

I prepare CSV exports with documented cleanup rules, a preserved source file, flagged exceptions, and a record-count reconciliation. My worked synthetic sample shows conservative duplicate handling and a repeatable Python workflow. Scope is agreed before work begins.