SpotDrop

• 

  Logo

• 

  Landing Page

• 

  Check In Form

• 

  Dashboard

• 

  Location Page

Inspiration

We were inspired partially by the Transit app's community-driven information on bus crowding. We wanted to take this concept and apply it to study locations to solve a shared problem we had of trying to find the best study spots without having to check manually.

What it does

SpotDrop shows you study spots near you, how busy they are throughout the day, and community reviews and photos, so that you can find the perfect spot to study before you even leave.

How we built it

We built our backend in Go using go-chi for routing and the mongo driver for DB interactions. We used MongoDB Atlas for our database, including GridFS for large file storage (photos) and MongoDB's geo features for querying based on location. The front end was built using TailwindCSS, Typescript, and SvelteKit. We used SvelteKit for server-side rendering and some processing of data before it reaches the browser. We used Railway hosting for automatic deployments whenever we merged PRs on GitHub.

Challenges we ran into

One of the main challenges we ran into was staying organized while four people worked on the same project. We separated into front-end and back-end, but even then, we were stepping on each other's toes, which slowed things down a lot, especially near the beginning. Once we built out our core functionality and focused on modularity, development went a lot smoother.

Accomplishments that we're proud of

Three out of four members of our team had never done a hackathon, and getting to the finish line with a working project is something that we are all proud of.

What we learned

Rapid implementation was one of the biggest lessons from this project. We learned that generative AI is most effective not when writing entire systems from scratch, but when paired with strong architectural foundations and well-designed boilerplate. By first establishing clean, repeatable patterns, especially through concise, reusable functions and strict layer interactions, we were able to use Gen AI to scale that code horizontally across the codebase. Instead of repeatedly hand-writing similar implementations, we used AI to extend proven patterns quickly and consistently. This significantly accelerated development while reinforcing good software structure, and showed us that Gen AI can act as a force multiplier when combined with disciplined engineering practices.

What's next for SpotDrop

We'd like to continue adding features as well as improving existing features that we decided not to pursue due to the time constraint, such as signing in with Google or other SSO providers.

Built With

• go
• mongodb
• railway
• sveltekit
• tailwind
• typescript