
Answer-first summary for fast verification
Answer: vsc.create_direct_access_index()
The question asks for the next logical function call when testing a vector store setup with Databricks managed embeddings and default embedding model. The community discussion shows a clear consensus: vsc.create_direct_access_index() (Option C) is optimal for testing scenarios because it allows manual addition of documents and embeddings without requiring a Delta table, making it ideal for minimal test setups. Option B (vsc.create_delta_sync_index()) is more suitable for production workflows where automatic synchronization with an existing Delta table is needed, but the question emphasizes testing. Option A (vsc.get_index()) retrieves an existing index rather than creating one, and Option D (vsc.similarity_search()) is for querying, not setup. The upvoted comments from adaine (2 upvotes) and DavidMiller (1 upvote) strongly support Option C, while Option B has mixed support with lower relevance to testing.
Author: LeetQuiz Editorial Team
Ultimate access to all questions.
A Generative AI Engineer is using the following code to test setting up a vector store:
from databricks.vector_search.client import VectorSearchClient
vsc = VectorSearchClient()
vs_index = vsc.create_delta_sync_index(
endpoint_name="...",
index_name="my_vector_index",
source_table_name="my_catalog.my_schema.my_chunked_docs",
pipeline_type="TRIGGERED",
primary_key="id",
embedding_source_column="text"
)
from databricks.vector_search.client import VectorSearchClient
vsc = VectorSearchClient()
vs_index = vsc.create_delta_sync_index(
endpoint_name="...",
index_name="my_vector_index",
source_table_name="my_catalog.my_schema.my_chunked_docs",
pipeline_type="TRIGGERED",
primary_key="id",
embedding_source_column="text"
)
Assuming they intend to use Databricks managed embeddings with the default embedding model, what should be the next logical function call?

A
vsc.get_index()
B
vsc.create_delta_sync_index()
C
vsc.create_direct_access_index()
D
vsc.similarity_search()
No comments yet.