Inspiration
The used car market has gotten especially gnarly recently due to countless world factors, but the demand and desire to own a car hasn't weakened. We wanted to create CarGo to help people find cars in their area more easily, focusing on the most important criteria to the average consumer and scouring a part of the market nobody else has tried at scale.
What it does
CarGo is a Discord bot that searches Craigslist listings for used cars within a user-defined radius. The way it differentiates itself from the rest is that it searches multiple Craigslist boards for sellers in this radius, and also allows for searching by parameters that are important to used car buyers: mileage, price, and more. The user enters their preferences as the bot prompts them, and when it's whittled the results down to a managebable number, it returns a short, bite-size list of links for the user to peruse at their leisure.
How we built it
CarGo was built on the Nextcord framework, Discord's official method for building Discord bots using Python. Our script iterates over a dataset of all Craigslist listings in the United States and filters it down by the given parameters one by one. We then hosted the bot on a spare Upcloud instance that one of our group members had.
Challenges we ran into
We initially wanted to make CarGo an app, but lack of experience on the part of our team prevented that. The pieces are definitely there, and the format would have absolutely suited it, so it's a bit of latent potential. The server we hosted on is also a bit slow, and because databases are an inherently computationally expensive application, the user experience is a bit degraded. It would have been nice to have a more powerful server. On the other hand, the coding process itself went smoothly.
Accomplishments that we're proud of
We're especially proud of our fighting spirit. Our team consists entirely of freshman-aged members who were relatively unskilled with coding, and we came together to come up with this idea. We fought through the disorganization and the roadblocks of knowledge to build CarGo from the ground up in record time. Though we lacked budget to host an expensive server, we still made optimizations to make both the code short and readable and the end product run in a reasonably width of time. We're also proud of the distance formula we wrote. Calculating the straight distance between a zip code and the approximate latitude-longitude of each listing was no small feat, and the formula is something to behold.
What we learned
We learned about how hackathons work. This was the first hackathon for all our group members, and we learned a lot about the process of developing an idea with such little time. We hope to come back the second time with an even bigger and better idea, and with more knowledge under our belt.
What's next for CarGo
This is probably the end of the line for CarGo as a Discord bot. If we do plan to develop it further, it would be in the form of a mobile app, and with a nicer UI. We feel the idea is not only sound, but also temporally relevant, so we may indeed consider going down this route.
Log in or sign up for Devpost to join the conversation.