Sourabh Godse

Most of my time here was spent deep inside GCP — building and maintaining ETL pipelines that fed healthcare data into BigQuery for reporting and analytics teams. The data volumes were large, the schemas were messy, and the tolerance for errors was low (healthcare data tends to work that way). A big part of the role was performance work. I'd get handed a query that was taking minutes, or a pipeline that was quietly costing too much, and figure out what was actually wrong — whether that was missing partitioni

FIGmd is a healthcare analytics company, so the data I worked with was clinical — quality measures, reporting feeds, the kind of data where a wrong number in a pipeline has real downstream consequences. My main work was SQL-heavy: writing, reviewing, and tuning queries across large datasets that powered product dashboards and compliance reports. When queries were slow, I dug in — rewrote joins, restructured logic, got them to a place where reporting deadlines weren't a constant anxiety. I also handled a lot