AIP-C01 Domain 2: RAG and Knowledge Integration

A team must store embeddings in vector database. Which option meets the requirement?

Multiple choice — select one answer

Answer choices

  1. a. Sync jobs and S3 event-driven ingestion
  2. b. OpenSearch Serverless vector engine or Aurora pgvector
  3. c. Fixed-size vs hierarchical chunking in KB
  4. d. RAG evaluation metrics and human review

Architectural breakdown & explanation

Vector DBs enable similarity search.

Correct answer(s): b. OpenSearch Serverless vector engine or Aurora pgvector

Exam domain: Domain 2: RAG and Knowledge Integration (24% weight on the official guide).

Official documentation & study links

Practice this certification

This page is one question from the AWS Certified Generative AI Developer – Professional bank. Run a full timed practice exam with domain-weighted random questions in your browser.