nerdexam
Google

PROFESSIONAL-MACHINE-LEARNING-ENGINEER · Question #197

You have been tasked with deploying prototype code to production. The feature engineering code is in PySpark and runs on Dataproc Serverless. The model training is executed by using a Vertex AI custom

The correct answer is C. Use the Kubeflow pipelines SDK to write code that specifies two components:. To create a scalable and maintainable end-to-end production process for disconnected PySpark feature engineering and Vertex AI model training, use Kubeflow Pipelines to define and orchestrate these steps as interconnected components.

Submitted by fatima_kr· Apr 18, 2026ML pipeline operationalization

Question

You have been tasked with deploying prototype code to production. The feature engineering code is in PySpark and runs on Dataproc Serverless. The model training is executed by using a Vertex AI custom training job. The two steps are not connected, and the model training must currently be run manually after the feature engineering step finishes. You need to create a scalable and maintainable production process that runs end-to-end and tracks the connections between steps. What should you do?

Options

  • ACreate a Vertex AI Workbench notebook. Use the notebook to submit the Dataproc Serverless
  • BCreate a Vertex AI Workbench notebook. Initiate an Apache Spark context in the notebook and
  • CUse the Kubeflow pipelines SDK to write code that specifies two components:
  • DUse the Kubeflow pipelines SDK to write code that specifies two components

How the community answered

(28 responses)
  • A
    4% (1)
  • B
    7% (2)
  • C
    75% (21)
  • D
    14% (4)

Why each option

To create a scalable and maintainable end-to-end production process for disconnected PySpark feature engineering and Vertex AI model training, use Kubeflow Pipelines to define and orchestrate these steps as interconnected components.

ACreate a Vertex AI Workbench notebook. Use the notebook to submit the Dataproc Serverless

Using a Vertex AI Workbench notebook for orchestration would be a manual or semi-manual process, lacking the scalability, maintainability, and automatic connection tracking required for a production process.

BCreate a Vertex AI Workbench notebook. Initiate an Apache Spark context in the notebook and

Using a Vertex AI Workbench notebook for orchestration would be a manual or semi-manual process, lacking the scalability, maintainability, and automatic connection tracking required for a production process.

CUse the Kubeflow pipelines SDK to write code that specifies two components:Correct

The Kubeflow Pipelines SDK is designed for building, deploying, and managing scalable, end-to-end ML workflows. Defining each step (Dataproc Serverless and Vertex AI custom training) as a component allows Kubeflow Pipelines to orchestrate execution, track dependencies, and provide lineage, fulfilling all requirements for a production process.

DUse the Kubeflow pipelines SDK to write code that specifies two components

This option is largely identical to C, which describes the correct approach for orchestrating ML pipelines.

Concept tested: MLOps orchestration with Kubeflow Pipelines

Source: https://cloud.google.com/vertex-ai/docs/pipelines/introduction

Topics

#ML Pipelines#Workflow Orchestration#Kubeflow Pipelines SDK#Production Deployment

Community Discussion

No community discussion yet for this question.

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