nerdexam
Databricks

GENERATIVE-AI-ENGINEER-ASSOCIATE · Question #3

A small and cost-conscious startup in the cancer research field wants to build a RAG application using Foundation Model APIs. Which strategy would allow the startup to build a good-quality RAG…

The correct answer is B. Pick a smaller LLM that is domain-specific. Choosing a smaller, domain-specific LLM (option B) is the optimal balance of cost and quality for a specialized use case like cancer research. Domain-specific models are pre-trained or fine-tuned on relevant medical/scientific data, so they perform better on niche queries than…

Generative AI Solution Design and Optimization

Question

A small and cost-conscious startup in the cancer research field wants to build a RAG application using Foundation Model APIs. Which strategy would allow the startup to build a good-quality RAG application while being cost- conscious and able to cater to customer needs?

Options

  • ALimit the number of relevant documents available for the RAG application to retrieve from
  • BPick a smaller LLM that is domain-specific
  • CLimit the number of queries a customer can send per day
  • DUse the largest LLM possible because that gives the best performance for any general queries

How the community answered

(32 responses)
  • A
    13% (4)
  • B
    78% (25)
  • C
    6% (2)
  • D
    3% (1)

Explanation

Choosing a smaller, domain-specific LLM (option B) is the optimal balance of cost and quality for a specialized use case like cancer research. Domain-specific models are pre-trained or fine-tuned on relevant medical/scientific data, so they perform better on niche queries than large general-purpose models - often at a fraction of the inference cost. Option A (limiting documents) degrades quality. Option C (limiting queries) harms user experience and doesn't reduce per-query cost. Option D (largest LLM) maximizes cost and is overkill for a specialized domain where a smaller domain-tuned model can outperform it.

Topics

#RAG#LLM Selection#Cost Optimization#Domain Specificity

Community Discussion

No community discussion yet for this question.

Full GENERATIVE-AI-ENGINEER-ASSOCIATE Practice