We mine data from several sources, spanning 48 million loans issued to millions of individual consumers. We cross-reference that data across multiple streams to produce clean, enriched credit datasets.
We then use that data to train sophisticated deep learning models in order to assess credit risk in a variety of contexts. Our models allow us to better discriminate between performing and non-performing loans.
We leverage our models to make smart credit decisions that result in robust, performant credit portfolios.
We work with a variety of data types and data sources to build credit models for any consumer credit product. Traditional credit data, macroeconomic data, transactional data, unstructured data: you name it we work with it.
We work fast to develop practical and effective solutions for any problem involving credit.
We are not hamstrung by traditional approaches to credit risk, which has allowed us to develop deep learning techniques to create state-of-the-art credit models.
We manage client investments in alternative lending platforms. Using our proprietary models we actively select the best loans available, focusing on building profitable portfolios resilient to macroeconomic stress.
Our proprietary credit rating system helps our partners price loan products for any customer in the United States.
We develop bespoke credit models for select clients and their specific product lines.
This graph shows the relationship between two of the many variables we use in our modelling, and reveals a unique insight. Can you guess what it is? Contact us if you want to find out.