The customer is a FinTech start-up providing consumer credit services for individuals. A complicated process of getting a consumer credit in traditional banks is one of the main features of the financial market where our customer operates. Because of this, the microfinance market is currently experiencing a sustainable growth. The customer’s company promptly took one of the leading positions in this area offering a wide range of high-tech services.
Challenge
Orion was asked to develop an enterprise-grade contact center from scratch, and it was for the customer’s debt collection department. Due to the rapid growth of their business, the client needed a solution with scalability and increased performance. Our team developed a solution of both front and back ends — all based on open source solutions. One of the requests from the client was that the solution include a feature for automatically placing a large number of outbound phone calls.
Solution
Our team achieved all the objectives in just three months, and the first official release of the enterprise-grade contact center was ready for launch when promised. The product was developed using a state-of-the-art technology stack and agile practices.
The core part of the solution used an open source platform that’s powered by a predictive algorithm. For optimal workload among all currently logged-in agents, the algorithm defines the number of outbound calls to be placed at every moment. The system performs calls, and once the subscriber answers the call, it’s immediately forwarded to an available agent. Additionally, using an early media analysis module, the product also defines the reason for unsuccessful dialing attempts — if the line is busy or if the subscriber is currently offline, etc.
We were asked to ensure the product’s stability, scalability, and high performance. Our team adopted an integrated approach and managed to improve the product’s scalability by revising the architecture of the open-source solutions.
Results
The solution was launched with 50 agents in less than 6 months, and it was successfully scaled to 3,000 agents working simultaneously. New agents are able to join the workflow as they please and in real-time due to the modern, user-friendly interface.
Major Functions:
- Feature-rich contact center solution
- Integration with third-party software and systems (CRM, Underwriting, etc.)
- Support for basic and advanced call processing features
- Advanced statistics and reporting
- Built-in-role-based access
Technologies:
Backend:
- Java EE
- Spring Framework
- Apache HDFS
- RabbitMQ
- FreeSWITCH
- ELK Stack
- Spark Streaming
- Kamailio
Database:
- Hbase
- Apache HB
- Redis
- Sybase ASE
- MySQL
Frontend:
- JavaScript
DevOps:
- Docker
