Inspiration
Other applications of AI, combined with this competition being finance-related, brought about an AI financial analyst.
What it does
Recieves a person's portfolio and a news (or informational) document, then presents an analysis of how different sectors will react, as well as a buy-hold-sell analysis of the uploaded portfolio.
How we built it
We first brainstormed to figure out what our solution would look like and what we needed to incorporate. Once we finished the brainstorming session, we split up and did individual research. We then figured out how to set up an AI agent of a Gemini model through Agno with a Google API. We then worked on the backend code to allow file uploads and add some other features. After finishing the backend code, we worked on the HTML script to make the webpage look good. We finally had to incorporate the two together to create the functioning site.
Challenges we ran into
Hosting the website, integrating all the code together to complete the site, changing the code due to changes in ideas, and changing the code again after moving from a Llama AI model to a Gemini model.
Accomplishments that we're proud of
We're proud to have ended up with a functioning website that does what it was designed to do. The UI (frontend) looks great. We're proud of its versatility, diverse set of functions, and its capacity to go beyond just the requirements of this datathon.
What we learned
We learned how to host a website through AWS, so we got some experience with Amazon products. We also learned lots about having files as inputs and how to incorporate this feature.
Log in or sign up for Devpost to join the conversation.