Inspiration
Staying productive during long work sessions can be difficult, and the Pomodoro method is a proven way to increase efficiency. The goal of this project is to not only have a usable Pomodoro timer web app but also to use AI as a way to increase engagement and enjoyment during breaks.
What it does
The project is a standard web app with a customizable Pomodoro timer. Additionally, the project obtains user inputs and provides relevant music recommendations to the user during break periods.
How we built it
We used nextjs, typescript and tailwindcss for the frontend and backend. The music genre model was developed and trained in Pytorch using roBERTa and set up using Docker and a Flask API.
Challenges we ran into
There were challenges in developing the music genre model as the lack of data made it difficult to train a good model. There were also challenges in implementing a working database and implementing the settings to customize the timer length.
Accomplishments that we're proud of
The Pomodoro timer is visually appealing with a clear minimalist theme and is user-friendly. Additionally, we integrated a machine learning feature into an app with a real-world application.
What we learned
We learned that despite simple web apps like a Pomodoro timer being seemingly easy to implement, in actuality, many components go into a user-friendly web app and have to seamlessly work together.
What's next for Code Monkeys - pomoAI
Improve the music recommendation model, add more break recommendation activities (e.g. basic exercises) and include additional timer features (e.g. the ability to create multiple timers).
Log in or sign up for Devpost to join the conversation.