Deployment Patterns and Tools, teaches you some of the standard ways you can get your ML system into production. In particular, the chapter will focus on hosting solutions in the cloud.

Chapter 6, Scaling Up, teaches you how to take your solutions and scale them to massive datasets or large numbers of prediction requests using Apache Spark and serverless infrastructure.

Chapter 7, Building an Example ML Microservice, walks through how to use what you have learned elsewhere in the book to build a forecasting service that can be triggered via an API.

Chapter 8, Building an Extract Transform Machine Learning Use Case, walks through how to use what you have learned to build a pipeline that performs batch processing. We do this by adding a lot of our newly acquired ML engineering best practices to the simple package created in Chapter 4, Packaging Up.