Design and Implement Models and Embeddings DP-800 Practice Questions (Page 5)

Questions 21–25

1. 21application · medium

   A retail company has a product catalog in Azure SQL Database. They generate embeddings for product descriptions to power a recommendation engine. The descriptions change infrequently, but when they do change, the embeddings must be updated within minutes. The team wants to minimize custom code and operational overhead. Which approach should they use?

   • AManually regenerate embeddings whenever a product description changes.
   • BSchedule a nightly Azure Data Factory pipeline to regenerate embeddings for all products.
   • CUse a database trigger to call the embedding API synchronously for every insert and update.
   • DEnable Change Tracking on the product table and use an Azure Function with SQL trigger binding to regenerate embeddings for changed rows.
   Reveal Solution Discussion Select an answer first
2. 22foundation · easy

   What is the primary purpose of managing the lifecycle of an external AI model in a database solution?

   • ATo ensure the model remains compatible with the database schema
   • BTo ensure the model is accessible to all database users
   • CTo reduce the cost of running the model
   • DTo keep the model updated with the latest data and improvements
   Reveal Solution Discussion Select an answer first
4. 23application · medium

   A financial services company has deployed an external model in Azure SQL Database for sentiment analysis of customer feedback. The model vendor releases a new version with improved accuracy and lower latency. The company wants to update the model without disrupting ongoing operations. What should they do first?

   • AKeep the old model and never update it
   • BDelete the existing model and recreate it with the new version
   • CImmediately update the production external model to the latest version
   • DTest the new model version in a staging environment before updating the production external model
   Reveal Solution Discussion Select an answer first
5. 24application · medium

   A manufacturing company maintains a parts database in Azure SQL Database. They use embeddings for a semantic search tool. The parts table has a 'Description' column that is updated frequently. The team wants to keep embeddings synchronized with minimal latency and without writing custom polling code. Which Azure service should they use?

   • AAzure Data Factory with a scheduled pipeline
   • BAzure Analysis Services
   • CSQL Server Integration Services (SSIS)
   • DAzure Logic Apps with a SQL trigger
   Reveal Solution Discussion Select an answer first
6. 25foundation · easy

   Which column should be excluded from an embedding to avoid introducing noise and reducing model performance?

   • AProduct name
   • BCustomer review text
   • CInternal tracking code
   • DProduct category
   Reveal Solution Discussion Select an answer first