GregLab | Exam Prep

Applications of Foundation Models

RAG Architecture

RAG connects retrieved source content with a foundation model response. The end-to-end flow ingests documents, chunks them, embeds chunks, stores vectors, embeds the user query at runtime, retrieves similar chunks, injects them into the prompt, and generates a grounded answer.

Concepts

  • RAG commonly ingests documents, chunks text, creates embeddings, stores vectors, retrieves relevant chunks, and sends context to a model.
  • RAG is usually preferred when knowledge changes frequently or must stay external to model weights.
  • Retrieval quality depends on chunking, metadata, embedding model choice, ranking, and source document quality.
  • At ingestion time, RAG loads documents, chunks them into passages, creates embeddings such as Amazon Titan Embeddings, and stores vectors in a vector database.
  • At query time, RAG embeds the user question, runs similarity search such as cosine similarity, retrieves the top matching chunks, injects them into the model prompt, and generates a grounded response.
  • Amazon OpenSearch Serverless is the default vector store for Amazon Bedrock Knowledge Bases, with other supported stores including Aurora PostgreSQL with pgvector, Pinecone, and Redis Enterprise Cloud.
  • Amazon OpenSearch Service is a managed provisioned search and analytics engine for log analytics, full-text search, monitoring, and vector search; Amazon OpenSearch Serverless auto-scales with pay-per-use capacity and native Bedrock Knowledge Bases integration.

Exam tips

  • RAG is preferred when private knowledge changes or should not be baked into model weights.
  • Chunking, metadata, embedding model choice, and ranking affect retrieval quality.
  • Bedrock Knowledge Bases provide managed RAG over supported data sources and vector stores.
  • For query-time RAG, embed the user question before similarity search, then pass retrieved chunks into the FM prompt.
  • OpenSearch Serverless is the default Bedrock Knowledge Bases vector store; do not confuse it with provisioned Amazon OpenSearch Service clusters.
  • Choose Amazon OpenSearch Service when you need managed cluster capacity control; choose OpenSearch Serverless for auto-scaling, no capacity planning, and Bedrock Knowledge Bases RAG integration.

Free AWS Certified AI Practitioner prep

Build focused AIF-C01 quizzes from exam domains, topics, and AWS services.

Practice with exam-style multiple-choice and multiple-response questions, clearly labeled supplemental exercises, score breakdowns, explanations, and a compact reference for this lane's official exam domains.

Build a quiz

Exam Weights

Quiz builder

Choose your practice set

Mode

Exam fidelity: AWS lists multiple choice and multiple response for this exam. Ordering, matching, and case-study items are supplemental learning exercises; their results stay in overall study accuracy but do not count toward exam-style accuracy. Difficulty labels describe this site's scenario complexity, not an AWS-published question rating.

Reference

AIF-C01 topics and service map

Study links

AIF-C01 resources