Production MLOps at GS Caltex with Amazon SageMaker Canvas, part 1
GS Caltex connects Amazon SageMaker Canvas to Amazon SageMaker Pipelines so process engineers' no-code models run in production. This post shows how ML engineers reuse validated preprocessing flows and models, cutting deployment from about a week to a few hours and running batch inference every 30 minutes.
First seen by Oriel on .
AI summary
GS Caltex automated the handoff from no-code Amazon SageMaker Canvas model development to production SageMaker Pipelines, cutting deployment time from about a week to a few hours.
- Process engineers validate preprocessing flows and models in SageMaker Canvas, while ML engineers manage production configuration, approvals, deployment, scheduling, and execution history.
- The five-step pipeline prepares inputs, runs Data Wrangler preprocessing, removes the placeholder target, performs SageMaker Batch Transform, and assembles predictions with preprocessed data.
- Amazon EventBridge starts batch inference every 30 minutes, avoiding continuously running endpoints and meaningfully reducing inference costs.
- Production reuse requires pinning the validated container, preserving learned transformations, matching schemas and feature order, and preventing overlapping executions that share fixed S3 input paths.
- Model Registry stores approval status but does not enforce it during inference, so deployment code must verify approval and update schedules if approval is revoked.
#amazon sagemaker #mlops #batch inference #amazon eventbridge
Generated by gpt-5.6-sol. Check the original for details.
Aliases
| Source | External ID | URL |
|---|---|---|
| AWS Machine Learning Blog | 760cbc8b70c56b832a6d83eb6dfab47ce2fe8a3b | https://aws.amazon.com/blogs/machine-learning/production-mlops-at-gs-caltex-with-amazon-sagemaker-canvas-part-1/ Read discussion |