Inspiration

During these tough times, with record-high inflation levels and low financial literacy, FinHelp is aiming to raise awareness for financial education.

What it does

It guides the user through a jurney of financial assessment, which enables the user to be profiled. The questions are simple to answer and target areas of high financial significance. The data gathered will then be processed and based on the answers, FinHelp will recommend personalized financial instruments, suggestions and objectives to reach your goals. All powered by AI!

How we built it

We integrated with GPT-4, the latest AI engine provided by OpenAI. Seamlessly deployed to Amazon EC2 as a docker container.

Challenges we ran into

Initially, we did several brainstorming sessions in order to come up with an effective profiling, but as we learned to use the LMM, we made it come up with its own questions (after many training sessions). We wanted to make every piece of financial advice to come from the AI.

Accomplishments that we're proud of

We made full functional flows for distinctively profiled users, ensuring all types of users can make use of out tool! We even made it work with silly or unintelligible inputs from the user.

What we learned

  • We worked hard to understand how to make the AI do the work for us!

What's next for FinHelp

  • Add some more reading resources
  • Add some tools for the users (eg: budget tracker)
  • Motivators for the user
  • Improve the model with longer memory
+ 14 more
Share this project:

Updates