Databricks-GenAI-Assoc Assembling and Deploying Apps Practice Question
An engineer is packaging a custom PyFunc model that wraps an open-source LLM and must deploy it to a Databricks Model Serving endpoint. The endpoint must load the model from Unity Catalog and expose it through a REST API. Which two actions are required to make the deployment succeed? (Choose two.)
⚠ Common exam trap
The trap here is focusing on endpoint scaling or compute choices, when deployment success actually hinges on declared dependencies and Unity Catalog permissions.
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
✓
Register the logged model in Unity Catalog and grant the serving identity USE CATALOG, USE SCHEMA, and EXECUTE or SELECT privileges on the model version.
A successful custom model deployment requires a correctly logged artifact with all dependencies declared, and a Unity Catalog registration with privileges granted to the serving identity. Environment build failures and permission errors are the two dominant causes of endpoint creation failure. Flavor conversion, attached clusters, and scale-to-zero settings are unrelated to whether the deployment itself can complete.
Answer analysis
Option-by-option breakdown
For each option: why learners choose it and why it is or isn't the right answer here.
- ✓
Register the logged model in Unity Catalog and grant the serving identity USE CATALOG, USE SCHEMA, and EXECUTE or SELECT privileges on the model version.
Why this is correct
Model Serving resolves the model by its Unity Catalog three-level name, and the endpoint's identity must be able to read that model version. Without the catalog, schema, and model privileges, endpoint creation fails with a permission error even though the artifact itself is valid. Granting these is mandatory for Unity Catalog-based serving.
- ✗
Set the endpoint's scale-to-zero configuration to disabled so the model stays in memory permanently.
Why it's wrong here
Scale-to-zero affects cold-start latency and cost, not whether the deployment succeeds. Disabling it is an availability and performance choice, and it does not address environment build or permission failures. The deployment can succeed with scale-to-zero enabled, so this is not a required action.
- ✗
Attach the model to a Databricks cluster that stays running so the endpoint can proxy requests to it.
Why it's wrong here
Model Serving provisions its own managed compute and does not proxy to an interactive cluster. Keeping a cluster running adds cost without enabling the endpoint, and the endpoint would not use it. Deployment succeeds or fails based on the logged artifact and permissions, not on an attached cluster.
- ✓
Log the model with mlflow.pyfunc.log_model, including a pip_requirements entry for every runtime dependency the wrapper imports.
Why this is correct
Serving rebuilds the environment from the logged artifact, so every imported library must be declared in pip_requirements or conda_env. Missing entries cause the endpoint to fail during environment build or model load. Explicitly pinning dependencies makes the deployment reproducible and is a prerequisite for a successful endpoint load.
- ✗
Convert the PyFunc wrapper to a scikit-learn estimator so the serving container recognizes the flavor.
Why it's wrong here
Model Serving supports the pyfunc flavor directly, so converting to scikit-learn is unnecessary and would likely break the wrapper's LLM logic. The pyfunc flavor is the standard way to serve arbitrary Python inference code, and no flavor conversion is required for a custom LLM wrapper to deploy.
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.