1 Getting started with MLOps and ML engineering

This chapter introduces the practice of taking machine learning beyond notebooks into reliable, scalable production systems. It frames MLOps as the discipline that closes the gap between experimentation and real-world operation, emphasizing that most failures stem not from model quality but from system reliability, data complexities, and process gaps. The authors adopt a hands-on path that builds confidence through practical patterns, automation, and reproducibility, guiding readers from core concepts to a working platform and concrete projects. The goal is to help practitioners of varied backgrounds—data scientists, software engineers, and ML engineers—navigate the full journey with clarity and momentum.

The ML life cycle is presented as iterative and orchestration-driven: start by validating that ML is the right tool, then progress through problem formulation, data collection and preparation, data versioning, training, evaluation, and business-facing validation. These steps are assembled into automated pipelines for consistency and speed. The transition to dev/staging/production introduces end-to-end automation with CI triggers, robust deployment practices (containerization, scaling, rollback strategies), and comprehensive monitoring across system performance, data/model drift, and business impact. Retraining is treated as a recurring, automated process, activated by schedules or performance thresholds to maintain model quality over time.

To execute this effectively, the chapter outlines the essential skill set: strong software engineering fundamentals, practical ML proficiency with common frameworks, data engineering literacy, and a bias toward automation—plus just enough Kubernetes to be productive. Readers are guided to incrementally build a modern ML platform centered on Kubeflow and Kubeflow Pipelines, then extend it with capabilities such as a feature store (to reduce training-serving skew), a model registry (for lineage and promotion), and CI/CD-driven model deployment. Tool choices are pragmatic and adaptable (build vs buy is context-dependent), and the lessons generalize to LLMOps. The chapter culminates with a roadmap of three projects—an OCR system, a tabular movie recommender, and a RAG-based documentation assistant—that anchor the concepts in realistic architectures and workflows.

The experimentation phase of the ML life cycle

The dev/staging/production phase of the ML life cycle

MLOps is a mix of different skill sets

The mental map of an ML setup, detailing the project flow from planning to deployment and the tools typically involved in the process

Traditional MLOps (right) extended with LLMOps components (left) for production LLM systems. Chapters 12-13 explore these extensions in detail.

An automated pipeline being executed in Kubeflow.

Feature Stores take in transformed data (features) as input, and have facilities to store, catalog, and serve features.

The model registry captures metadata, parameters, artifacts, and the ML model and in turn exposes a model endpoint.

Model deployment consists of the container registry, CI/CD, and automation working in concert to deploy ML services.

• The Machine Learning (ML) life cycle provides a framework for confidently taking ML projects from idea to production. While iterative in nature, understanding each phase helps you navigate the complexities of ML development.
• Building reliable ML systems requires a combination of skills spanning software engineering, MLOps, and data science. Rather than trying to master everything at once, focus on understanding how these skills work together to create robust ML systems.
• A well-designed ML Platform forms the foundation for confidently developing and deploying ML services. We'll use tools like Kubeflow Pipelines for automation, MLFlow for model management, and Feast for feature management - learning how to integrate them effectively for production use.
• We'll apply these concepts by building two different types of ML systems: an OCR system and a Movie recommender. Through these projects, you'll gain hands-on experience with both image and tabular data, building confidence in handling diverse ML challenges.
• Traditional MLOps principles extend naturally to Large Language Models through LLMOps - adding components for document processing, retrieval systems, and specialized monitoring. Understanding this evolution prepares you for the modern ML landscape.
• The first step is to identify the problem the ML model is going to solve, followed by collecting and preparing the data to train and evaluate the model. Data versioning enables reproducibility, and model training is automated using a pipeline.
• The ML life cycle serves as our guide throughout the book, helping us understand not just how to build models, but how to create reliable, production-ready ML systems that deliver real business value.

FAQ