AIP-C01 AWS Certified Generative AI Developer – Professional Free Practice Questions and Answers

A multilingual support assistant searches millions of document embeddings and filters by date, agency, and document type. The team wants managed vector search integrated with Bedrock Knowledge Bases. What is a strong fit?

• A. OpenSearch Serverless vector search with metadata filtering
• B. A single EC2 instance with local text files
• C. Route 53 Resolver with forwarding rules
• D. CloudTrail event history as the embedding store

A RAG application misses exact technical terms while returning semantically similar but irrelevant passages. Which retrieval approach can improve both exact matching and semantic recall?

• A. Hybrid retrieval that combines lexical search with vector search and reranking
• B. Increase the language model temperature
• C. Remove all document metadata before indexing
• D. Use only a larger foundation model without changing retrieval

A sensitive application in private subnets must call Bedrock without sending traffic through the public internet. Access should be limited to approved endpoints and roles. What should the team configure?

• A. Bedrock interface VPC endpoints with scoped IAM policies and endpoint conditions
• B. A public subnet and an unrestricted internet gateway
• C. A public S3 bucket containing model credentials
• D. A CloudFront distribution in front of the Bedrock API

Many teams share prompt templates across regions. Administrators require version history, review approval, and an audit trail of prompt management actions. Which approach best supports these needs?

• A. Use Bedrock Prompt Management versioning and governance workflows, and record API activity with CloudTrail
• B. Store prompt text in local files and email versions for review
• C. Use a CloudWatch alarm as the prompt version database
• D. Put prompts in an unversioned public S3 bucket

A support tool must display Bedrock responses progressively to thousands of concurrent users. The application needs a persistent, bidirectional connection for response chunks. Which design fits?

• A. A WebSocket API integrated with a streaming backend that forwards generated chunks
• B. A synchronous REST request that waits for the full response
• C. A daily batch job that stores responses in S3
• D. A CloudWatch dashboard embedded in the application