Databricks-GenAI-Assoc Application Development Practice Question
Exhibit
{
"model_name": "customer_support_llm",
"task": "llm/v1/chat",
"endpoint_config": {
"auto_capture_request_payload": false,
"traffic_config": {
"routes": [{"served_model_name": "v1", "traffic_percentage": 100}]
}
}
}Refer to the exhibit. A developer wants to enable monitoring for their deployed LLM endpoint. Given the current configuration, what must the developer change to ensure that request/response logs are captured for analysis?
⚠ Common exam trap
Candidates frequently attempt to create custom logging logic or external monitoring scripts, missing the built-in configuration flag 'auto_capture_request_payload' which is specifically designed for this purpose.
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
✓
Update 'auto_capture_request_payload' to true.
The current configuration has 'auto_capture_request_payload' set to false, which prevents the logging of inference traffic. By updating this flag to true, the system will start capturing payloads into a Delta table. This is critical for monitoring model performance, drift, and quality. Enabling this feature allows data teams to perform retrospective analysis, which is essential for iterating on model prompts and fine-tuning configurations based on real user interactions.
Answer analysis
Option-by-option breakdown
For each option: why learners choose it and why it is or isn't the right answer here.
- ✗
Change the task to 'llm/v1/completions'.
Why it's wrong here
Changing the task type does not enable logging. The 'llm/v1/chat' task is correct for conversational models. Modifying the task type would likely cause a mismatch between the expected schema and the model architecture, potentially breaking the endpoint and failing to address the primary goal of enabling payload capture.
- ✓
Update 'auto_capture_request_payload' to true.
Why this is correct
Setting 'auto_capture_request_payload' to true instructs the inference service to log incoming requests and outgoing responses to a managed Delta table. This is the direct configuration setting required to enable observability, ensuring that all data passed to the model is stored for later quality assessment and monitoring purposes.
- ✗
Add a new route with 0% traffic to the 'traffic_config'.
Why it's wrong here
Traffic configuration manages version routing and A/B testing, not observability. Adding a route with zero percent traffic does not enable logging functionality. It merely creates an inactive path to a model version, which is irrelevant to the requirement of capturing inference logs for the currently active production model.
- ✗
Increase the 'traffic_percentage' to 200%.
Why it's wrong here
Traffic percentages must sum to 100%. Setting the value to 200% is an invalid configuration that would likely be rejected by the API validation logic. Furthermore, traffic allocation has no correlation with the logging or monitoring of request payloads, which is controlled solely by the capture settings provided.
About these practice questions
One of 330 original Databricks-GenAI-Assoc practice questions on Courseiva, each with a full explanation and wrong-answer analysis — not exam dumps or protected exam content. 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.