HackStack

Project Info

Munch Match - Stateful-AI-Based Tinder for Restaurants

Devpost

Inspiration

We wanted to create a web app for people who have trouble finding new food and cuisines to try, and for those who are too busy to cook their own food but want to expand their palettes.

What it does

Munch Match allows users to swipe left or right on food locations near them, swiping right if they like it and left if they’re uninterested. Based on their swipes, our web app will recommend more locations that are similar to the liked restaurants and adapt accordingly. It also has a map marking all the restaurants you’ve liked, showing you which location it is and where it is relative to you.

How we built it

We used a React.js framework to build the full-stack web app connecting to a PostgreSQL database through the Supabase API. We began by constructing a basic .html and .js site that filtered through a sample set of restaurant data with a basic interface, restaurant information, semi-working images, and two buttons. We then built on this foundation for the remaining duration, implementing Letta AI to adapt and push out restaurant recommendations dependent on the user's feedback. After, we made a map section for the liked restaurants to see each of the locations, as well as a page to see which restaurants the user as liked.

Challenges we ran into

One of the primary challenges was the wifi issue that impacted our ability to collaborate. Instead, we utilized the time where the wifi was down to focus our efforts on ideation, brainstorming, and outlining our project plan. We also struggled with database management as it was most of our first times working with databases. After hours of debugging and reading documentation, we were able to have reliable loading image files drawn from the Yelp public database.

Accomplishments we're proud of

We are proud to finish our first real hackathon project. Our team has spent a lot of time on building our product and knowing that we built something that was functional and useful is amazing.

What we learned

We learned a lot about how to collaborate as a team. It was difficult at first to decide on an idea that everyone was passionate about and find ways to split up tasks, but it ended up being very fun! We also learned to quickly adapt and learn things we previously haven't even heard of or interacted with (i.e. Supabase and AI implementation).

What's next

We’d like to add a section where users can connect with their contacts on their phone, allowing the contacts to see which restaurants they’ve liked, creating a gateway for a social aspect on the web app. From there, the users can make plans with their contacts to go and try those new restaurants together if they choose to.

Analysis

Compare with all teams

View

Metric

Figures cover GitHub contributors during the hackathon window. A co-authored commit counts in full for each author, so per-member totals add up to more than the whole-team figures.

Technology

Found in codeClaimed only
  • CSSIn code
  • HTMLIn code
  • JavaScriptIn code
  • ReactIn code
  • SupabaseIn code
  • Node.jsClaimed
  • PostgreSQLClaimed

5 of 7 appear in the indexed code. 2 claimed on Devpost could not be matched to code, which may simply mean the tool leaves no trace in the repository.

AI coding agents

No AI coding agent signals were found in this repository.

Detected from committed agent config files and commit authorship. Absence of a signal is not proof an agent was unused.

Codebase size

Source size

89 KB

Source files

16

Counts recognized source files only; vendored directories, binaries and lockfiles are excluded, so this is smaller than the repository on disk.

0 stars