Inspiration
We were inspired by a drive to make loans easily accessible to everyone as well as to revolutionize the fintech world.
What it does
Our project allows banks to make accurate predictions on the credit risk of people with no previous credit history, by analyzing debit card transactions.
How we built it
We used TensorFlow to cluster transaction data and label users as low or high credit risk, then used convolutional neural networks to create a model to predict whether a person was low or high risk.
Challenges we ran into
We initially had many problems with implementing the neural networks, as well as deciding on specific design choices. However, we were able to overcome these challenges through hard work and teamwork.
Accomplishments that we're proud of
I'm very proud of my time for creating such a powerful tool in such a short amount of time at this hackathon. Not to mention the fast learning abilities of our time, especially in creating a usable, good looking interface using next js.
What we learned
We learned much about handling large amounts of data, as well as creating neural networks in tensorflow and designing ui with react and next js.
What's next for Credless
If the judges like our pitch, hopefully we'll become a business someday and truly revolutionize the credit and debit card world!
Log in or sign up for Devpost to join the conversation.