Question

Two ML teams at a logistics company both consume a shared 'shipment_features' set from Vertex AI Feature Store. The routing team requests a schema change that adds a new derived feature and modifies the logic of an existing one. The demand-forecasting team's production models must continue to reproduce predictions using the exact feature definitions they trained on, while the routing team needs the updated definitions immediately. What approach best satisfies both teams while preserving reproducibility?

Accepted Answer

Create a new feature view / feature version with the updated definitions and keep the existing version active, allowing each team to reference the version their models depend on.

Answer

Overwrite the existing feature values in place and instruct the forecasting team to retrain their models against the new definitions before the change is applied.

Answer

Export the Feature Store data to a BigQuery table snapshot and have both teams read directly from BigQuery instead of the Feature Store.

Answer

Disable online serving for the shared feature set and require both teams to compute features independently in their own notebooks.