Designing Machine Learning Systems: An Iterative Process for Production-Ready Applications

A while ago, I picked up Designing Machine Learning Systems: An Iterative Process for Production-Ready Applications from Amazon and just had it lying around - but once I picked it up, it was definitely a great read!

If you're picking up this book with the expectation of learning some very specific tools, you are going to be dissapointed. But that's not what this book is about - it's about describing the chaotic ecosystem that is machine learning in production.

While Huyen doesn't necessarily go into depth with what tools you should use, she definitely gives you some ideas as to how to structure your machine learning pipeline such that it is easier to deploy your cool models! Tips and tricks for getting labelled data, how to deal with imbalanced data were great, and in the later chapters on the importance of model monitoring, reproducability and different deployment methods, Huyen introduced ideas and concepts that I had not come across before.

All in all, Designing Machine Learning Systems is a great book and it's a great read for everyone interested in machine learning, and especially for data scientists who all can benefit from good practices in software engineering!