PMLE Collaborating to manage data and models Practice Question

Which TWO statements about Vertex AI Feature Store are correct? (Choose 2)

⚠ Common exam trap

Google Cloud often tests the misconception that Vertex AI Feature Store is tightly coupled to Vertex AI models or that it performs automatic feature engineering, when in fact it is a decoupled storage and serving layer that supports any ML framework and requires explicit feature engineering steps.

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

✓

Feature Store provides a centralized repository for feature data.

Option D is correct because Vertex AI Feature Store acts as a centralized repository where feature data is registered, versioned, and managed as feature groups and features, so multiple teams and models can share a single consistent source of features. Option E is correct because Feature Store supports both online serving, which returns low-latency feature values for real-time predictions, and offline serving, which reads historical feature values in bulk for training and batch scoring. Options A, B, and C are incorrect: Feature Store does not automatically perform feature engineering transformations (you define and ingest the transformed values yourself), it is not limited to numerical features (it supports types such as strings, booleans, and arrays as well), and it is not restricted to Vertex AI models since features can be served to any application or model via the online/offline APIs.

Answer analysis

Option-by-option breakdown

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

✗
Feature Store automatically applies feature engineering transformations.
Why it's wrong here
Feature Store stores and serves precomputed feature values; transformations are authored in BigQuery or Dataflow before ingestion. It would be the right tool for online serving of engineered features, but automatic in-store transformation is not a capability it provides.
✗
Feature Store can only store numerical features.
Why it's wrong here
Feature Store supports numerical, categorical, string, and embedding value types, so restricting it to numerical features is false. It would be correct if only numeric tensors were servable, but the online store handles multiple types across entities and feature groups.
✗
Feature Store can only be used with Vertex AI models.
Why it's wrong here
Feature Store serves features to any serving system, including custom models on Compute Engine, GKE or on-premises, via online serving APIs; it is not bound to Vertex AI models. It is tempting because its tightest integration is with Vertex AI training and prediction, where it would be the natural choice.
✓
Feature Store provides a centralized repository for feature data.
Why this is correct
Vertex AI Feature Store acts as a centralised repository, letting teams register, version and share feature data across projects and models instead of duplicating feature engineering per pipeline, which is the defining architectural property this statement asserts.
✓
Feature Store supports both online and offline serving.
Why this is correct
Vertex AI Feature Store separates online serving, which returns low-latency feature values for real-time prediction, from offline serving, which reads historical feature values from the offline store for training and batch scoring, so both serving modes are supported.

Go deeper

PMLE Collaborating to manage data and models Practice Question

Why each option matters

Option-by-option breakdown

Related to this question