PMLE Practice Question: Collaborating Within and Across Teams to Manage Data and Models
A data science team needs to share features across multiple ML models while ensuring consistency between training and serving. Which approach best achieves this?
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
✓
Use Vertex AI Feature Store to define and serve features for both training and online prediction
Vertex AI Feature Store provides a central repository where features are defined once and reused across models, reducing training-serving skew.
Answer analysis
Option-by-option breakdown
For each option: why learners choose it and why it is or isn't the right answer here.
- ✗
Store features in a shared BigQuery dataset without versioning
Why it's wrong here
Without versioning, BigQuery feature values change in place, so training reads one snapshot while serving reads another, breaking consistency. It is tempting because a shared dataset does centralise features, and would suffice if the requirement were only storage consolidation rather than reproducible point-in-time parity.
- ✗
Export features to CSV files shared via Cloud Storage
Why it's wrong here
CSV exports in Cloud Storage are static snapshots, so serving reads stale values rather than the live feature, and there is no shared transformation logic. It is tempting because files are simple to distribute, and would suit one-off batch scoring, not continuous training-serving consistency.
- ✓
Use Vertex AI Feature Store to define and serve features for both training and online prediction
Why this is correct
Vertex AI Feature Store provides a centralised repository where feature values are defined once and served consistently to both training jobs and online prediction, eliminating training-serving skew. This shared definition satisfies the consistency requirement while allowing multiple models to reuse the same features.
- ✗
Each team maintains its own feature engineering code in separate pipelines
Why it's wrong here
Separate pipelines duplicate transformation logic, so training and serving drift whenever one team edits its code. It is tempting because independent pipelines give teams autonomy and would work where models share no features, but here the requirement is cross-model consistency from one definition.
Go deeper
Related to this question
About these practice questions
One of 775 original PMLE practice questions on Courseiva, each with a full explanation and wrong-answer analysis — not exam dumps or protected exam content. Learn why practice questions differ from exam dumps →
JA
Written by Johnson Ajibi, MSc IT Security
Senior Network & Security Engineer · founder of Courseiva
This PMLE practice question is part of Courseiva's free Google Cloud 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 PMLE exam.