[ Avengers ] - Credit Assessment Microservice

Inspiration

Singapore's current process of assessing and approving loans is still not optimized, which creates obstacles for both borrowers and lenders. Given the problem's urgency and its consequences on economic development, our team believes that Machine Learning can be applied to simplify the loan application process and give users a holistic view of each other's credibility and capability.

What it does

The API is able to help us classify credit score buckets together with calculations on the expected loan amount for SMEs and prediction the possibility of whether a consumer’s loan application is accepted or not. The purpose of this API is to facilitate paperwork processes for banks and also provide initiatives for borrowers by giving them tools to predict the amount of loan that will be given by banks and the possibility of being accepted.

How we built it

The FastAPI microservice and the two machine learning models are hosted on an EC2 instance. The API is protected by an AWS API Gateway and is to be consumed by the general client and our mobile application using REST API. In addition, users can interact with the microservice using Swagger documentation.

Built With

• fastapi
• flutter
• pytorch