How would you design a scalable ETL pipeline to handle large-scale biotech data from multiple sources?
