A company needs a serverless, internal-only chat interface for employees to search a large internal document knowledge base. Which combination of steps will satisfy these requirements?

The correct combination involves Bedrock, Lambda/API Gateway, and S3. Amazon Bedrock offers a serverless way to access foundation models like Anthropic Claude, suitable for the chat interface's AI capabilities. AWS Lambda and Amazon API Gateway provide a serverless architecture for the chat interface itself, handling requests and invoking the Bedrock model without managing servers. Amazon S3 is a cost-effective and scalable object storage service ideal for storing vector database dumps and embeddings, which are crucial for the knowledge base search. Incorrect options: SageMaker JumpStart with Llama is a valid option for foundation models, but Bedrock is generally preferred for fully managed, serverless access to a variety of FMs. EC2 instances introduce server management overhead, violating the "serverless" requirement. Amazon RDS for MySQL is a relational database, not optimized for storing vector embeddings, which are better suited for specialized vector databases or object storage like S3.