What it does
Effectively simple, text - based user interaction system that is activated when the user texts our AI program when in need. The program then asks the user a few questions in order to perfectly pair it with the nearest shelter using the latest, state of the art, geosynchronous technology. Our algorithm takes into account the gender, sexuality, and parental status of the user to redirect them to not only the nearest shelter, but also one that best suits/specializes in their needs and that in real - time updates the availability of beds for that night so a user isn't directed to a full shelter that will have no choice but to turn them away. We purposefully chose to use this text based system (as opposed to creating an app) in order to provide cross platform support for all cellphones regardless of model. This way everyone in need is reached and we do not discriminate against users that can't run certain apps, lack data, or web application software/accessibility.
How we built it
-Decided LAMP stack
-Chose to work with Google Maps API and Twilio for external services
-Split team up based on strengths and assigned work packages accordingly
-Note: Our project uses mock data and live data. Live data can and will eventually replace all mock data.
Challenges we ran into
-Managing our stack (keeping things clean)
-Brainstorming the most efficient yet effective ways to solve the problem of homelessness and how best to help shelters/organizations in this same quest for the betterment of humanity.
-Knowing which features to prioritize
Accomplishments that we're proud of
-Being able to offer everyone this service (all platform compatibility)
-Ensuring that a user is directed to a place with actual available resources, thus effectively a system that guarantees zero rejection
-Gearing results to specific user needs/situations
-Our integration of Twilio
-Taking unfamiliar APIs and adapt/work with them despite lack of previous experience. Our adaptability.
What we learned
-Importance of having a very strong design document at the start. Lack of specifics slowed us down.
-Syntax is not our friend.
What's next for Project: End Homelessness
-Cleaner databases
-More statistical/predictive analysis
-Expansion of our system to alert users via text of nearby soup/food kitchens, holiday gift/food giveaways for kids/families throughout the year, and job opening opportunities.
Log in or sign up for Devpost to join the conversation.