mAe.I

Inspiration

We were inspired to create an easy-to-use application to help people figure out if they are ready to purchase a home. Our hope was that we could help anyone, regardless of their financial status, to make progress toward achieving their goal of buying a home.

What it does

mAe.I is a full-stack web application that takes in a user's financial information such as credit card payments, estimated mortgage payment, etc. and creates an AI generated response explaining whether a user is ready to purchase a home. The response would also provide advice for improving certain metrics such as credit score to help prepare a user to purchase a house in the future.

How we built it

We used Next.js, TailwindCSS, and the OpenAI API to develop an application that takes user input and engineers a prompt that is passed to a Large Language Model that analyzes the tokens to generate a set of suggestions to help improve a person's credibility for buying a home.

Challenges we ran into

Due to the time constraints, we had to significantly downscale our ambitions and use OpenAI's API instead of fine-tuning our own model to provide responses more catered to our use-case. We also faced challenges representing data of other buyers in a clear and concise manner.

Accomplishments that we're proud of

Our UI design and the accuracy of our suggestions for users are accomplishments we are proud of.

What we learned

The team collectively improved at front-end development skills as well as prompt engineering for use by a Large Language Model.

What's next for mAe.I

Incorporating fine-tuning of Llama, an AI model from Meta, will help us make significant improvements in the detail of suggestions for a user. Displaying more data in multiple formats will also help users get a better idea of how individual factors affect their ability to buy a home.