NextOccasion

Inspiration

As high schoolers balancing heavy course loads and extracurriculars, finding the right summer programs, hackathons, or internships is incredibly overwhelming. It often feels like the best opportunities are hidden behind hours of endless, disorganized web searches. We realized that it was difficult to find valuable extracurriculars that actually fit our specific skills, interests, and schedules. We wanted to create a tool that cuts through the noise and delivers perfectly tailored, high-value options directly to students.

What it does

NextOccasion is a personalized, AI-driven opportunity scout. Users create a profile detailing their background, grade level, skills, and interests. Using this context, NextOccasion generates a dynamic carousel of highly relevant extracurriculars—ranging from specialized internships to local volunteer roles. Users can seamlessly click through tailored recommendations and save their favorites into a persistent, personalized "Library" so they never lose track of an application deadline.

How we built it

We built a clean frontend utilizing HTML, CSS, and JavaScript. To handle the backend logic, we integrated Firebase Realtime Database to manage user authentication and securely store user profiles and their saved event libraries. The core recommendation engine is powered by the Gemini API. We utilized structured prompt engineering to pass the user's profile data to Gemini, which returns a cleanly formatted JSON array of customized events that our JavaScript dynamically renders into the UI.

Challenges we ran into

One of the biggest challenges was moving beyond a static concept to a fully interactive app within a tight hackathon timeframe. Specifically, ensuring the Gemini API returned consistently formatted JSON data that our frontend could safely parse required careful prompt tuning and error handling. Additionally, managing asynchronous data flows—like syncing a user's local screen with their Firebase database to handle real-time "saves" and "deletes" in the Library—was a complex hurdle we had to overcome.

Accomplishments that we're proud of

We are incredibly proud to have built a fully functioning, full-stack application from scratch. Going from a blank canvas to a polished product with user profiles, database management, and an AI integration in a single weekend is a massive win for our team. Most importantly, we built a practical, highly customized tool that genuinely solves a major pain point we face in our own lives as students preparing for our futures.

What we learned

We significantly deepened our understanding of asynchronous JavaScript, handling external API requests, and managing state between a web frontend and a backend database like Firebase. We also learned how to effectively engineer prompts for a Large Language Model so it acts as a reliable, structured data generator rather than just a conversational chatbot.

What's next for NextOccasion

We plan to improve our recommendation algorithm by adding advanced, user-controlled filters (e.g., toggling between strictly online vs. in-person events). We also want to implement direct application links and a Google Calendar integration so users can automatically sync application deadlines directly from their NextOccasion Library.