Databricks-GenAI-Assoc Application Development Practice Question
A GenAI engineer is deploying a chain that calls an external LLM API. The chain must not block the serving thread while waiting on the remote API, and the endpoint must handle many concurrent requests. Which implementation approach should the engineer choose when logging the model?
⚠ Common exam trap
The trap here is treating a remote LLM call like local computation, where a synchronous client seems adequate, when in fact blocking I/O caps concurrency on the endpoint.
Answer choices
Why each option matters
Answer the question above first, then reveal the full breakdown to understand why each option is right or wrong.
Correct answer & explanation
✓
Implement predict as an async function that awaits an async HTTP client, so the event loop can interleave other requests while waiting on the external API.
Calling an external LLM API is I/O-bound, so the serving process should not block a thread while waiting. Writing predict as an async function with an async HTTP client lets the event loop serve other requests during those waits, improving concurrency on each replica. Synchronous clients, table-based queues, and single-concurrency settings all undermine throughput or break the synchronous API contract.
Answer analysis
Option-by-option breakdown
For each option: why learners choose it and why it is or isn't the right answer here.
- ✗
Batch all incoming requests by writing them to a Delta table and have a separate job poll the table and call the external API.
Why it's wrong here
Introducing a table-based queue turns a synchronous inference API into an asynchronous batch workflow, which breaks the request-response contract expected from a serving endpoint. Clients would not receive answers in the same call. This design also adds latency and operational complexity, so it does not satisfy the concurrency requirement.
- ✓
Implement predict as an async function that awaits an async HTTP client, so the event loop can interleave other requests while waiting on the external API.
Why this is correct
Model Serving supports async predict, and an async HTTP client releases the event loop during network waits, allowing other requests to be handled concurrently on the same replica. This directly addresses the non-blocking requirement and improves throughput for I/O-bound chains. It is the recommended pattern for calling external LLM APIs from a served model.
- ✗
Use a synchronous HTTP client inside predict and increase the endpoint's scale-to-zero timeout so queued requests eventually complete.
Why it's wrong here
A synchronous client blocks the serving thread for the duration of each remote call, so concurrency is limited by the number of worker threads. Raising the scale-to-zero timeout changes idle behavior, not throughput under load. This approach would cause head-of-line blocking and poor concurrency, failing the scenario's requirement.
- ✗
Set the endpoint's concurrency to one and rely on the external API's own load balancing to absorb parallel requests.
Why it's wrong here
Setting concurrency to one serializes requests at the endpoint, guaranteeing that the deployment cannot handle many concurrent calls regardless of the external API's capabilities. The external API's load balancing is irrelevant if the endpoint itself admits one request at a time. This option directly contradicts the scenario's throughput goal.
About these practice questions
This Databricks-GenAI-Assoc question is part of Courseiva's 330-question bank — original exam-style content with full explanations and wrong-answer analysis, never real exam questions or exam dumps. Learn why practice questions differ from exam dumps →
JA
Written and reviewed by Johnson Ajibi, MSc IT Security
Senior Network & Security Engineer · founder of Courseiva
Last reviewed September 2026 · checked against the official Databricks exam blueprint
This Databricks-GenAI-Assoc practice question is part of Courseiva's free Databricks certification practice question bank. Courseiva provides original exam-style practice questions with explanations, topic-based practice, mock exams, readiness tracking, and study analytics to help learners prepare for the Databricks-GenAI-Assoc exam.