Building Retrieval Augmented Generation (RAG) workflows with Amazon OpenSearch Service

Retrieval Augmented Generation (RAG) is a powerful technique that enhances the accuracy and relevance of AI-generated responses by combining information retrieval with text generation. In this course, you will explore how to build RAG solutions using Amazon OpenSearch Service, a fully managed search and analytics solution. In this course you will discover how embeddings, which are vector representations of textual data, can significantly improve search and information retrieval capabilities. We will delve into the process of creating effective retrievers for RAG using the advanced features of Amazon OpenSearch Service.

• Course level: Intermediate
• Duration: 60 minutes

Activities

This course includes presentations, demonstrations, and assessments.

Course objectives

In this course, you will do the following:

• -Explain the workflow of large language models (LLMs) and identify their limitations. -Describe RAG and illustrate how it improves LLM performance. -Identify and outline the steps involved in building a knowledge base for RAG. -Compare different search methods that OpenSearch supports. -Demonstrate the use of Amazon OpenSearch Serverless with Amazon Bedrock Knowledge Bases to generate embeddings and retrieve relevant information.

Intended audience

This course is intended for the following job roles:

• Technical field community
• Analytics solutions architects (SAs)
• Database administrators
• Database engineers
• Data architects
• Solutions architects (SAs)
• Data platform engineers
• Search specialists

Prerequisites

We recommend that attendees of this course have the following:

• A basic knowledge of OpenSearch Service or Amazon OpenSearch Serverless on the AWS.
• Completed the Level 100 course on Skill Builder for OpenSearch Service, Getting Started with Amazon OpenSearch Service.
• Completed the Level 200 course on Skill Builder for OpenSearch Service, Empower Search with AI using Amazon OpenSearch Service.

Course outline

Module 1: Introduction to building Generative AI Applications

• Understanding RAG
• Building the Knowledge Base

Module 2: Building RAG Applications

• Using Amazon Bedrock Knowledge Bases

Module 3: Resources

• Learn More
• Contact Us