QuickVest AI

Inspiration

Our main source of inspiration for this project was definitely the Fidelity track. We thought it was really interesting because it wanted us to incorporate AI in such a way that it streamlines the entire account creation process, which is regularly not the main usage of AI. We also wanted to knock out two birds with one stone, while also keeping our project efficient, leading to our integration of the Google Gemini API. Setting up an investing account can honestly take a large chunk of time, and sometimes, people may not know exactly what they are getting into, so overall, we wanted to really fix this problem.

What it does

QuickVest AI streamlines investment account creation through a 7-step process: personal info, AI-powered identity verification (selfie + ID scan), document uploads, personalized investment recommendations based on goals/risk tolerance, and an always-available AI chatbot for real-time advice.

How we built it

Frontend: Next.js 14 with TypeScript, Tailwind CSS for Fidelity-inspired design, Framer Motion for smooth animations. AI: Google Gemini API (gemini-2.5-flash) for chat, recommendations, and vision-based identity verification State: React hooks with localStorage in React

Challenges we ran into

This project didn't go completely smoothly at all. There were definitely setbacks and bugs, like loading errors (an infinite loading screen), the user couldn't continue where they left off, there were some CSS issues, and even more. Also, integrating the Google Gemini API was a whole battle by itself. Although we did have some of these issues, we solved a good number of them to get our project where we want it to be, at least within the short window of the hackathon.

Accomplishments that we're proud of

We are pretty proud of our UI/UX design, as we think it makes the program flow really smoothly. We are also very proud of our Google Gemini API integration, as we were able to incorporate it into so many areas of our program, as listed in the What it does section. This really pushed forward the main focus of the program to guide investors and help them open accounts.

What we learned

We learned that prompt engineering is quite important for our requests to Gemini, as we don't want the user to receive essay-length responses when talking to the chatbot bot and we want the verification system to work properly. We also learned a little more about the technologies we used in our tech stack for this project.

What's next for QuickVest AI

Next for QuickVest AI, we want to actually use a database to store the data instead of putting it all into localStorage, as this data is meant to travel to other places where you can set up investment accounts. Also, we would like to incorporate more AI-based features, like an investment tutorial with AI feedback.

Built With

• css
• gemini
• next.js
• react
• tailwind
• typescript