Credify Uncollateralized Lending & Credit Score

The global web3 market has reached record heights with Bitcoin and Ethereum showing a rapid growth of 9200% and 33500% respectively (as per the data from 2016-2022) and more people trying to become a part of it, hence, it won’t be an over exaggeration to say that uncollateralized lending is no longer far out-of-term.

This project will benefit the user which has very less or no asset to give as a collateral and still wants to opt for a loan in the same manner. The method ensures that the borrower is at the side of lower risk but on the other hand, the institution needs some metrics to risk analyze the profile of the borrower, that is exactly where our algorithm for calculating the credit score for the user comes into picture.[financial benefit to the borrower, while also taking care the lender can lower the risk by looking at the metrics of risk profile to take a calculated risk to earn high yields The process becomes efficient and faster as the lender can analyze the credit risk profile of the applicant and approve the loan faster. Through uncollateralized means, the loan can be applicable to wider use-cases other than automobile and real-estate.

Since DeFi is still a very young and evolving space, trustless lending is something which involves its own set of unique challenges! Conceptual Challenges The first challenge that we ran into was the issue of not being able to identify the user-uniquely (we thought of using Polygon ID Connect, but it doesn’t exactly solve the issue at hand).
Managing users who default their repayments on the platform. This in a way is linked to the first point of not being able to uniquely identify the user (in a scenario where a defaulter makes a new wallet and uses that address to disguise as a new user).

Technical Challenges At the heart of the project, lies the algorithm which analyzes the wallet history of a web3 user and tries to compute the credit score without bias. Designing this algorithm was our first hurdle which we faced along the way. However, after focusing on few important metrics obtained from the wallet transaction’s history of the user, we were able to narrow down the procedure based on which our platform assigns the score (to a new user), for returning user, it is recommended to borrow/lend using the lending protocol built hosted on the smart contract of Credify, for better assessment of the risk associated with that user. Preventing bot intervention and manipulation of the algorithm by borrowing and repaying in split minutes to get a better credit score. We tackled this challenge by introducing a ‘cool-down period’ of 5 minutes after the amount is borrowed to the borrower which locks him to do any transaction during that time.