Book review: Designing Machine Learning Systems

Here’s a review post on “Designing Machine Learning Systems” that I’ve finished reading.

Designing Machine Learning Systems

Machine learning systems are both complex and unique. Complex because they consist of many different components and involve many different stakeholders. Unique because they're data dependent, with data varying wildly … - Selection from Designing Machine Learning Systems [Book]

www.oreilly.com

One of the most unique point of this book is that it gives us a holistic view of ML systems. ML systems don’t consist of only ML algorithms but involves lots of components such as data infrastructure, serving, monitoring, ML platform, etc.

As this book focuses on the holistic view, it doesn’t pretty much mention each component of ML systems. If you want to know a specific component in detail, it would be better to find an another one.

Having said that, I’d highly recommend reading it. Especially for newgrads, reserchers who come from academia and engineers who come from non-ML areas, this book would be a great fit as it tells us what it looks like to build ML systems and make it keep working in a sustainable way.