Question 29
Domain 3: Applications of Foundation ModelsA company wants to improve its chatbot's responses to match the company's desired tone. The company has 100 examples of high-quality conversations between customer service agents and customers. The company wants to use this data to incorporate company tone into the chatbot's responses. Which solution meets these requirements?
Correct answer: D
Explanation
Amazon Bedrock fine-tuning uses labeled examples to adapt a foundation model to a specific style or behavior, which fits using 100 high-quality conversations to match the company’s tone. It is designed to “incorporate company tone into the chatbot’s responses” by training the model on the company’s own examples.
Why each option is right or wrong
A. Use Amazon Personalize to generate responses.
Amazon Personalize is for recommendations and personalization, not chatbot response generation.
B. Create an Amazon SageMaker HyperPod pre-training job.
SageMaker HyperPod supports large-scale model training infrastructure, not supervised tone adaptation from examples.
C. Host the model by using Amazon SageMaker. Use TensorRT for large language model (LLM)
SageMaker hosting with TensorRT is a deployment optimization path, not a training method for company tone.
D. Create an Amazon Bedrock fine-tuning job.
Amazon Bedrock fine-tuning is the documented mechanism for adapting a foundation model to domain-specific behavior using a smaller, curated dataset; AWS states that fine-tuning updates the base model on a smaller task-specific corpus rather than requiring pre-training from scratch. With only 100 high-quality agent-customer conversations, the company has exactly the kind of labeled examples used in a custom fine-tuning job to shift response style and tone, whereas retrieval or prompt-only approaches do not change the model’s underlying behavior.