data engineer in fintech

The company’s data lakehouse stores millions of financial events daily, including bets, wallet transactions, and live odds, and serves over 12 million active users.

• Own the lakehouse architecture, including bronze/silver/gold layers, Iceberg/Delta tables, and schema evolution
• Ingest operational data through CDC streaming with Kafka and Debezium, handling late and duplicate events
• Design data layouts for speed and cost, including partitioning, compaction, file sizing, and query performance on Trino/Athena/Snowflake
• Own retention and archival, including storage tiering, regulatory retention, immutability, and GDPR deletion
• Ensure data correctness through freshness SLAs, drift detection, and reconciliation against source wallet and ledger systems
• Own governance, including catalogs and lineage, row/column access control, PII masking, encryption, and audit trails
• Monitor ingestion health, data anomalies, and cloud storage and compute spend

• 2 Years of hands-on experience in a production lakehouse environment
• Experience with lakehouse architecture, bronze/silver/gold layering, an open table format (Iceberg, Delta, or Hudi), and schema evolution
• Experience with data layout and query optimization at TB+ scale, including partitioning, compaction, file sizing, and query performance on Trino/Athena/Snowflake
• Production experience with a cloud lakehouse/DWH: Snowflake, Databricks, or BigQuery
• Experience with CDC and streaming ingestion using Kafka and Debezium or equivalent, including late, duplicate, and out-of-order events
• Strong SQL and data modeling skills, with relational knowledge to reason about captured OLTP systems, especially financial ledgers
• Experience with correctness practices, including freshness SLAs, drift detection, and reconciliation against source wallet/ledger systems
• Experience with governance, including catalogs, lineage, row/column access control, PII masking, retention, and GDPR deletion
• Experience with cloud object storage (S3 or GCS), storage tiering, and archival
• Python and an orchestrator such as Airflow or Dagster

• Hybrid schedule: 2 days per week from the office