We took our inspiration from the decentralized nature of Web 3.0. Zapit will allow individuals from all walks of life to lend and help individuals in financial difficulty. Similar services available charge individuals unsustainable rates of interest which does not address the problem of financial difficulty for the borrower. Our credit score card system allows for identification of credit worthiness to ensure investor funds are not at risk through making use of the service. We took the UI inspiration from modern dating applications with investors and borrowers given the option to select a potential match depending on the interest rate, term and amount of the loan.
What it does
Zapit takes a number of variables input by potential borrowers and investors and aligns their needs based on matching inputs. Zapit creates a peer to peer network of individuals will to invest in potential borrowers and enables lending across boarders. Our machine learning function addresses the risk associated with lending and determines an individuals credit worthiness based on a range of factors taken into consideration for our score card model.
How we built it
We built Zapit using the Django web frame work, Microsoft Azure web application and Postgres database on Azure. Our machine learning model has been created using Jupyter notebook. Our market research conducted via an anonymous survey provided results from 80 individuals and provided feedback that individuals are willing to lend to others via platform. Predominant feedback was risk aversion and methods of capital being collected. Zapit services this market through our gig economy model.
Challenges we ran into
Our challenges related to our Machine Learning model and the ways in which to best score individuals based on previous lending.
Accomplishments that we're proud of
We are proud to have achieved our goal of making use of the Azure platform the leverage our product. We have a deeper understanding of technologies required to progress our platform and our Machine learning model will aid in risk mitigation.
What we learned
We learnt much about credit scoring and Machine Learning and methods by which to rate credit worthiness. Our ambition is to enable lending to multiple regions and allow capital access to areas not serviced previously. Leveraging the peer to peer aspect of Web 3.0 we are confident in our approach to Zapits use case and potential.
What's next for Zapit
Next for Zapit is to integrate payment processing and further develop our Machine Learning model to mitigate unnecessary risk associated with the platform and ensure lenders and borrowers feel comfortable using the service.