Machine Learning Engineering on AWS
This 3-day course teaches how to build, deploy, orchestrate, and operationalize machine learning solutions at scale on AWS.
The Machine Learning Engineering on AWS course is designed for professionals who want to learn ML engineering on AWS. Through a balanced mix of theory, labs, and activities, participants gain practical experience with Amazon SageMaker AI and analytics tools such as Amazon EMR to develop robust, scalable, and production-ready ML applications.
Day 1 covers ML fundamentals on AWS, responsible ML, analyzing ML business challenges, data processing, exploratory data analysis, data transformation, and feature engineering. Day 2 looks at choosing a modeling approach, including SageMaker built-in algorithms and Autopilot, and then at training, evaluating, and tuning models and at model deployment strategies. On Day 3 participants learn how to secure ML resources, apply MLOps and automate deployment with CI/CD pipelines, and monitor model performance and data quality, including detecting data drift.
The course includes presentations, demonstrations, group exercises, and seven hands-on labs. Labs cover data preparation with Amazon SageMaker Data Wrangler and Amazon EMR, training and tuning models with SageMaker AI, shifting traffic, using SageMaker Pipelines and the SageMaker Model Registry, and monitoring a model for data drift.
The course is aimed at current and in-training machine learning engineers who may have little prior experience with AWS. DevOps engineers, developers, and SysOps engineers can benefit as well. It is recommended that attendees are familiar with basic machine learning concepts, have working knowledge of Python and libraries such as NumPy, Pandas, and Scikit-learn, and have a basic understanding of cloud computing and AWS. Experience with Git is beneficial but not required.