An ecommerce company wants to migrate on-premises data ingestion pipelines to AWS without managing servers. The solution must orchestrate existing Python and Bash scripts without refactoring. Which option provides the least operational overhead?

Amazon MWAA is the best choice because it's a fully managed service for Apache Airflow, which is excellent for orchestrating complex data pipelines. It natively supports Python and Bash scripts, allowing the company to migrate existing code without refactoring, directly addressing the core requirement. Being fully managed, it minimizes operational overhead. AWS Lambda is serverless but is designed for short-running, event-driven functions, not complex pipeline orchestration, and would require significant refactoring to handle Bash scripts directly. AWS Step Functions is good for orchestrating workflows but requires defining state machines, which would involve refactoring existing scripts into individual steps. AWS Glue is an ETL service, not primarily an orchestration tool for arbitrary scripts, and while it supports Python, it's not ideal for directly running existing Bash scripts without modification.