nerdexam
Google

PROFESSIONAL-MACHINE-LEARNING-ENGINEER · Question #268

You recently trained an XGBoost model on tabular data. You plan to expose the model for internal use as an HTTP microservice. After deployment, you expect a small number of incoming requests. You…

The correct answer is D. Use a prebuilt XGBoost Vertex container to create a model, and deploy it to Vertex AI Endpoints. To productionize an XGBoost model as an HTTP microservice with minimal effort and latency for a small number of requests, use a prebuilt Vertex AI container and deploy it to Vertex AI Endpoints.

Submitted by andres_qro· Apr 18, 2026ML pipeline operationalization

Question

You recently trained an XGBoost model on tabular data. You plan to expose the model for internal use as an HTTP microservice. After deployment, you expect a small number of incoming requests. You want to productionize the model with the least amount of effort and latency. What should you do?

Options

  • ADeploy the model to BigQuery ML by using CREATE MODEL with the
  • BBuild a Flask-based app. Package the app in a custom container on Vertex AI, and deploy it to
  • CBuild a Flask-based app. Package the app in a Docker image, and deploy it to Google
  • DUse a prebuilt XGBoost Vertex container to create a model, and deploy it to Vertex AI Endpoints.

How the community answered

(67 responses)
  • A
    6% (4)
  • B
    18% (12)
  • C
    4% (3)
  • D
    72% (48)

Why each option

To productionize an XGBoost model as an HTTP microservice with minimal effort and latency for a small number of requests, use a prebuilt Vertex AI container and deploy it to Vertex AI Endpoints.

ADeploy the model to BigQuery ML by using CREATE MODEL with the

Deploying to BigQuery ML serves predictions within BigQuery, not as an external HTTP microservice for low-latency individual requests.

BBuild a Flask-based app. Package the app in a custom container on Vertex AI, and deploy it to

Building a custom Flask app and container for a standard XGBoost model involves more effort than leveraging Vertex AI's prebuilt containers designed for this purpose.

CBuild a Flask-based app. Package the app in a Docker image, and deploy it to Google

Deploying a custom Flask app to Google (e.g., Cloud Run) involves more manual setup and configuration compared to the streamlined, managed model deployment offered by Vertex AI Endpoints with prebuilt containers.

DUse a prebuilt XGBoost Vertex container to create a model, and deploy it to Vertex AI Endpoints.Correct

Using a prebuilt XGBoost Vertex AI container significantly reduces deployment effort by providing a ready-to-use serving environment. Deploying this to Vertex AI Endpoints creates a managed HTTP microservice with low latency, perfectly suited for online predictions with minimal operational overhead.

Concept tested: Vertex AI Endpoints with prebuilt containers

Source: https://cloud.google.com/vertex-ai/docs/predictions/prebuilt-containers

Topics

#Model Deployment#Vertex AI Endpoints#Prebuilt Containers#XGBoost

Community Discussion

No community discussion yet for this question.

Sign up to join the discussion
Full PROFESSIONAL-MACHINE-LEARNING-ENGINEER Practice