This session, we focus on common use cases and design patterns for predictive analytics using Amazon EMR. We address accessing data from a data lake, extraction and preprocessing with Apache Spark, analytics and machine learning code development with notebooks (Jupyter, Zeppelin), and data visualization using Amazon QuickSight. We cover other operational topics, such as deployment patterns for ad hoc exploration and batch workloads using Spot and multi-user notebooks. The intended audience for this session includes technical users who are building statistical and data analytics models for the business using tools, such as Python, R, Spark, Presto, Amazon EMR, Notebooks.