A team needs a foundation model to adopt a consistent brand tone and domain terminology in its responses. They want results fast without training infrastructure. What should they try FIRST?
A.Fine-tuning the model on brand documents
Fine-tuning works but is slower, costlier — try prompting first.
B.Prompt engineering with style instructions and few-shot examples
Correct. Instructions + examples shape style with zero training.
C.Continued pre-training from scratch
Pre-training is a massive undertaking reserved for base-model creation.
D.Deploying a larger model
A bigger model doesn't inherently know your brand voice.
Why this answer
Prompt engineering — system prompts with style instructions plus few-shot examples — is the fastest, cheapest lever for tone/terminology and requires no training. Fine-tuning is the escalation path when prompting can't capture the pattern.
Correct answer: B · Last reviewed 2026-10-26
Official AWS references
Practice this interactively
Answer without seeing the solution first in the Generative AI & Bedrock practice session.
Practice Generative AI & Bedrock