MLOps engineers have to deal with a glut of tools and SaaS applications, not to mention technical debt clogging the system. Such complexity requires a comprehensive approach. The Databricks platform provides all the critical components for end-to-end MLOps and LLMOps in one place. This exhaustive book shows you how to use Databricks to build and manage a robust ML system that delivers on your business's needs. Maria Vechtomova guides you through MLOps principles and explains how Databricks handles the machine learning lifecycle holistically, from data preparation to model deployment and monitoring, and enables data engineers, data scientists, and MLOps engineers to collaborate seamlessly. To put all the pieces together, you'll navigate two ML projects: a real-time ML application and an LLM-based system that highlights LLM-specific Databricks features. Understand the Databricks components for MLOps and LLMOps Unpack ML Model Serving architectures Track your machine learning experiments and register your models Build an ML application that uses Feature and Model Serving, and Model Serving with automatic feature lookup Deploy a real-time ML application and an LLM-based application Monitor your AI applications on Databricks Understand how MLOps principles fit into AI governance
