A company wants to build, with as little effort as possible, a RAG application on Amazon Bedrock that ingests internal documents and answers based on their content. Which Bedrock feature is the BEST fit?

1 / 1
Select an answer
CorrectB

Explanation

Choosing the Bedrock feature for building a RAG that answers based on internal documents.

  • 1ingests internal documents and answers based on their contentSearch external data and use it in answers = RAG
  • 2with as little effort as possibleBedrock Knowledge Bases, which provides ingestion through search as a managed service
AIncorrect

Bedrock Guardrails

Bedrock Guardrails is a safety feature that suppresses harmful output and undesirable topics.

It is not a feature for building a RAG that answers based on internal documents, so it is incorrect.

BCorrect

Bedrock Knowledge Bases

Correct. Bedrock Knowledge Bases is a managed RAG-building feature that ingests internal documents and provides embedding, vector storage, and search. It lets you produce answers grounded in external data with little effort.

CIncorrect

SageMaker JumpStart

SageMaker JumpStart is a model hub that provides pretrained models and ready-made solutions.

It is not the Bedrock feature for building a managed RAG that answers based on internal documents, so it is incorrect.

DIncorrect

Amazon Q in QuickSight

Amazon Q in QuickSight is a feature that operates and creates BI dashboards in natural language.

It is not the Bedrock feature for building a RAG that answers based on internal documents, so it is incorrect.

Key Takeaway

Remember how the correct answer, 'Bedrock Knowledge Bases,' works.
・It provides, as a managed service, the ingestion, chunking, embedding, vector storage, and search of internal documents.
・Without building RAG from scratch, you can construct answers grounded in external data.
Its purpose differs from Guardrails (harmful suppression), JumpStart (model hub), and Q in QuickSight (BI).