Situation at the Project Launch
The bank has a 30-year track record in corporate, mid-market, and retail banking. The bank’s team has extensive experience in M&A transactions, asset consolidation, and provides financing and advisory services for complex, structured deals.
With the rapid growth of the auto lending direction in 2020, the performance and flexibility requirements for the existing Decision Engine for Lending increased significantly. The old system could not cope with the growing volume of applications, had scalability limitations, and required significant manual involvement. This hindered operational efficiency and limited further business growth.
The transition to a new system became a logical step in the development of the bank’s digital infrastructure.
Assigned Task
The bank set the following goals:
- Implement a modern and scalable Decision Engine for auto lending;
- Ensure the ability to configure business logic independently, without involving developers;
- Maintain integration with both external and internal data sources;
- Complete the project entirely remotely — from contract signing to production launch.
Implementation
The bank acquired the low-code 2RS Platform as a tool for internal, independent development and process configuration. This significantly reduced the “time to launch” and allowed the bank to maintain control over business logic on the user side.
The project was fully implemented remotely within 5 months — all stages, including training, prototyping, and deployment, were carried out online.
Achieved Results
The project demonstrated how a bank can within 5 months and without external development, deploy a powerful Decision Engine system fully tailored to its needs while building internal low-code development expertise. This became an important step in the client’s digital transformation and gave momentum for further product scaling.
- The system is operating across all 36 bank’s branches and processes applications without manual involvement at the initial stage;
- Over 1,900 unique loan applications are processed daily, with peaks of 25+ applications per minute;
- The system integrates with 50+ external data sources and supports 40+ underwriting strategies;
- Bank business units independently manage filters and processing rules, without involving IT.
Development and Outlook
- Vendor support is used only when updates to the platform itself are needed — not for custom project work;
- The platform is now used not only for credit risk management, but also for other internal systems, expanding the bank’s product ecosystem;
- With each new platform release, additional capabilities are unlocked — including Python script integration, enabling advanced functionality and flexibility.