CERTIFIED-MACHINE-LEARNING-PROFESSIONAL Question #47: Real Exam Question with Answer & Explanation

Question

Which of the following MLflow Model Registry use cases requires the use of an HTTP Webhook?

Options

• AStarting a testing job when a new model is registered
• BUpdating data in a source table for a Databricks SQL dashboard when a model version transitions
• CSending an email alert when an automated testing Job fails
• DNone of these use cases require the use of an HTTP Webhook
• ESending a message to a Slack channel when a model version transitions stages

Explanation

Option B requires an HTTP Webhook because updating a source table for a Databricks SQL dashboard involves calling the Databricks SQL REST API - an external HTTP endpoint. MLflow's Job Webhooks can only trigger Databricks Jobs directly; they cannot make arbitrary HTTP API calls, so only an HTTP Webhook can invoke the SQL REST API to refresh or update the dashboard's underlying data.

Why the distractors are wrong:

• A (starting a testing job) can be handled by a Job Webhook, which is purpose-built to trigger Databricks Jobs - no HTTP Webhook needed.
• C (email alert on job failure) is a standard Databricks Job notification setting, not an MLflow Model Registry webhook at all.
• E (Slack message on stage transition) is the classic HTTP Webhook example, but it does NOT require one - you could instead use a Job Webhook to trigger a Databricks Job that sends the Slack message, making it optional rather than mandatory.

Memory tip: Ask yourself "does this target an external HTTP endpoint that can't be reached any other way?" Only B is locked into needing a direct HTTP call (the SQL REST API) with no Job Webhook alternative - the other scenarios either use Job Webhooks natively or fall outside the Model Registry webhook system entirely.

No community discussion yet for this question.