Machine Learning Platform Engineering Build an internal developer platform for ML and AI systems

Get a free eBook (PDF or ePub) from Manning as well as access to the online liveBook format (and its AI assistant that will answer your questions in any language) when you purchase the print book.

Delivering a successful machine learning project is hard. This book makes it easier. In it, you’ll design a reliable ML system from the ground up, incorporating MLOps and DevOps along with a stack of proven infrastructure tools including Kubeflow, MLFlow, BentoML, Evidently, and Feast.

A properly designed machine learning system streamlines data workflows, improves collaboration between data and operations teams, and provides much-needed structure for both training and deployment. In this book you’ll learn how to design and implement a machine learning system from the ground up. You’ll appreciate this instantly-useful introduction to achieving the full benefits of automated ML infrastructure.

In Machine Learning Platform Engineering you’ll learn how to:

• Set up an MLOps platform
• Deploy machine learning models to production
• Build end-to-end data pipelines
• Effective monitoring and explainability

About the technology

AI and ML systems have a lot of moving parts, from language libraries and application frameworks, to workflow and deployment infrastructure, to LLMs and other advanced models. A well-designed internal development platform (IDP) gives developers a defined set of tools and guidelines that accelerate the dev process, improving consistency, security, and developer experience.

About the book

Machine Learning Platform Engineering shows you how to build an effective IDP for ML and AI applications. Each chapter illuminates a vital part of the ML workflow, including setting up orchestration pipelines, selecting models, allocating resources for training, inference, and serving, and more. As you go, you’ll create a versatile modern platform using open source tools like Kubeflow, MLFlow, BentoML, Evidently, Feast, and LangChain.

What's inside

• Set up an end-to-end MLOps/LLMOps platform
• Deploy ML and AI models to production
• Effective monitoring, evaluation, and explainability

About the reader

For data scientists or software engineers. Examples in Python.

About the author

Benjamin Tan Wei Hao leads a team of ML engineers and data scientists at DKatalis. Shanoop Padmanabhan is a software engineering manager at Continental Automotive. Varun Mallya is a senior ML engineer at DKatalis.

Table of Contents

Part 1
1 Getting started with MLOps and ML engineering
2 What is MLOps?
3 Building applications on Kubernetes
Part 2
4 Designing reliable ML systems
5 Orchestrating ML pipelines
6 Productionizing ML models
Part 3
7 Data analysis and preparation
8 Model training and validation: Part 1
9 Model training and validation: Part 2
10 Model inference and serving
11 Monitoring and explainability
Part 4
12 Designing LLM-powered systems
13 Production LLM system design
A Installation and setup
B Basics of YAML

Benjamin Tan is a product manager and principal engineer for sata Science at DKatalis where he leads a team of talented machine learning engineers, data scientists, and data engineers. He is also the author of The Little Elixir and OTP Guidebook and Building an ML Pipeline with Kubeflow (liveProject) from Manning, and Mastering Ruby Closures.

Shanoop Padmanabhan is a software engineering manager at Continental Automotive, where he leads a team of software engineers focusing on machine learning based perception for autonomous vehicles.

Varun Mallya is a machine learning engineer working at DKatalis where he is responsible for the setup and maintenance of the Bank’s machine learning platform.