Outsy

Inspiration

We’ve all had moments where a group wants to hang out, but everyone wants something different. One person cares about price, another cares about distance, and someone else wants a specific type of activity. We wanted to build an app that removes the back-and-forth and helps a group quickly find something that works for everyone. Our app is called OUTSY.

What it does

OUTSY helps groups discover and plan outings based on their location, budget, interests, availability, and transportation preferences. Instead of giving everyone the same generic recommendation, it considers each person’s preferences and ranks destinations based on how well they match the group.

How we built it

We built the app with Next.js and TypeScript, using Supabase for the backend, database, authentication, and real-time features. We use Google Maps/Places for location, destination, and transportation data, while Gemini helps explain recommendations and generate personalized outing plans and itineraries. Our recommendation engine runs on the backend and scores destinations based on factors such as distance, travel time, cost, shared interests, event availability, and transportation. Gemini then turns those results into understandable recommendations and plans.

Challenges we ran into

Our biggest challenge was the scope. We wanted to combine real-time location, live APIs, maps, group recommendations, AI, voting, chat, and cost splitting into one product, while none of us had previous hackathon or significant technical experience. We had to decide what was truly essential for the MVP and prioritize reliability over building every possible feature. We focused on making the group-based recommendation work first, with the map and AI-generated itinerary as the main supporting features.

Accomplishments that we're proud of

We’re most proud of turning a common social problem — “What should we do?” — into a technical recommendation problem that can actually be solved. Our app combines individual preferences into a group-level recommendation instead of simply showing a list of popular places. We’re also proud of combining a traditional scoring system with Gemini: our code determines which options are the best match, while AI explains the reasoning and creates the final outing plan.

What we learned

We learned that building a useful product is not just about adding more features. It’s about identifying the one problem that matters most and making that experience reliable. We also learned how different technologies can work together: APIs provide real-world data, our backend handles the recommendation logic, and AI makes the results easier to understand and more personalized. Most importantly, as first-time hackathon participants, we learned how to quickly learn unfamiliar technologies and turn an idea into a working product.

What's next for OUTSY

Our current MVP is limited by API reliability, a simple recommendation system, and basic safety information. We would improve it by using more reliable real-time data and making recommendations more personalized based on users’ past choices and feedback. We also plan to improve real-time location sharing, group chat, voting, and payment splitting. Ultimately, we want to make the entire process of planning, enjoying, and paying for an outing more seamless.