Ganit

The client was operating on a legacy data architecture that lacked scalability and did not support unstructured Loan Origination System (LOS) data. Critical loan origination and loan management data remained siloed across multiple systems, and the organization did not have a centralized data warehouse to support enterprise-wide reporting and analytics.

Ganit modernized the client's data infrastructure by building a centralized Data Warehouse on AWS to eliminate data silos and support long-term scalability. The implementation unified Loan Origination System (LOS) and Loan Management System (LMS) data, deployed automated data pipelines with governance and authentication, and leveraged Amazon S3, Amazon EMR, AWS Glue, Amazon Athena, Amazon QuickSight, AWS Lambda, AWS Database Migration Service (DMS), AWS Identity and Access Management (IAM), and Amazon CloudWatch.

The client needed to migrate its entire data infrastructure from Azure to AWS Databricks while managing complex dependencies across hundreds of enterprise data assets. The migration required maintaining row-level accuracy, preserving historical data, and avoiding disruption to ongoing analytics and logistics operations

Ganit designed and implemented an automated, dependency-aware migration framework to move the client's data infrastructure from Azure to AWS Databricks. The solution addressed dependency mapping across more than 500 tables, over 100 views, and 30 data pipelines while maintaining row-level validation across 11 nested models. Historical data synchronization was implemented using AWS DataSync to prevent data loss, and compute and storage performance were optimized through redundant table removal and dataset consolidation.