Question

A financial services company is building a customer-facing assistant on Amazon Bedrock. During testing, the model gives technically correct answers but uses casual language and occasionally references topics outside the company's investment products. The team wants responses that consistently adopt the persona of a professional, compliance-aware financial advisor who only discusses the company's offerings. Which prompt engineering approach BEST achieves this with minimal engineering effort?

Accepted Answer

Add a system prompt that defines the assistant's role as a professional financial advisor and instructs it to only discuss the company's investment products.

Answer

Fine-tune the foundation model on thousands of labeled advisor conversations to change its tone.

Answer

Lower the temperature parameter to 0 so the responses become more formal and on-topic.

Answer

Increase the max tokens setting so the model has more room to produce a professional response.