Foca | Devpost

Inspiration

I first had the idea of Foca when a good friend of mine and myself are doing accountability study sessions on Discord. We both have great ambitions but find it hard to focus with the constant distractions around us and also get bored of the work we do (especially uni work). Wanting to create a startup, I thought why can’t we make a more specific platform to do what we are doing on Discord and make it better? Foca was born. Other inspirations and motivations include, the movie “The Social Network”, app: Strava, Github, and BeReal.

What it does

Foca is a web platform that allows users to track their productivity when completing online tasks. Foca is unique in that you don’t have to report your activity. This AI-enabled product analyses snapshots of the user’s screen to determine whether the apps they are using in their productive session are relevant to their task at hand. Furthermore, there is a social aspect where users can share their completed tasks and boast their productivity to other users in their circle.

How we built it

We split the workload into three parts. King worked on the main page of the website and AI, Cal worked on the timer and most of the frontend and Sanket structured the database and created the visualisations. The Frontend was created using React.js and Chakra UI. Visualisations and AI interaction were done in Python and SQLite was used as a database management system.

Challenges we ran into

We have a database with over 500 images which is our own proprietary dataset and training our own model on such a small dataset would not be possible, as we wanted our AI to have a high prediction accuracy. Finding an AI model that has already been trained on a lot of data to do the task for us was our biggest challenge. The frontend and backend of the app were not integrated due to an issue with Flask that we encountered last minute. Due to time constraints we were unable to get this working. This challenge could easily be solved with more time for troubleshooting.

Accomplishments that we're proud of

Coming up With a product that can be used to solve a real world problem. Creating ‘good’ functional components with limited time. Picking up new skills while building this product to add features and build something meaningful. Most importantly collaborating together as a team and being highly productive; building Foca in a three day time span is something we are really proud of.

What we learned

Produce an iteration under pressure. Importance of soft skills in a professional setting. The effort that goes into starting a startup. On the technical side of things we learnt a lot of new ways to code, managing databases and most importantly using github features.

What's next for Foca

We want to train our own AI model using our own proprietary dataset and have Foca be used in schools and other institutes to create a competitive atmosphere for students and promote learning. In further iterations, Foca can be a fully integrated product that gives users their productivity data in real time and nudges them back on track if their attention deters.