GS Caltex has linked Amazon SageMaker Canvas to Amazon SageMaker Pipelines to deploy machine learning workflows into production. By connecting these systems, the organization enables no-code models created by process engineers to function in live operational environments. The integration allows machine learning engineers to reuse previously validated preprocessing flows alongside the models themselves.
The implementation significantly accelerates the operational lifecycle of machine learning workloads at the company. Deployment durations have dropped from roughly a week to only a few hours. In addition to faster release cycles, the resulting pipeline routinely executes batch inference every 30 minutes.

