• Contents:

RAG - Basic Concepts

• RAG = Retrieval Augmented Generation
• 🔧 Allows an FM to reference data sources outside of its training data
  • Bedrock takes care of creating Vector Embeddings in the DB of your choice, i.e. a Knowledge Base, based on your data
  • Use where real-time data is needed to be fed into the FM
• Architecture diagram
• Answers to prompts will carry reference numbers

Knowledge Bases

• Data source documents are chunked, then processed by an embeddings model (creates vectors), then stored into a vector DB → knowledge base created
• ‼️ The embedding model doesn't need to be the same FM that uses RAG!
• Knowledge Base creation diagram
• RAG Vector DB types to learn for the exam:
  • Amazon OpenSearch Service – search & analytics DB
    • Real time similarity queries, store millions of vector embeddings
    • Scalable index management, and fast nearest-neighbor (kNN) search capability
    • 💡 Default to this DB unless you have reasons to choose another DB
  • Amazon DocumentDB [with MongoDB compatibility]
    • NoSQL database
    • Real time similarity queries, store millions of vector embeddings
  • Amazon Aurora – relational database, proprietary on AWS
  • Amazon RDS for PostgreSQL – relational database, open-source
  • Amazon Neptune – graph database
• RAG Data Sources
  • Amazon S3
  • Confluence
  • Microsoft SharePoint
  • Salesforce
  • Web pages (your website, your social media feed, etc…)
  • …etc (More added over time)

RAG Use cases

💡 Most typical use case is a chatbot with specific domain-knowledge, which can query external sources and knowledge bases

• Customer Service Chatbot
  • Knowledge Base – products, features, specifications, troubleshooting guides, and FAQs
  • RAG application – chatbot that can answer customer queries
• Legal Research and Analysis
  • Knowledge Base – laws, regulations, case precedents, legal opinions, and expert analysis
  • RAG application – chatbot that can provide relevant information for specific legal queries