At Logiciel Services, I build and operate a financial data lakehouse on Amazon S3 using Apache Iceberg, AWS Glue Data Catalog, Trino, and Spark on EMR on EKS. I migrated TB-scale trading and order data from Hive to Iceberg, reducing ETL costs by 10% and reporting latency by 40%.
I also built a Spark-based CAT-to-Blotter reconciliation engine and FINRA CAT ingestion pipelines. For trade blotter ingestion, I led the migration from Apache NiFi to Spark, creating 11 vendor connectors driven by a PostgreSQL schema registry.
At xLoop Digital, I developed real-time video processing and computer vision pipelines for a banking client. I also migrated SQL workflows to Spark and built batch ETL pipelines that processed more than 2M financial transaction rows daily.
