Inspiration
What inspired us for this project was the vast availability of unused cars that have a lot of functional parts. It also consumes a lot of time to disassemble, locate, and post the parts of those vehicles online, not to mention whether it will be sold or not.
What it does
The application helps the users do the tedious work of posting those unused parts online. The users have to enter their vehicle make, model, year, and color to get a list of parts of the vehicle. The part names, images, and estimated prices are to be fetched by AI. Then the user selects what parts they want to list and adjust the pricing. The app then posts the parts onto eBay and the website itself.
How we built it
We mainly used Reactjs for the frontend, and django for the backend. The complete app utilizes vision related AIs to find and detect the correct images for the auto parts.
Challenges we ran into
We ran into the challenge of integrating Google Cloud Vision and Microsoft Vision API onto our application. We then decided that it will be a better idea to train an AI model that can accomplish the desired task.
Accomplishments that we're proud of
We are proud to have overcome the small challenges that came along related to coding. We had several issues with reactjs which took a considerable portion of our time. Unironically, AI language models came in to our help when we were stuck at those sections with issues.
What we learned
We definitely learned how difficult it is to develop a near complete website in a span of 2 days. Other things we learned include getting familiar with the Google Cloud and Microsoft Azure interface.
What's next for eParts
The full version of the website is expected to have a fully functional marketplace, payment, seller-buyer contact, chatbot, and several other features including the options to post the listings on Amazon and other online platforms.
Log in or sign up for Devpost to join the conversation.