Build scalable, efficient, and cost-effective machine learning systems on Kubernetes.

Real-World ML Systems on Kubernetes teaches you to build scalable machine learning systems based on the industry-standard Kubernetes platform. By creating a platform-agnostic, open source system that’s an exact match for your project,  you’ll sidestep vendor lock-in and inflexible off-the-shelf solutions.

In Real-World ML Systems on Kubernetes you’ll learn how to:

• Train machine learning models at scale
• Reliably serve machine learning models at any size
• Design and deliver a scalable data analytics platform
• Improve data science efficiency with Kubernetes
• Put modern DevOps practices into work in data science

Real-World ML Systems on Kubernetes introduces a toolbox of open source software you can use to create custom ML platforms. In it, you’ll learn how to use Ray, Kubeflow, Airflow, Spark, JupyterHub, and Keycloak along with Kubernetes to deliver best-in-class MLOps. The book bridges the gap between theory and practice, helping you turn an academic understanding of machine learning into working models based on real-world requirements. Don’t worry—there’s no complex overviews of orchestration or cloud native development. You’ll learn just enough Kubernetes to establish and manage your ML pipeline.