CoursesFree Quizzes
NVIDIAKnowledge Integration and Data Handling

RAG & Cost Optimization — NCP-AAI Practice Question

A representative NVIDIA-Certified Professional Agentic AI (NCP-AAI) exam question on RAG & Cost Optimization. Work through it below, then read why each option is right or wrong.

This question is part of the free NVIDIA-Certified Professional Agentic AI practice test.

Short answer

The correct answer is D. Pre-index all historical outage records, billing disputes, and performance logs into a vector database during off-peak periods; use dynamic batching and caching during peak seasons to reuse embeddings for common queries..

Pre-indexing during off-peak periods amortizes embedding costs and leverages caching to handle seasonal spikes efficiently—essential under a cost ceiling. Dynamic batching and embedding reuse are standard RAG optimization techniques documented in NVIDIA retrieval guidance. On-demand embedding is cost-prohibitive during peak load. Full-text search sacrifices semantic quality needed for nuanced customer issues. Separate pipelines per data type multiply infrastructure and embedding overhead rather than reduce it.

The Question

A telecommunications provider is building a retrieval pipeline to support customer service agents explaining service outages, billing discrepancies, and network performance issues. The provider faces a cost ceiling and experiences seasonal demand spikes (peak during winter storms, holidays). What is the most cost-effective data preparation strategy for this scenario?

AMaintain separate retrieval pipelines for each data type (outages, billing, performance) with independent embeddings to maximize parallelism.
BStore raw documents in a data lake and perform full-text search only, avoiding embedding costs entirely but sacrificing semantic retrieval quality.
CEmbed and index data on-demand for every query, ensuring freshness but accepting higher compute costs during peak demand.
DPre-index all historical outage records, billing disputes, and performance logs into a vector database during off-peak periods; use dynamic batching and caching during peak seasons to reuse embeddings for common queries.Correct

Why D is correct

Pre-indexing during off-peak periods amortizes embedding costs and leverages caching to handle seasonal spikes efficiently—essential under a cost ceiling. Dynamic batching and embedding reuse are standard RAG optimization techniques documented in NVIDIA retrieval guidance. On-demand embedding is cost-prohibitive during peak load. Full-text search sacrifices semantic quality needed for nuanced customer issues. Separate pipelines per data type multiply infrastructure and embedding overhead rather than reduce it.

Why the other options are wrong

Option A: Maintain separate retrieval pipelines for each data type (outages, billing, performance) with independent embeddings to maximize parallelism.

Option A does not satisfy the requirement in the scenario. Review the explanation above: the correct choice (D) is the only one that fully meets every constraint stated in the question.

Option B: Store raw documents in a data lake and perform full-text search only, avoiding embedding costs entirely but sacrificing semantic retrieval quality.

Option B does not satisfy the requirement in the scenario. Review the explanation above: the correct choice (D) is the only one that fully meets every constraint stated in the question.

Option C: Embed and index data on-demand for every query, ensuring freshness but accepting higher compute costs during peak demand.

Option C does not satisfy the requirement in the scenario. Review the explanation above: the correct choice (D) is the only one that fully meets every constraint stated in the question.

Key idea: RAG & Cost Optimization

Pre-indexing during off-peak periods amortizes embedding costs and leverages caching to handle seasonal spikes efficiently—essential under a cost ceiling. Dynamic batching and embedding reuse are standard RAG optimization techniques documented in NVIDIA retrieval guidance. On-demand embedding is cost-prohibitive during peak load. Full-text search sacrifices semantic quality needed for nuanced customer issues. Separate pipelines per data type multiply infrastructure and embedding overhead rather than reduce it. On the NCP-AAI exam, questions in the "Knowledge Integration and Data Handling" domain test whether you can map a scenario's constraints to the right choice. Read the requirement carefully, eliminate options that violate any single constraint, and pick the one that satisfies all of them with the least operational overhead.

Ready to see how you'd score?

Take the free practice quiz and find out which NVIDIA-Certified Professional Agentic AI domains you need to focus on. No signup required.

Take the free quizSee all 213+ practice questions

Practice 5 similar questions

Same cert, same or adjacent domain. Use these after reviewing the explanation.

1Agent Architecture & Router Agent — NCP-AAI Practice QuestionAgent Architecture and Design2Approval Controls & Cost Management — NCP-AAI Practice QuestionHuman-AI Interaction and Oversight3Agent Integration & Tool Registry — NCP-AAI Practice QuestionAgent Development4Measurement & Release Strategy — NCP-AAI Practice QuestionEvaluation and Tuning5Explainability & Audit Trail — NCP-AAI Practice QuestionDeployment and Scaling

Related NVIDIA-Certified Professional Agentic AI Practice Questions

Agent Architecture & Router Agent — NCP-AAI Practice Question→Approval Controls & Cost Management — NCP-AAI Practice Question→Agent Integration & Tool Registry — NCP-AAI Practice Question→

Official sources used

Verify program and exam details at the source before scheduling.

docs.nvidia.comFunctions And Function Groups: Functions.htmldocs.nvidia.comDocumentation: Agents