Bacancy Technology
32 Case Studies
A Bacancy Technology Case Study
The US consumer lending and card issuing firm faced significant challenges with fraud detection due to siloed data from three separate systems, hourly batch scoring that caused delays, a lack of data lineage, and costly data retention. This prevented real-time risk analysis and created compliance issues. Bacancy Technology was engaged to address these problems through their data engineering and governance services.
Bacancy Technology built a governed streaming lakehouse on Databricks and Delta Lake, using Kafka for ingestion and Unity Catalog for governance. This solution consolidated the data sources, enabled fraud scoring within seconds of each transaction, and provided full column-level lineage. The results included reducing fraud scoring latency from an hour to under a minute, unifying three data sources into one lakehouse, and maintaining years of queryable history on low-cost storage, allowing the client to stop fraud in progress and meet regulatory requirements.
a US consumer lending and card issuing firm