Question

A retail company has deployed a custom demand-forecasting model on Vertex AI. Over time, the data science team notices the model's accuracy is degrading as customer buying patterns shift. Leadership wants a repeatable, automated process to retrain and redeploy the model when performance drops, without engineers manually rebuilding everything each time. Which Google Cloud capability best addresses this need?

Accepted Answer

Vertex AI Pipelines to automate and orchestrate the ML workflow (MLOps), including retraining and deployment

Answer

The Vertex AI pre-trained Vision API to replace the custom model

Answer

Gemini in the Google Cloud console to chat about the model's results

Answer

BigQuery BI Engine to accelerate the dashboards that display forecasts