Inspiration
We realized how much of a pain choosing the right movie for two people is, and we wanted to do something about it.
What it does
Users connect with a partner, and after rating a few movies, our service provides them with a list of movies that will maximize their enjoyment.
How we built it
Using web sockets, artificial intelligence and full-stack web development.
Challenges we ran into
Trying to deploy our AI model onto Heroku, our chosen deployment platform. The dependencies we were using were far too large, so Heroku wasn't willing to host them. Looking back, we should've used an external hosting service like AWS.
ADD IN CASE: In the end, we had to mock the model as it was simply too large for Heroku.
Accomplishments that we're proud of
Learning sockets from scratch and successfully implementing them. For most of us, it was the first time building a full-stack application from start to finish, so we're proud of how visually appealing and user-friendly our web-app turned out.
What we learned
We learned a lot about deployment - we realized how underrated this task can be, as we faced way more challenges deploying than we did coding. We also learned a lot about sockets, and our "AI guy" in our team really enjoyed his challenge of building such a complex structure
What's next for MatchFlix
Finishing the AI model and adding an authentication system so users only have to take the rating quiz once. Then, users will be able to add their friends and immediately get a list of movies between them that can be updated by our AI as it gets more training data.
Log in or sign up for Devpost to join the conversation.