Databricks-ML-Assoc Practice Test 18 — 15 Questions

Question 1

A machine learning engineer is deploying a model to a Databricks Model Serving endpoint and needs to send feature values that are computed by a separate upstream pipeline. The engineer wants the endpoint to accept a JSON payload describing a single record with named fields. Which approach correctly describes how the client should format the request?

Choose an answer to begin — your selection is scored in the full session.

Accepted Answer

Send a JSON body with a dataframe_records or dataframe_split field containing the named columns matching the model signature.. For models logged with a pandas DataFrame signature, Databricks Model Serving accepts JSON using dataframe_records or dataframe_split, where each record is an object of column names to values. This preserves the named-column contract and allows signature validation. Positional array formats and non-JSON encodings do not match the DataFrame interface and would either fail validation or misalign features.

Answer

Send a JSON body containing a base64-encoded serialized pandas DataFrame along with a pickle content type.

Answer

Send the record as form-encoded key-value pairs in the request body with a content type of application/x-www-form-urlencoded.

Answer

Send a JSON body with an inputs field containing a nested array of numeric values in the exact column order of training.

Question 2

A data scientist has a custom Python model wrapped in an MLflow pyfunc flavor and needs to serve it on Databricks Model Serving. The model's preprocessing requires a library that is not part of the default serving environment. What is the correct way to make that dependency available to the endpoint?

Accepted Answer

Log the model with the required library listed in the model's conda environment or pip requirements so it is captured in the model artifact.. Model Serving reconstructs the runtime environment from the dependency specification stored with the logged model. To bring in a library beyond the default environment, the dependency must be captured in the model's conda environment or pip requirements when the model is logged, which guarantees the endpoint installs it reproducibly. Cluster-level installs and storage paths do not reach the serving container.

Answer

Upload the library wheel to DBFS and reference its path in the endpoint's environment variables.

Answer

Add the library as a cluster library on the all-purpose cluster where the model was trained.

Answer

Install the library interactively on the driver node before creating the endpoint.

Question 3

A team has just registered a new version of a fraud detection model in the workspace Model Registry. Before routing production traffic to it, they want to send a small percentage of live requests to the new version and compare its predictions against the current production model. Which Databricks Model Serving feature should they use?

Accepted Answer

Traffic splitting between served entities. Traffic splitting on a Databricks Model Serving endpoint allows multiple served entities, each pointing to a different model version, to share incoming requests by percentage. This makes it possible to route a small fraction of live traffic to a new version and compare its behavior against the existing production model before promoting it fully. Other listed features address logging, capacity, or naming, not request routing.

Answer

Inference tables

Answer

Provisioned throughput

Answer

Model aliases

Question 4

A team has registered a model in Unity Catalog and wants to serve it with Databricks Model Serving. Their security policy requires that the endpoint access the model without embedding long-lived credentials, and they want the endpoint to use a dedicated service principal for accessing downstream feature tables. Which configuration should they apply to meet these requirements?

Accepted Answer

Grant the endpoint's service principal access to the model and feature tables via Unity Catalog privileges, and configure the endpoint to use that service principal.. Unity Catalog allows Databricks Model Serving endpoints to run under a service principal identity, with access governed by Unity Catalog privileges on models and tables. Granting the service principal `EXECUTE` on the registered model and `SELECT` on feature tables enables secure, credential-free access. This design avoids embedding long-lived tokens and provides auditable, least-privilege access for production inference workloads.

Answer

Mount the feature tables as a DBFS path and configure the model to read them using a cluster-scoped credential passthrough token.

Answer

Enable token-based authentication on the endpoint and pass the token in each inference request header from the calling application.

Answer

Configure the Serving endpoint with a personal access token stored in the endpoint's environment variables.

Question 5

An ML engineer needs to give an external application a stable HTTPS URL to call a registered model served by Databricks Model Serving. The application must authenticate with a token and must not be able to modify the endpoint configuration. Which approach best meets these requirements?

Accepted Answer

Grant the external application a service principal with a token and assign it only the permissions required to query the serving endpoint.. Model Serving endpoints are invoked over a stable HTTPS URL using bearer-token authentication. Issuing the external application a service principal token scoped to query the endpoint satisfies the authentication requirement while preventing configuration changes, since the principal lacks management permissions. Sharing an admin token over-privileges the client, and the remaining choices either bypass the endpoint or remove authentication entirely.

Answer

Expose the endpoint through a public network with no authentication so the external application can call it directly with the URL alone.

Answer

Embed the model's artifacts in the external application and call the model locally instead of using the serving endpoint.

Answer

Share the serving endpoint's URL and provide the external application with a Databricks personal access token belonging to a workspace admin.

