Data & ML
Production ML Training Pipeline
S3 → Spark feature engineering → Feast → SageMaker → MLflow → A/B deployment
AI Prompt
“Draw a production ML training pipeline: raw data in S3 → feature engineering (Spark on EMR) → feature store (Feast) → model training (SageMaker) → experiment tracking (MLflow) → model evaluation → model registry → A/B deployment with shadow mode.”
Try this in AIDrawIO
Free — no credit card required
Refine it with follow-up prompts
After generating the base diagram, use these prompts to iterate and add detail — the same way a real architect would refine a whiteboard sketch.
- 1
“Add Great Expectations data validation before feature engineering”
Try this follow-up - 2
“Show feedback loop from production predictions back into training”
Try this follow-up - 3
“Add data drift detection and model performance degradation alerts”
Try this follow-up
How AIDrawIO generates this diagram
- 1.You paste the prompt above into the chat input.
- 2.AIDrawIO sends it to your chosen AI model (Claude or Gemini).
- 3.The model returns draw.io-compatible XML — rendered instantly in the canvas.
- 4.Export as SVG, PNG, or XML. Edit any element manually or with follow-up prompts.