A book about practical problems

Available now on Amazon and O'Reilly.

example

I wrote this book to give readers tools to solve the most common practical ML problems based on my experience mentoring hundreds of Data Scientists and ML Engineers. This book will help you build practical applications that are powered by ML.

To build ML powered applications, you'll need to convert product goals to an ML approach, explore and label datasets, debug models, and plan deployment strategies. These are hard problems, and they are rarely covered in textbooks. This book is dedicated to them.

We will go through every step of the ML process together, and help you accomplish each of them by sharing a mix of methods, code examples, and advice from me and other experienced practitioners.

To hear more about what the book covers, I encourage you to:

Covers 95% of the job of a Data Scientist

Data Scientists often complain that training models is only 5% of the job, with 95% of their time spent narrowing down product use cases, wrangling data, and deploying their work. This book's goal is to share approaches and advice to better tackle this part of the role, the 95%.

Building Machine Learning Powered Applications (BMLPA) covers the process of ML, from product idea to deployment. It particularly focuses on aspects outside of model training.

The book is concrete and practical. It contains detailed code examples and explanations at every step of the way.

Code notebooks, case study, and interviews

Overall, BMLPA includes:

  • Fifteen notebooks to illustrate concepts.
  • An end-to-end case study demonstrating how to use tools.
  • Four discussions with industry leaders about practical realities of the field.
  • Code snippets for common tasks.
  • Practical tips for less trendy aspects of ML such as dataset creation and labeling, model debugging and model deployment.

Reviews and aknowledgements

The book made FloydHub's list of best ML books ever published.

It has also received praise from engineers and leaders at the best tech companies in the world. See what they had to say about the book.

“So many books about machine learning skip the hardest parts: refining the problem, debugging models, and deploying to customers. But this book focuses on them so you can move your projects from an idea to making an impact."

Alexander Gude, Staff Data Scientist, Intuit

“ML models need to be integrated into data products and larger systems to be useful. This is a crucial and hard skill to master. I recommend this excellent book by Emmanuel Ameisen. It covers the entire end-to-end process of building and managing data products."

Jeremy Howard, Founder & Deep Learning researcher, fast.ai

“If you are looking for practical advice on how to get ML models into production, what could go wrong and what to watch out for, this is your book. I wish I had it 10 years ago. Lots of the lessons I had to learn the hard way."

Lukas Tencer, Senior Manager, ML at Twitch

“From product thinking, to infrastructure, to the inner workings of machine learning models, this book gamely tackles all of the areas that an MLE needs to be successful. […] It's so good to FINALLY find a book that discusses deploying and monitoring ML applications and building CI/CD pipelines for ML. Whether you're coming to machine learning engineering by way of data science or by way of software engineering this book holds something for you."

David Stevens, Software Engineer, Peloton

“It is so full of best practices, it should become mandatory for all ML’ers."

Darvish Shadravan, Machine Learning, Salesforce

“If you're a practitioner looking to understand the end-to-end process of developing machine learning based products, then this is the book for you. Emmanuel superbly describes each stage of machine learning development, from framing the problem to designing, implementing and operating the models and data pipelines. This book will show you how to build real machine learning systems."

Luigi Patruno, Founder, MLinProduction.com

“This book was sorely needed in the ML world. There are tons of books out there that detail how ML algorithms work, but this is the first I've come across that explicitly details how to make ML projects work."

Jon Krohn, Chief Data Scientist, Untapt

“Having worked with Emmanuel as Head of AI at Insight, I vouch for how fantastic his guidance is on this topic."

Jake Klamka, Founder, Insight Data Science

“the first book I’ve read that's written the way I write books: build an actual product from end to end. […] Badass!"

Russell Jurney, O'Reilly Author

“If you're looking to pick up the skills to break into ML Engineering, I highly recommend this book!"

Jeremy Karnowski, VP of product, Insight Data Science

“In the jungle of publications about ML, this book provides a unique hands-on and principled set of tools to really get you through a project from start to finish. A must read to any working data scientist or data engineer out there. Can't recommend it enough."

Amazon review

“I think everyone who wants to work on machine learning projects should read this book. It's a good and quick read and can be referred back to again and again."

Goodreads review

“This book answered so many questions I had about a transition between an ML playground experiment to having an ML-powered product. Lots of practical examples mixed with insightful interviews. Extremely glad I picked this book up!"

Goodreads review

Want to try it before you buy it?

In order to help you make sure this book is the right for you, I'm sharing a free PDF of the first chapter which shares tools to go from product goals to ML approaches, along with the table of contents to give you an overview of the topics. If you'd like to learn more about building ML powered applications, order a copy below!

Available now on Amazon and O'Reilly.