College life is supposed to be about more than lectures, assignments, and exams. It's about meeting people, exploring interests, discovering new hobbies, and finding communities where you belong.
But here's the problem: not every student knows which events to attend. College announcements are scattered across different platforms, and students often miss opportunities simply because they don't know what's happening or which events would genuinely interest them.
I built Collexa to make that experience more personal.
Imagine a friend who wants to participate in college life but doesn't know where to start. They might enjoy design, music, technology, gaming, or creative activities, yet struggle to find events that match their interests.
Instead of making them scroll through endless announcements, Collexa aims to help them discover events that fit their vibe.
The experience follows a simple journey:
Discover: Explore a dedicated platform for college events.
Choose Your Vibe: Express your interests and preferences.
Personalize: Connect those preferences with a more relevant event discovery experience.
Explore: Find opportunities to participate, meet like-minded people, and make college life more memorable.
My goal with Collexa is simple: make discovering college experiences feel less overwhelming and more exciting.
I want students to spend less time searching for opportunities and more time actually experiencing them.
Demo
Live Demo: [Add your deployed Collexa link here]
Source Code: [Add your Collexa GitHub repository link here]
Screenshots / Demo Video: [Add a short walkthrough showing the landing page, Choose Your Vibe flow, authentication, and event discovery experience.]
How I Built It
Collexa is built with a modern frontend stack, with a strong focus on user experience, personalization, and visual storytelling.
Technology stack:
React + Vite: For building a responsive and component-based frontend.
GSAP: For smooth, purposeful animations and transitions.
Supabase: For backend services, authentication, and persistent user data.
Google OAuth and email/password authentication: To give users convenient sign-in options.
AI-powered personalization: A direction for making event discovery more relevant to individual interests.
I designed the onboarding experience around a simple principle: personalization should begin with the user, not with a long and complicated setup process.
The current journey moves from the landing page to Choose Your Vibe and then authentication, with the aim of creating a personalized event discovery experience.
I also wanted the interface to feel like a thoughtfully designed product rather than a typical college event listing website. The visual design uses a warm, light foundation, a darker hero section, carefully arranged event cards, and restrained animation.
Note: The current stack and implemented features are listed above. An open-weight model, local inference, or an AI recommendation pipeline should only be described as implemented if it is actually connected to Collexa.
Why Does Open Innovation Matter?
I believe personalization should not have to mean giving up control over your data or depending entirely on a single AI provider.
For a platform like Collexa, open innovation could make a meaningful difference.
An open-weight model running locally could help match student interests with event descriptions, explain why an event might be relevant, and generate personalized suggestions without requiring every interaction to be sent to an external AI service.
It could also give developers the freedom to experiment with different models, adapt recommendation behaviour, and work within limited budgets. That matters when building something intended to be accessible to students and college communities.
Open approaches can also make experimentation more accessible to independent developers like me. They create opportunities to learn, customize, and build beyond the limitations of a fixed, closed API.
For Collexa, the long-term vision is to combine personalized event discovery with AI that can be inspected, adapted, and run in ways that respect users' needs.
I want to be clear that this is the direction I would like to take Collexa, rather than claim that local AI inference is already implemented.
Built with the belief that college life gets better when people find the experiences and communities that feel right for them
Top comments (0)