GS Caltex Connects SageMaker Canvas to Production MLOps
Understand the story
On November 14, 2026, the AWS Machine Learning Blog published a firsthand case study: South Korean energy and chemical company GS Caltex connected Amazon SageMaker Canvas with SageMaker Pipelines to bring machine learning to production. Process engineers build models in Canvas with no code, while ML engineers handle model deployment and inference scheduling, achieving automated handoff between the two roles. By reusing validated preprocessing pipelines and models, the team cut model deployment time from about one week to a few hours. For inference, they replaced an always-on endpoint with batch inference run every 30 minutes, significantly reducing inference costs.
AI overview of reporting · Updated 2 h前
Reporting timeline
Explore different aspects of the story through its reporting.
- AWS Machine Learning BlogProduction MLOps at GS Caltex with Amazon SageMaker Canvas, part 1
Korean energy and chemical company GS Caltex connected Amazon SageMaker Canvas with SageMaker Pipelines, letting process engineers build models without code in Canvas while ML engineers handle deployment and scheduled inference. By automating handoffs and reusing validated preprocessing pipelines and models, deployment time dropped from about 1 week to a few hours, and batch inference every 30 minutes replaced a always-on endpoint, significantly cutting inference costs.
Story heat trend
Not enough continuous observations to show a trend yet.