Inspiration
As programmers who all work in areas unrelated to finance, we wanted to make a simple project that could help us understand what financial documents and terms mean in context.
What it does
By the time the project is complete, we hope to have a simple website with 3 to 4 tabs, with the pages being a simple homepage, a chat bot where documents can be uploaded and analyzed, a page that stores uploaded documents, and an optional sign in page for those who want to store their documents.
How we built it
We decided to experiment with using AI to build entire structures rather than smaller chunks of code. We specifically used Google Gemini to help us build the base for our chat bot and optimize it to respond to financial questions rather than work as a copy of Gemini. We used primarily used React.js, JavaScript, Node.js, HTML, and CSS to build the structure of the website and planned our UI using Figma.
Challenges we ran into
The main issue we found with using AI to code without already having a background in a specific language or framework was that we couldn't easily identify what was going wrong even when an issue was very simple or constrained to a single character. We also started working on separate devices without starting from the same GitHub repository, so we struggled to put our code together.
Accomplishments that we're proud of
We are proud of the integration of Gemini into our website and having a simple website which is easy to navigate.
What we learned
We've learned a lot about how coding with AI works, React.js formatting, and how to code/refine a chat bot.
What's next for FinDocs
We want to improve the chat bot functionality to include advice on how to act in response to relevant data.
Log in or sign up for Devpost to join the conversation.