nerdexam
Databricks

GENERATIVE-AI-ENGINEER-ASSOCIATE · Question #88

A Generative AI Engineer using the code below to test setting up a vector store: Assuming they intend to use Databricks managed embeddings with the default embedding model, what should be the next…

The correct answer is B. vsc.create_delta_sync_index(). After creating a VectorSearchClient, the next step when using Databricks managed embeddings is to create a Delta Sync index, which automatically handles embedding generation using the default model.

Vector Store Management and Indexing on Databricks

Question

A Generative AI Engineer using the code below to test setting up a vector store:

Assuming they intend to use Databricks managed embeddings with the default embedding model, what should be the next logical function call?

Exhibit

GENERATIVE-AI-ENGINEER-ASSOCIATE question #88 exhibit

Options

  • Avsc.get_index()
  • Bvsc.create_delta_sync_index()
  • Cvsc.create_direct_access_index()
  • Dvsc.similarity_search()

How the community answered

(41 responses)
  • A
    2% (1)
  • B
    88% (36)
  • C
    2% (1)
  • D
    7% (3)

Why each option

After creating a VectorSearchClient, the next step when using Databricks managed embeddings is to create a Delta Sync index, which automatically handles embedding generation using the default model.

Avsc.get_index()

vsc.get_index() retrieves an existing index by name and will fail if called before the index has been created.

Bvsc.create_delta_sync_index()Correct

vsc.create_delta_sync_index() creates a Vector Search index that syncs from a Delta table and uses Databricks-managed embeddings to automatically embed source data - this is the correct next call when intending to use the default managed embedding model without providing custom embeddings.

Cvsc.create_direct_access_index()

vsc.create_direct_access_index() is used when the caller will supply pre-computed embeddings directly, not when using Databricks managed embeddings.

Dvsc.similarity_search()

vsc.similarity_search() queries an existing index and cannot be called before the index has been created and populated.

Concept tested: Databricks Vector Search index creation with managed embeddings

Source: https://docs.databricks.com/en/generative-ai/vector-search.html

Topics

#Databricks Vector Search#Vector Store Creation#Managed Embeddings#Delta Lake Integration

Community Discussion

No community discussion yet for this question.

Full GENERATIVE-AI-ENGINEER-ASSOCIATE Practice