AWS

Learn how to seamlessly integrate machine learning capabilities into your SQL queries using Amazon Athena ML and Amazon SageMaker. This hands-on workshop will guide you through building and deploying ML models with Amazon SageMaker, and then using those models directly from your SQL queries in Amazon Athena.

You'll discover how to perform ML inference without moving your data out of your data lake, significantly simplifying your architecture and accelerating time to insights. Through practical exercises, you'll learn to implement both built-in ML functions and custom ML models in your SQL workflows.

Workshop Agenda:

• Introduction and Setup
  • Overview of Amazon Athena ML capabilities
  • Understanding SageMaker integration patterns
  • Environment setup and configuration
• Hands-on Lab 1: Built-in ML Functions
  • Using Athena's built-in ML functions
  • Implementing common ML use cases
  • Best practices for performance optimization
• Hands-on Lab 2: Custom ML Models
  • Deploying SageMaker endpoints
  • Integrating custom models with Athena
  • Real-time inference in SQL queries
• Advanced Topics and Best Practices
  • Cost optimization strategies
  • Security considerations
  • Production deployment patterns
• Q&A and Wrap-up

Target Audience:

• Data Engineers and Analysts who want to incorporate ML capabilities into their SQL workflows
• ML Engineers looking to simplify model deployment and inference
• Data Scientists interested in making their models more accessible to SQL users
• Solutions Architects designing ML-enhanced analytics solutions

Requirements:

• Bring your own laptop [no laptop will be provided onsite]
• Basic SQL knowledge
• Familiarity with AWS Console
• Understanding of basic ML concepts (helpful but not required)
• AWS account with necessary permissions (will be provided during the workshop)

What You'll Learn:

• How to use Athena ML to perform inference on your data lake
• Techniques for integrating SageMaker endpoints with Athena
• Best practices for security and performance optimization
• Real-world patterns for production deployment
• Cost optimization strategies for ML inference at scale