Beginning MLOps with MLFlow: Deploy Models in AWS SageMaker, Google Cloud, and Microsoft Azure

Practical MLOps, from model building to deployment

Beginning MLOps with MLFlow: Deploy Models in AWS SageMaker, Google Cloud, and Microsoft Azure is a hands-on guide to turning machine learning work into something that can actually run in the real world. The book starts with data analysis and model construction, then moves into the MLOps mindset: tracking experiments, managing model versions, and pushing models through repeatable deployment workflows.

The focus is refreshingly practical. Rather than staying at the level of theory, the authors work through scikit-learn and PySpark modeling, then show how MLFlow fits into the development process before moving on to deployment in AWS, Azure, and Google Cloud. For readers trying to connect training notebooks to production systems, that progression matters.

What the book covers

• Data analysis using a credit card dataset
• Building logistic regression models in scikit-learn and PySpark
• Hyperparameter tuning and model validation
• MLFlow tracking for runs, parameters, metrics, graphs, and models
• Deployment workflows in AWS SageMaker, Azure, and Google Cloud
• Databricks integration and MLFlow Model Registry usage

Why it stands out

This title is especially useful for readers who already understand basic machine learning and want to see how MLOps changes the day-to-day workflow. The book connects experimentation, reproducibility, and cloud deployment in a single path, which makes it easier to understand how production ML systems are assembled.

Good fit for

Data scientists, ML engineers, and technically inclined developers who want a structured introduction to MLOps with a strong emphasis on MLFlow and cloud deployment will find this a practical, readable starting point.