Implementing Generative AI on AWS

This unit focuses on practical implementation and customization of generative AI using AWS services, centering on Amazon Bedrock, Amazon Q, and supporting tools. Students will learn model evaluation, prompt and flow management, RAG strategies, agents, knowledge bases, guardrails, and security practices to prepare models for production.

Learning Objectives

• Analyze AWS generative AI services (including Amazon Bedrock and Amazon Q) and supporting components to select appropriate architectures for given application requirements
• Apply model customization and prompt-management techniques—such as embeddings, fine-tuning strategies, prompt engineering, Bedrock Flows, and GraphRAG—to improve task-specific performance
• Demonstrate setup and configuration of Bedrock features, including prompt management, flows, agents, knowledge bases, and guardrails while implementing recommended security and data protection practices
• Evaluate foundation model performance and deployment readiness using model evaluation tools, RAG approaches, monitoring strategies, and documented acceptance criteria to prepare for production deployment