Question

Two data science teams at your company are prototyping churn-prediction models in separate Vertex AI Workbench notebooks. One team uses TensorFlow, the other uses XGBoost. Leadership wants a single, consolidated view to compare all runs across both teams using the same metrics (AUC, log loss) and to reproduce the exact hyperparameters and datasets that produced the best model. Which approach best supports this cross-team comparison and reproducibility?

Accepted Answer

Have each team log runs to a shared Vertex AI Experiment, recording metrics, parameters, and dataset artifacts, then compare runs in the Experiments UI.

Answer

Have each team export their metrics to separate CSV files stored in team-specific Cloud Storage buckets and manually merge them in a spreadsheet.

Answer

Have each team save their best model to the Vertex AI Model Registry and rely on the registry's version list to compare training metrics.

Answer

Have each team push their notebooks to a shared Git repository and re-run every notebook locally to reproduce the metrics for comparison.