Production MLOps at GS Caltex with Amazon SageMaker Canvas
According to AWS Machine Learning: GS Caltex uses Amazon SageMaker Canvas and Pipelines to connect process engineers' models to production workflows. This integration allows ML engineers to deploy models within a few hours instead of a week, significantly lowering inference costs by running batch predictions every 30 minutes.
According to AWS Machine Learning:
GS Caltex, a South Korean energy and chemical company, divides AI model development between process engineers and machine learning engineers. Process engineers build and validate models visually in Amazon SageMaker Canvas without writing code. Meanwhile, ML engineers configure Amazon SageMaker Pipelines for deployment and scheduled inference.
To bridge the gap between development and production, GS Caltex automated the handoff process. ML engineers previously spent about a week on integration tasks like setting data paths and connecting preprocessing jobs. With the new automated setup, they can deploy models within a few hours. The company now runs batch inference every 30 minutes, which eliminates the need to keep costly real-time prediction endpoints running continuously and lowers overall costs.
View selection rationale
Importance 55/100
About this brief
- Content type
- News summary
- Source type
- First-party / official source
- Original source
- AWS Machine Learning ↗
- Source published
- Process
- AI-assisted discovery, translation, summarisation, classification and scoring; standards and corrections are maintained by the editorial desk.
- Verification
- The original URL and publication metadata are retained. This brief is not independently fact-checked unless explicitly stated.
- Published by
- Polyemergence Editorial Desk · PolyEmergence Pte Ltd.
- Conflict disclosure
- No financial or commercial conflict declared.
