How do you configure AI Gateway inference tables?

By configuring a separate logging service like Prometheus

By manually creating a Delta table and inserting logs

By using the MLflow tracking API to log requests

By enabling inference logging on the AI Gateway endpoint

What type of data is stored in AI Gateway usage tables?

Only the text of the requests and responses

The model's training loss and accuracy

Token usage, costs, and other consumption metrics

The configuration of the AI Gateway endpoint

What is the purpose of configuring AI Gateway inference tables?

To store the model's training data for fine-tuning

To manage the routing of requests to different models

To enforce rate limits on the gateway

To log and track LLM or agent requests and responses

A data science team has enabled inference logging for a chat model serving endpoint. They want to create a daily dashboard that shows the average latency and the distribution of response lengths for requests that contain a specific prompt template. Which approach is most efficient?

Query the inference table with a SQL query that filters on the prompt template and aggregates latency and output token counts.

Write a custom script that calls the endpoint and logs the metrics to a separate table.

Enable Agent Monitoring and use its built-in latency charts.

Use the AI Gateway usage table to get latency and response length metrics.

What is the primary purpose of an inference table for a live LLM endpoint?

To manage the scaling of the endpoint based on traffic

To store the model's configuration and environment variables

To track and analyze the behavior of the live endpoint by storing request and response data

To store the model's training data for future fine-tuning

DatabricksCertified Generative AI Engineer Associate

Domain 6Objective 2

Monitoring and Logging for Deployed Applications GENERATIVE-AI-ENGINEER-ASSOCIATE Practice Questions (Page 3)

Part of the Section 6: Evaluation and Monitoring domain, which makes up ~17% of our current practice bank. Databricks does not publish an official question count, but from its 90-minute exam (~35–60 total, ~6–10 in this domain), expect 1–2 from this objective — we provide 25 practice questions to prepare you well beyond it. (estimate)

25questions here

5free pages

6concepts

Questions 11–15

11foundation · easy
How do you configure AI Gateway inference tables?
Select an answer first
12foundation · easy
What type of data is stored in AI Gateway usage tables?
Select an answer first
13foundation · easy
What is the purpose of configuring AI Gateway inference tables?
Select an answer first
14application · medium
A data science team has enabled inference logging for a chat model serving endpoint. They want to create a daily dashboard that shows the average latency and the distribution of response lengths for requests that contain a specific prompt template. Which approach is most efficient?
Select an answer first
15foundation · easy
What is the primary purpose of an inference table for a live LLM endpoint?
Select an answer first

Finished these 5 questions?

Review the revealed explanations, or continue through the curriculum.

Next 5 questions →Back to GENERATIVE-AI-ENGINEER-ASSOCIATE

Free Basic Practice is a study aid with revealable answers — not a scored exam. Examers.io is independent and not affiliated with or endorsed by Databricks. “GENERATIVE-AI-ENGINEER-ASSOCIATE” is a trademark of its owner, used for identification only.