Question 6

A data scientist has registered a scikit-learn model in Unity Catalog as `ml_prod.churn.model_v3` and wants the Databricks Model Serving endpoint to automatically pick up newly registered model versions as they are promoted to the `champion` alias. Which configuration should the data scientist use when creating the serving endpoint?

Accepted Answer

Serve the model using the alias-qualified path `ml_prod.churn.model_v3@champion`.. Alias-qualified paths in Unity Catalog let a serving endpoint follow a movable pointer. When a new model version is registered and the `champion` alias is reassigned to it, the endpoint resolves the alias and begins serving the new artifact. This enables zero-touch promotion workflows, keeps governance in Unity Catalog, and avoids endpoint recreation or explicit version pinning.

Answer

Export the model with `mlflow.sklearn.save_model` to a DBFS path and point the endpoint at that directory.

Answer

Serve the model by its explicit version number, `ml_prod.churn.model_v3/3`, and re-create the endpoint after each promotion.

Answer

Serve the model by its registered name only, `ml_prod.churn.model_v3`, and rely on Databricks to always use the latest version.

Question 7

A team queries a Databricks Model Serving endpoint through the serving client and receives an error indicating the endpoint is not ready. They confirmed the endpoint exists. Which condition most directly explains why requests fail until it clears?

Accepted Answer

The endpoint is in a state such as Updating or Not Ready while its configuration is being applied or a replica is still building.. When an endpoint is being created or its configuration is being applied, it passes through transitional states and cannot serve requests until at least one replica is ready. A not-ready error for an endpoint known to exist points to that provisioning or updating state, which resolves on its own once the deployment completes and the endpoint reports Ready.

Answer

The caller's personal access token has expired.

Answer

The model version referenced by the endpoint has been deleted from the registry.

Answer

The serving client is using the wrong workspace URL for the endpoint.

Question 8

A data scientist is using Databricks Feature Store to create a training dataset for a model. They define a feature table with a primary key and a timestamp column. They then create a training set using create_training_set with the feature table and a label DataFrame. They notice that the training set contains null values for some features, even though the feature table has no nulls. What is the most likely reason for the nulls in the training set?

Accepted Answer

The label DataFrame contains timestamps that are earlier than the earliest timestamp in the feature table, so no feature values exist for those times.. In point-in-time feature lookups, Feature Store retrieves the most recent feature values as of the label timestamp. If the label timestamp precedes all feature timestamps for a given primary key, no feature value exists, resulting in nulls. This occurs when the feature table's time range does not cover the label events' times, often because the feature data starts later than the labels.

Answer

The timestamp column in the feature table is not sorted in ascending order, causing point-in-time lookups to fail and return nulls.

Answer

The feature table was created without specifying a timestamp column, so Feature Store cannot perform point-in-time lookups and returns nulls.

Answer

The primary key columns in the feature table and label DataFrame do not match in data type, causing join failures and nulls.

Question 9

An ML engineer is building an automated CI/CD pipeline that, after validating a new model version, must programmatically move it to the `champion` alias in Unity Catalog so the production Model Serving endpoint begins using it. The pipeline runs in a Databricks job using a service principal. Which action accomplishes the promotion programmatically?

Accepted Answer

Call the MLflow client method `set_registered_model_alias` with the model name, alias `champion`, and the target version, using a client connected to the Databricks registry URI.. Programmatic promotion in Unity Catalog is done through the MLflow client's alias APIs: `set_registered_model_alias` moves the alias to the validated version, and any serving endpoint resolving that alias adopts the change. Endpoint inference calls, custom Delta lookup tables, and delete-and-recreate maneuvers do not perform an atomic alias update and either fail outright or introduce availability risk.

Answer

Write the model version number into a Delta table that the endpoint reads at request time to decide which version to load.

Answer

Delete the current `champion` alias and recreate it by registering the model again under the same name with a new version number.

Answer

Invoke the endpoint's REST API with an update payload that includes the new model version number, forcing the endpoint to reload the model.

Question 10

A team is preparing to deploy a registered MLflow model to a Databricks Model Serving endpoint. They want to capture every request and response for later monitoring and debugging, and they also want the endpoint to remain available during a rolling model version update. (Choose two.)

Accepted Answer

Enable inference tables on the serving endpoint.. Capturing every request and response is achieved by enabling inference tables, which log payloads to a Delta table. Keeping the endpoint available during a model version change is achieved by updating the served version in place, which triggers a rolling update rather than endpoint deletion. Together these satisfy both the observability and availability requirements without downtime or manual log plumbing.

Answer

Enable verbose logging in the MLflow model's Python code and write logs to DBFS.

Answer

Delete the existing endpoint and create a new one pointing at the new model version.

