SNOWPRO-ADVANCED-MLOPS-ENGINEER Practice Test

Question 1 of 5

Single choice

A team wants to train, register and serve a model without moving customer data out of Snowflake. Which architecture fits this requirement best?

Choose one answer

A. Train with Snowpark ML, register the model in the Snowflake Model Registry, and run inference inside Snowflake
B. Send a full extract of each table to a third-party training service and receive back a model file
C. Export the tables nightly to an external object store and train on a separate cluster that the data team manages
D. Copy the feature tables to a laptop for training, then upload the model file to a shared drive by hand

Choose an answer or reveal it to continue.

Keyboard shortcuts:A to D or 1 to 4answerEntercheck the answer←→previous or nextFflag for review