Question

A company is building a customer-support chatbot on Amazon Bedrock. They want every response to consistently adopt a polite, professional tone and to refuse to answer questions unrelated to their product line, without having to repeat these instructions in each user message. Which prompt engineering approach best achieves this?

Accepted Answer

Define a system prompt that sets the assistant's role, tone, and boundaries before the conversation begins

Answer

Increase the temperature parameter so the model produces more creative and polite responses

Answer

Reduce the number of tokens in each user message to save on inference costs

Answer

Fine-tune the foundation model on the company's full support ticket history before launch