AWS Machine Learning says GS Caltex uses SageMaker Canvas and Pipelines for faster model handoff

In a case study published by AWS Machine Learning, GS Caltex describes a workflow that hands validated models from process engineers in SageMaker Canvas to ML engineers running SageMaker Pipelines. The company automated the transfer of preprocessing flows, model configuration and approval steps into production jobs. GS Caltex says deployment time has fallen from about a week to a few hours, and batch inference now runs every 30 minutes, reducing inference costs.