Every one of us has struggled with being the "new kid" at school, or just trying to find a group with similar interests in general. We've all been there: you finally find a friend group, but your passions get talked over, and nobody asks what you bring to the table. Our team has struggled in the past with trying to find a friend group that shared our interests. We know what it is like to feel left out and isolated. With our app, SquadUp, this won't be a problem again. SquadUp allows users to interact directly by voice or by text with an AI, which then uses their preferences based on interest, experience, and commitment level to match them with other users. This allows teams to be created for certain group events like hackathons, study groups, or even just to find a friend group. SquadUp uses a team of Gemini agents to interview the user, summarize their responses, and make matching predictions through the use of an additional machine learning model. Our app also keeps track of the users location to match team members based on their area as well. We even allow the option for a user to create their own squad, or organization (as they are listed on our app). One challenge we faced was trying to figure out what exactly the scope of our problem was going to be. We had a lot of discussion about this because we could not really find a truly unique way to tackle the problem. Another challenge we faced is figuring out which server software to use as a host of our web app. We first tried using the Vultr hosting service, but their server stopped working and went down, so we could not host our web app on it anymore. We then had to run our web app and host it locally on a computer.

We used Grok to create an auto merge GitHub agent. Ran 3 Cursor swarms to build UI, test, and integrate AI features into our website. This cut down our development time by 80%.