SERVICES | SUCCESS STORIES | FEDERAL LENDER

JUNE 22, 2021
DATA MODERNIZATION, OBSERVABILITY AND MIGRATION FOR FEDERAL LENDER

Emergint's Data Modernization team was asked to complete a four-pronged modernization for a large U.S. lender.

Create intraday, real time insights of mortgage loan performance:

  • Remove slow batch jobs running against a legacy MIDAS core banking system

  • Use Qlik/Attunity to capture loan performance metrics and foreclosure events via Change Data Capture (CDC) against a mainframe

Modernize away from traditional Ab Initio ETL/ELT to PySpark-based ingestion within a hybrid, Big Data footprint

 

Orchestrate end-to-end pipelines for credit risk assessment of loan applicants:

  • Convert a large code base of legacy SAS code to Scala and Python

  • Use a containerized architecture in Kubernetes

  • Extract credit card transactions from a TSYS supported mainframe

  • Create a machine learning pipeline using Kubeflow to manage data ingestion, feature extraction, and gradient boosting

 

Migrate away from legacy IBM DB2 z/OS footprint to a Snowflake Cloud Data Warehouse