At Contractor, I led the migration of production databases from PostgreSQL to Google BigQuery, redesigning schemas and maintaining data integrity while minimizing downtime.
I also implemented CI/CD pipelines with Cloud Build and ArgoCD, and developed tests to validate data accuracy and consistency throughout the migration.
At Capgemini, I optimized a real-time PySpark ETL pipeline, reducing processing time by 75%. I also created a Dataflow pipeline that loaded data into BigQuery and fed a VertexAI model.
At Spotlite, I improved and maintained an ETL pipeline using Python and distributed computing with Dask. My academic work includes evaluating GANs for InSAR phase unwrapping and contributing to a publication comparing YOLO with DETR.

