Develandoo team participated in the Smart People Innovation event, held on November 21-22 at the Brody Studios in Budapest. It’s already the 4th year this event brings traditional corporates, startups and venture capitalists together to talk about digital ecosystems, collaboration, and company building. The 2 day event gathered more than 250 participants around case studies, product demos, and panel discussions.
At the beginning of the event, Develandoo’s CEO, Albert Stepanyan, introduced the company and the team’s presentation topic, ”Can AI get a mortgage?” within which we talked about our latest banking AI product Protogen.
In particular, our teammate from the Netherlands, CEO of Protogen, Martijn Imrich, presented our new product, its aim, and benefits. Imrich explained, credit risk rating and the underwriting process is a hard and tedious routine, which often requires a lot of information to be processed and scored by human labor.
So what is Protogen AI and how does it work? It represents a paradigm shift from the conventional approaches to AI-based credit risk rating enablement. It is based on deep learning models for credit default risk prediction. It takes into account social and open banking data to build credit history patterns including, a sentiment model for evaluating real-time customer relation status. It does this by having a recommender system that offers relevant cross and upsell suggestions, and it can be extended with behavioral models based on customer’s actions in real time.
Speaking about its uniqueness, Protogen’s CEO touched upon its accuracy vs speed ratio. He explained, it can reach an 80% test accuracy on the Home Credit default dataset, found on Kaggle, and can ensure a high speed of prediction where the inference is made instantly, since it doesn’t use any kind of blending of training and test data.
”Protogen hooks into your core rating process, by assessing the credit-related steps and automates the credit risk rating process with full lineage and audibility,” says Imrich.
He also spoke about the benefits of the product among which is the reduced time to market for credit risk models through streamlined implementation and deployment processes, the incorporated best practices in model development including support for sensitivity analysis, the enhanced model monitoring, benchmarking, analytical reporting, the improvements in risk rating process efficiency, and straight through processing.
The audience was quite interested in the talk and the product itself as during both the talk and after, people were involved in an interactive discussion.
Did it also grab your interest? Then we suggest you dive deep into the video of our team’s talk on how AI can get a mortgage.
- Artificial Intelligence