AidenAI: Your AI Financial Aid Advisor

The financial aid process is complex and difficult — but AidenAI harnesses the power of Google Gemini to help students from all backgrounds make informed decisions, and negotiate for better aid.

Comment 2

The Gist

Aiden AI harnesses the power of Google Gemini to analyze official scholarship offers and financial documents, in order to help students make a financially smart and informed college decision. The Aiden AI website allows users to upload the deluge of financial communication they receive from colleges, and understand their situation easily through an informal conversation with a helpful AI assistant.

Inspiration

As college students, we have all had to navigate the extensive and cumbersome financial aid process. We know how difficult it can be to read through large legal and financial documents, understand the complex terms, and decide which offer is better. We were inspired by the power of generative AI to provide people with new insights, and so AidenAI aims to make this process easier and more affordable to those who can’t afford a financial analyst through a personalized AI chatbot.

What it does

The AidenAI website allows users to inform Aiden of their background, finances, and educational goals, and differentiates itself further by allowing users to submit official aid offers, FAFSA documentation, and communications from school financial aid offices. With a generative AI targeted specifically at parsing these complex financial documents, AidenAI helps students to:

  • Unpack the financespeak by talking with Aiden
  • Learn the process of negotiating for more financial aid
  • Ask the what-ifs and hypotheticals that will inform their eventual decision

How we built it

We started by drawing mockups of the various pages and UI elements that would make up the outward-facing website, as well as connecting buttons to the database actions that would need to be taken. We settled on using the FReMP stack to build this application: Flask, React, MongoDB, and Python. We also used Vite for rapid prototyping of UI as well as bundling the app for Flask hosting. We also used the CapitalOne dataset for providing the AI with mock financial statements as well, and of course Google Gemini for powering Aiden himself.

Challenges we ran into

There were many technical challenges that came with incorporating so many technologies and libraries into a single project — specifically, we struggled to establish an effective development cycle for backend work, which needed to simultaneously access the external MongoDB database as well as our internally-hosted (fully RESTful!) API. With regards to teamwork and integrating our separate pieces of code successfully, we had some issues with branch merging, especially where multiple people were modifying our application’s shared stylesheets. However, we were able to avoid this issue for the most part by using effective version control techniques.

Accomplishments that we're proud of

We are extremely proud to have developed a fully functional, visually pleasing, and (hopefully!) useful application in only 36 hours! Each member was able to apply their own unique skillset, and each team member constantly discussed how their decisions could impact the tasks other members were completing, which helped us avoid grief down the line. Our team had extremely minimal hacking experience, so it was extremely rewarding to end with a well-designed and fully functioning product at the end of our 36-hour marathon!

What we learned

Our team learned a lot about Git and proper version control, as well as the need for great project file layout, in order to prevent merging issues as much as possible. Additionally, everyone greatly expanded their knowledge of React and Flask, and our backend engineers also learned how to use MongoDB Atlas and Cloudflare.

What's next for AidenAI

As we continue to develop the project, we hope to support more file formats for document upload, and also to work on fine-tuning Aiden’s responses so he can better aid students. Focusing on more minute fixes by improving the graphics and general UI/UX will also be an area of improvement.

Share this project:

Updates