Dexter is our proprietary statistical modeling and machine learning-based analytics engine that forms the core of KapitalTech.
It began as a simple rule-based decision support system using just 50 data points. With the deep credit insights that KapitalTech’s on-ground team has built over time, Dexter rapidly transformed into a highly robust engine to underwrite Indian businesses and an extremely flexible engine for customer risk and reliability score computation.
Today, Dexter takes into account over 2,500 data points simultaneously to deliver almost real-time appraisal of potential borrowers. Leveraging simple, yet profound analytics techniques viz. pattern matching, dimensionality reduction, probabilistic record linkage, etc., Dexter blends a wide range of data sets from multiple sources to reduce them into comprehensibly fewer number of metavariables.
Metavariables are processed through our proprietary binary modeling and credit analytics algorithms to generate optimized prediction for automated decision making. Dexter’s capability to judge accurately the ability and willingness of borrowers to repay minimizes the perceived risk, while extending the credit spectrum.