At bigbasket, I build production data pipelines for India-wide quick commerce, turning inventory, vendor, and transaction data into reliable systems with measurable business impact.
I led a Redshift-to-lakehouse migration using AWS DMS, dbt, Trino, Airflow, Iceberg, and EKS, authoring roughly 400 dbt models and migrating 50+ production jobs in six months. The platform reduced data freshness from D-1 to around 20 minutes, eliminated pipeline failures, and delivered about $62K per month in operational savings.
I’ve also built vendor-payables, stock-on-hand, ML personalization, and payout datamarts that improved stock availability, delivery-partner productivity, and targeted-SKU conversion. My work spans Python, PySpark, SQL, AWS, ClickHouse, and data modeling across high-availability lakehouse and warehouse environments.
