KickFlips AI

Inspiration

Something about me is that I conduct a side business where I buy and resell rare sneakers. To provide a brief background about the sneaker market, one can think about it like baseball cards or stocks. There are thousands upon thousands of shoes that exist, each having a different value for different reasons (i.e., color, limited production, collaboration with a celebrity, etc). As a reseller, how do you know which shoes are good to resell? Also, how do you know what to sell the shoe at to sell it quickly while still maximizing profit? Sure, some tools exist, such as recent sales on eBay or StockX, but there is no way to tell how long it stayed in someone's inventory, and there are no tools to help you determine a good buying price to maximize ROI, which is where my idea for KickFlips AI comes into play.

What it does

This website serves as a portfolio for resellers to track their sneaker investments. First, a user will create a login. Once this is done, they can start to add sneakers to their inventory with information such as purchase price, purchase date, size, condition, etc. Once a shoe has sold, they can mark it as sold and input any fees they may have had to pay. There is a tab on the top of this page that displays stats, such as average hold time, total revenue, etc. Another tab in this website is the "search" tab, which allows you to search for a specific shoe, such as "Jordan 1 Chicago". This will display all of the sales from other users on this website, which will show information such as their purchase price and hold time in their inventory. An AI will also be at the bottom of the page to offer insights, such as the performance of the shoe for a reseller, and whether or not it's a good investment. It breaks down recent sales and compares sizes, and tells you what price you need to buy it at to sell it quickly and make a profit.

How we built it

For the framework, we used Next.js and used Tailwind for the frontend and styling. For the backend, we used JavaScript and Supabase to store all of the data, such as shoes and users. We also used an API from Gemini to incorporate it into our program to analyze this data.

Challenges we ran into

The challenge I ran into was connecting the database to my code. It ended up being a policy program, where I needed to allow any user to modify the database. Another challenge I ran into was connecting the AI to my code. More specifically, it was tricky feeding my data to analyze. Instead, it would give general advice.

Accomplishments that we're proud of

I am proud of myself for trying a brand new tech stack. I was unsure if I would be able to get anything running, but I actually ended up executing my entire plan. This was also my first time using AI in a project.

What we learned

I learned how to use Supabase, Tailwind, Next.js, and AI in a personal project. I also learned time management skills, as I was working solo.

What's next for KickFlips AI

Next step is to offer more analytics that are useful for resellers. I also want to incorporate more filters in the search section to allow people to find the stats they need more quickly.

Built With

• api
• gemini
• javascript
• next.js
• supabase
• tailwind