Answer

Set the endpoint's `min_instances` to 0 to reduce cost during the update.

Question 11

A team wants to monitor a production Model Serving endpoint for data drift and to capture the exact request payloads and responses for later auditing. They want this logging to happen automatically without adding code to the client application. What should they configure on the endpoint?

Accepted Answer

Enable inference tables on the endpoint so requests and responses are automatically logged to a Delta table for monitoring and auditing.. Inference tables are the native Model Serving feature that logs request and response payloads to a Delta table automatically, enabling drift monitoring and auditing without touching client code. Client middleware, container logs, and MLflow experiments either require application changes, lack payload structure, or are not built for high-volume inference logging.

Answer

Attach an MLflow experiment to the endpoint so each prediction is recorded as a run with parameters and metrics.

Answer

Add client-side logging middleware in the application that writes each request and response to a Delta table before and after calling the endpoint.

Answer

Configure the endpoint to write its container stdout and stderr to a log delivery location in cloud storage.

Question 12

A team is standing up a real-time Databricks Model Serving endpoint for a fraud model. Requests will carry several numeric features, and the team wants the endpoint to reject malformed payloads with a clear client error rather than silently scoring them, and to avoid cold-start latency during business hours. Which two actions should the team take? (Choose two.)

Accepted Answer

Configure a non-zero minimum number of serving instances so replicas stay warm.. Declaring an MLflow model signature gives the endpoint a schema contract so invalid payloads are refused before scoring, and holding a minimum instance count above zero keeps replicas loaded so requests are not delayed by provisioning. Together these satisfy both the fail-fast validation goal and the warm-capacity latency goal, whereas scale-to-zero, inference tables and registry choice affect cost or governance only.

Answer

Set the endpoint's scale-to-zero behaviour so idle capacity is released between requests.

Answer

Register the model in the workspace Model Registry instead of Unity Catalog.

Answer

Enable inference tables to persist the request and response payloads.

Question 13

A data scientist has trained a scikit-learn model and wants to expose it for real-time inference through Databricks Model Serving. The model is currently logged as an MLflow run artifact but has not been registered anywhere. What must the data scientist do before creating the serving endpoint?

Accepted Answer

Register the model in Unity Catalog and note the model name and version.. Databricks Model Serving deploys registered model versions, referencing a Unity Catalog model name and version. Logging a model during training stores artifacts but does not make the model deployable until it is registered. Once registered, the endpoint can be created against that version, and the serving layer uses the stored signature and environment to build the container. Raw file uploads, format conversions, and scheduled jobs do not satisfy this prerequisite.

Answer

Convert the model to an ONNX file for cross-platform compatibility.

Answer

Export the model to a pickle file and upload it to DBFS.

Answer

Create a Databricks job that runs the model on a schedule.

Question 14

A data scientist trains a model with a feature engineering pipeline and wants batch scoring to happen nightly on a Delta table using Databricks, producing predictions that downstream dashboards read. The scoring job must scale with data volume and be re-runnable if it fails. Which approach best fits these requirements?

Accepted Answer

Use MLflow's predict API to load the model in a Spark job and score the Delta table with a distributed DataFrame.. Nightly bulk scoring on a Delta table is a distributed batch workload, so the model should be loaded with the MLflow predict API inside a Spark job that reads the feature table and writes predictions back to Delta. That approach scales horizontally with data volume and integrates cleanly with Databricks job scheduling and retries, unlike endpoint-per-row calls, dashboard-embedded pickles or single-threaded SQL.

Answer

Deploy the model to a real-time Model Serving endpoint and call it row by row from a notebook loop.

Answer

Export the model to a pickle file and have the dashboard application load and score records on demand.

Answer

Convert the model to a SQL UDF and run a single-threaded query against the feature table.

Question 15

A platform team is rolling out a new Databricks Model Serving endpoint for a churn model. They must ensure the endpoint can be queried by an external application and that only authorized callers can invoke it. Which TWO actions should they take? (Choose two.)

Accepted Answer

Grant the calling principal permission on the served model or endpoint and issue a Databricks personal access token or OAuth token for authentication.. Securing an endpoint for external use requires both an authenticated identity and a permission grant. A service principal with an OAuth token is the preferred machine identity, and granting it permission on the endpoint or served model enforces authorization. Using a personal access token for a human or service identity works similarly. Together these actions let the external application reach the endpoint while ensuring unauthorized callers are rejected.

Answer

Expose the endpoint only through a cluster-scoped init script that writes the scoring URL into the application's configuration.

Answer

Disable authentication on the endpoint so the external application can call the REST API without credentials.

Answer

Enable inference tables on the endpoint so that requests are logged and therefore implicitly authenticated.