It's always hard to find a restaurant that perfectly suits you, and browsing Yelp right before mealtime can be such a chore. By taking a page out of the e-dating's handbook, we've created Meet Your Eat -- a fun, engaging experience to find a tasty dining establishment nearby that perfectly matches your tastes!
Try it out here: meetyoureat.online
What it does
The app gives you a personalized selection of restaurants, which you can further refine through the tried-and-true e-dating experience.
tl;dr Tinder for restaurants
How we built it
The technology stack was React for the front end and Flask for the back end.
Infrastructure ❤ GCP
- For the persistent datastore, we used Firebase Cloud Storage.
- For web-hosting, we used Google Cloud Deploy.
- For a recommendation prototype, we used a Cloud Function.
- For restaurant information, we used the Google Maps Places API.
Challenges we ran into
We had different ideas about how we should deploy the different parts of the app. Luckily, our design was modular enough that we could adapt both deployments to fit together cohesively.
There was an attempt to make a recommendation system. However, the performance and implementation of the system weren't quite there, so currently nothing in the frontend or backend calls this system. Luckily for us, because we don't call the Cloud Function, it incurs no cost!
Accomplishments that we're proud of
We're proud that we were able to make a fun, interactive experience in such a short amount of time.
What we learned
- How to deploy an app server and a backend server via Google Cloud
- How to use React Springs and Gestures for good looking animations on the frontend
What's next for Meet Your Eat
- Implementing a real machine learning framework to learn the types of restaurants the user likes/dislikes.
- Collect temporal locality (like whether it's nearing a holiday, or nearing supper time) and add that to the recommendation system.