AWS Certified AI Practitioner AIF-C01 Practice Question

Knowledge Bases for Amazon Bedrock provides a managed Retrieval-Augmented Generation (RAG) workflow. After you load the policy documents into S3 and configure a vector store, you can invoke the RetrieveAndGenerate API. Bedrock automatically retrieves relevant chunks, appends them to the prompt, and calls the chosen foundation model, all within the same managed operation. Amazon Kendra offers search but does not invoke a model, SageMaker fine-tuning requires custom code for retrieval, and using Neptune with Lambda lacks an integrated RAG pipeline.