GS Caltex connects Amazon SageMaker Canvas to Amazon SageMaker Pipelines, allowing no-code models built by process engineers to run in production. According to the post, ML engineers reuse validated preprocessing flows and models, which reduces deployment time from roughly a week to a few hours. The resulting setup runs batch inference every 30 minutes. This is part 1 of a series on the company’s MLOps approach.
Why it matters
Bridging no-code model building with production pipelines can shorten the path from a domain expert’s prototype to an operational system. The reported reduction in deployment time and the regular batch inference cadence point to a workflow aimed at making engineer-built models usable in day-to-day operations.
Who should care
ML engineers, MLOps teams, and enterprises using SageMaker Canvas and Pipelines may find the described integration relevant, particularly those looking to operationalize models created by non-specialist staff.