Avent

Here is a project write-up for Avent, following the requested format:

Inspiration

The idea for Avent was inspired by the need to help new and international students quickly discover campus events that match their interests. Navigating a large university environment can feel overwhelming, so we wanted to create a tool that recommends events while allowing students to connect with clubs, communities, and friends through shared experiences and icebreaker prompts.[1]

What it does

Avent enables students to select a few events they are interested in upon registration, then uses AI to analyze choices and provide personalized recommendations for future campus activities and club events. It generates a custom interest profile, shares AI-powered icebreaker conversation starters, and can build a daily plan based on recommended events and popular campus locations.[1]

How we built it

We built Avent using React as the frontend framework in VS Code, with tailwind CSS for design and style customization. The AI logic leverages Google Gemini's generative models for analyzing user-selected events and generating recommendations, profiles, and icebreakers. Images are stored in the public directory and displayed with JSX, while event and theme data is managed locally in state using React hooks.[1]

Challenges we ran into

We encountered integration issues with the Gemini API, file extension and JSX errors when initially trying to run code directly with Node.js, and styling/image path problems. Accessing AI results in the correct format and ensuring real-time responsiveness across selection phases required trial and error.[1]

Accomplishments that we're proud of

We're proud of delivering a seamless event discovery and recommendation workflow, providing actionable daily plans and conversation starters uniquely matched to student interests. The real-time, dynamic theme updates and intuitive React interface help create a welcoming campus experience.[1]

What we learned

We learned the importance of using proper React tooling for JSX support, efficient state management with hooks, and robust error handling when consuming generative AI APIs. Using theme-based design helped personalize experiences, and collaborating on a team required clear role division and documentation.[1]

What's next for Avent

The next steps for Avent include integrating campus authentication, expanding AI to more types of campus opportunities, developing mobile-friendly layouts, supporting notifications, and collecting live feedback data to continuously optimize recommendations and student engagement.[1]

Built With

• react