Kubernetes and MLOps — A nice read during the summer holidays!
Recently I got the opportunity to read the “Machine Learning on Kubernetes: A practical handbook for building and using a complete open-source machine learning platform on Kubernetes” book.
The book begins with an overview of machine learning and describes the basics of MLOps. There is also a nice introduction on Kubernetes. These introductions are not broing at all :) and they make a lot of sens!
I found this approach very interesting, if you’re a pure k8s guy and know few or nothing about machine learning and MLOps, the introduction guides you well. On the other hand, if you’re a pure “data scientist” and not really into containers and clusters, then again, the introduction gives you a real good idea about what Kubernetes is, and then it’s easy to understand why and how leverage Kubernetes to build machine learning solutions on top of it.
After these fine introductions, the book guides the reader to steps such as model deployment and automation, building a complete ML project or even building your own data pipeline.
I really enjoyed reading this book, and I don’t pretend to be a data scientist or data engineer, but I have a clearer or more practical picture of how to combine the best of the breed to use the power of Kubernetes to deploy AI/ML based solution.
A very good read!