Databricks-GenAI-Assoc Evaluation and Monitoring Practice Question
A GenAI engineer notices that a RAG agent's retrieval stage is returning relevant chunks, but the generated answers frequently omit key facts present in those chunks. The team wants a single evaluation metric that isolates whether the generator is using the provided context. Which metric should they focus on?
⚠ Common exam trap
The trap here is choosing answer correctness because the final output is wrong, when the diagnostic question is which pipeline stage is failing and only groundedness isolates the generator's use of context.
Answer choices
Why each option matters
Answer the question above first, then reveal the full breakdown to understand why each option is right or wrong.
Correct answer & explanation
✓
Groundedness, which checks whether claims in the answer are supported by the retrieved context.
Groundedness is the metric that inspects whether the answer's claims are supported by the retrieved context. Since the retriever is already delivering relevant chunks, a low groundedness score points squarely at the generator ignoring or misusing that evidence, giving the team a precise diagnostic for the stage that is failing.
Answer analysis
Option-by-option breakdown
For each option: why learners choose it and why it is or isn't the right answer here.
- ✗
Answer correctness against a reference answer.
Why it's wrong here
Answer correctness compares the final response to a ground-truth answer. It would drop in this scenario, but it conflates retrieval and generation failures and does not isolate whether the generator ignored context. The team needs a diagnostic that pinpoints the generation stage specifically.
- ✓
Groundedness, which checks whether claims in the answer are supported by the retrieved context.
Why this is correct
Groundedness evaluates each claim in the generated answer against the retrieved context. When retrieval is good but the answer omits supported facts or invents unsupported ones, groundedness drops. It isolates the generator's use of context, which is exactly the failure described.
- ✗
p95 latency of the serving endpoint.
Why it's wrong here
Latency measures how long responses take, not whether they use retrieved context correctly. A fast answer that omits key facts would still show excellent latency. This metric is operational and cannot diagnose a generation-quality problem, so it is irrelevant to the scenario.
- ✗
Context recall of the retriever.
Why it's wrong here
Context recall measures whether the retriever surfaced all relevant information. The scenario states retrieval is already returning relevant chunks, so this metric would look healthy and would not reveal the generator's failure to use them. It diagnoses the wrong stage of the pipeline for this symptom.
About these practice questions
This Databricks-GenAI-Assoc question is part of Courseiva's 330-question bank — original exam-style content with full explanations and wrong-answer analysis, never real exam questions or exam dumps. Learn why practice questions differ from exam dumps →
JA
Written and reviewed by Johnson Ajibi, MSc IT Security
Senior Network & Security Engineer · founder of Courseiva
Last reviewed September 2026 · checked against the official Databricks exam blueprint
This Databricks-GenAI-Assoc practice question is part of Courseiva's free Databricks certification practice question bank. Courseiva provides original exam-style practice questions with explanations, topic-based practice, mock exams, readiness tracking, and study analytics to help learners prepare for the Databricks-GenAI-Assoc exam.