At Freed, I'm building a redesigned MongoDB-to-PostgreSQL pipeline to prevent schema-inference data loss across 500+ collections. It uses raw JSONB ingestion, metadata-driven schema inference, automated DDL generation, and Airflow DAGs.
I also maintain Freed's MongoDB-to-PostgreSQL CDC pipeline across GCP VMs and own the PostgreSQL Data Warehouse. I redesigned indexing to bring query latency from minutes to under a second and dashboard loads from around 30 minutes to under a minute.
At Genpact, I contributed to GE Vernova's migration from Greenplum and Talend to Databricks, migrating 50+ tables and developing 100+ production outgestion jobs. I also built a Databricks and dbt streaming data platform project using Delta Lake, incremental models, and SCD Type 2 snapshots.
