PMLE Practice Question: Collaborating Within and Across Teams to Manage Data and Models

Answer analysis

Option-by-option breakdown

For each option: why learners choose it and why it is or isn't the right answer here.

• ✗

  Vertex AI Workbench

  Why it's wrong here

  Workbench provides managed JupyterLab notebook environments for interactive development, not a centralised feature store. It cannot serve the same feature values to both training and online prediction, so training-serving skew persists. It would be the right choice for authoring and experimenting with model code, not for sharing engineered features across models.

• ✗

  Vertex AI Model Registry

  Why it's wrong here

  Model Registry catalogues and versions trained models and their artefacts; it stores no feature definitions or feature values. Sharing features across models requires a feature store, which serves identical feature data to training and serving. Model Registry would be correct for tracking model lineage, approval and deployment stages.

• ✓

  Vertex AI Feature Store

  Why this is correct

  Vertex AI Feature Store provides a centralised repository where features are computed once and served identically to both training and prediction pipelines, directly eliminating training-serving skew. Sharing one feature definition across multiple models satisfies the consistency requirement, since online and batch serving draw from the same managed source rather than duplicated transformations.

• ✗

  Vertex AI Experiments

  Why it's wrong here

  Vertex AI Experiments tracks and compares training runs, logging parameters and metrics; it stores no feature values and serves nothing at inference, so skew cannot be reduced. It is the right choice for auditing model iterations, not for sharing features between models.

Go deeper

Related to this question