SageMaker Canvas Iris ML Pipeline demonstrates an end-to-end low-code machine learning workflow on AWS, from data ingestion through real-time inference. The pipeline ingests the Iris dataset into S3, catalogs it via AWS Glue and Lake Formation, then uses SageMaker Canvas for automated data preparation and built-in model training without writing code. Trained models are registered and deployed to SageMaker Real-Time Endpoints, with predictions served through API Gateway and Lambda, while CloudWatch monitors both training metrics and endpoint performance. This architecture eliminates the need for custom ML code, enabling rapid model development and deployment for classification tasks. Fork this diagram on Diagrams.so to customize data sources, add batch inference, or integrate additional AWS services like SageMaker Feature Store. The modular design separates ingestion, training, and inference layers, making it easy to scale or swap components as requirements evolve.