Inspiring Hack Projects: Spill.it
May 3, 2017 / Developers, Hackathon Spotlight, Text Data Use Case
A few weeks ago we had the pleasure of supporting HackBeanpot, for the second year in a row, at Genuine HQ. We caught up with the winners of indico’s Machine Learning API prize — Alex, Noah, Brendan, and Jon, the creators of Spill.it.
What is Spill.it?
Spill.it is a party game that encourages your friends to “spill the tea.” Users can join on their phones to answer questions on celebrities, emotions, life events, and more. Using the indico API, we analyze each answer to determine which emotion, personality, or persona best represents it. Using this data, the game encourages friends to guess whose response matches each different emotion, personality, or persona. Correct answers earn points, and the player with the most points wins! – Brendan
What inspired you to build Spill.it?
We had recently played the Jackbox TV games as a group, and really enjoyed the social aspect of their game structure. When we saw the indico.io API presentation, we were immediately interested, and knew we wanted to integrate it into our project in some way. After some brainstorming, we realized we could use indico’s sentiment analysis as our own twist to the Jackbox games, and create our own social party game. – Jon
How did you build it?
We use indico as our back-end analysis platform. We created a few methods that use the API endpoints when our users submit questions to the game, and these methods are then triggered to grab these data points from the indico platform. Our back-end then takes the persona/personality/emotion that has the highest percentage and assigns that value to the submitted answer. This is stored in our database and then used in the game to prompt users to correctly match the characteristic to players in the room for points. – Brendan
How did Spill.it evolve throughout the hackathon?
We started off with a big idea, listing out all of the features and bells and whistles, before we started narrowing our scope as we went along. Some of those decisions came from technical limitations while others came from an agile background — we wanted to make sure we had a presentable MVP by the end of weekend. – Alex
What are your plans for its future?
In order to have a semi-finished product by the end of the hackathon, we focused mainly on getting the back-end of our application together. Therefore, in the future we’d like to build on our existing foundation by branding our application and putting together a more dynamic front-end. Beyond that, our stretch goals include creating a greater variety of question types using features such as indico’s Image API. – Noah