Explain how you would structure a data pipeline to ingest and normalize daily alternative data from multiple unstructured sources.
