← Back
AWS ML Blog

Production MLOps at GS Caltex with Amazon SageMaker Canvas, part 1

•16 min read•
#deployment#llm
Level:Intermediate
For:AI Engineers
✦TL;DR

GS Caltex accelerated its MLOps by linking Amazon SageMaker Canvas‑built models to SageMaker Pipelines, cutting deployment time from about a week to a few hours. Engineers use Canvas for visual data prep (Data Wrangler) and model training, then register the model in the SageMaker Model Registry. ML engineers convert these assets into Pipeline steps that run preprocessing and SageMaker Batch Transform on a 30‑minute batch schedule via Amazon EventBridge, dramatically lowering inference‑costs.

⚡ Key Takeaways

  • Deployment time dropped from ~1 week to a few hours after automating the Canvas‑to‑Pipelines handoff.
  • The workflow reuses a Data Wrangler preprocessing flow and a registered model version as Pipeline inputs.
  • Batch inference runs every 30 minutes, avoiding continuously running endpoints and reducing cost.
  • Production jobs are orchestrated with SageMaker Pipelines and triggered by Amazon EventBridge.
  • Assets must be registered in the SageMaker Model Registry and stored in Amazon S3 before they can be consumed by Pipelines.
💡 Why It Matters

Engineers can achieve rapid model roll‑outs and cost‑effective inference by treating no‑code Canvas artifacts as first‑class pipeline components, enabling tighter collaboration between process and ML teams.

✅ Practical Steps

  1. Build and validate a model in SageMaker Canvas, then register the model in the SageMaker Model Registry.
  2. Create SageMaker Pipelines steps that reference the Canvas‑generated Data Wrangler flow and the registered model, and schedule the pipeline with Amazon EventBridge for 30‑minute batch runs.

Want the full story? Read the original article.

Read on AWS ML Blog ↗

More like this

Introducing workload recovery and expanded infrastructure checks on Amazon SageMaker HyperPod

AWS ML Blog•#deployment

Introducing Unity Gateway Insights

Databricks Blog•#deployment

Introducing Gemini 4 Neon

Google DeepMind•#llm

DaisyRose-27B: Pushing the Frontier of Post-Training Efficiency

Hugging Face Blog•#llm

EXPLORE AI NEWS

Daily hand-picked stories on LLMs, RAG, agents and production AI — curated for engineers who ship.

BROWSE NEWS

GET THE WEEKLY DIGEST

Join engineers getting the Monday signal-over-noise AI breakdown. No spam, unsubscribe anytime.

LEARN AI ENGINEERING

Curated courses, research papers, repos and tutorials built for engineers leveling up in AI.

START LEARNING