Amazon SageMaker Pipelines can be automated across multiple AWS accounts and Regions, but monitoring them becomes complex as developers must manually switch between environments. Amazon SageMaker Studio only provides monitoring within a single account and Region. A solution is presented to centralize monitoring using Amazon CloudWatch custom dashboards, with a customizable AWS CDK example on GitHub. The solution uses a serverless, event-driven architecture that responds to SageMaker Pipeline events in real time, avoiding always-on monitoring. It follows a hub-and-spoke model, centralizing monitoring in a primary account and Region while lightweight components in secondary accounts forward data. The architecture includes two AWS CloudFormation stacks: the Dashboard stack, deployed in the primary account with CloudWatch dashboard, Amazon DynamoDB storage tables, and AWS Lambda functions; and the Forwarder stack, deployed in monitored accounts, using Amazon EventBridge to send enriched data to the monitoring hub.