"The story behind the university sorting hat."
UniSort was born out of a simple observation: choosing a university is often reduced to grades and location, ignoring the vibrant, distinct cultures that define the "Big Four" experience.
We wanted to create something that felt less like a form and more like a journey—a digital sorting hat that considers your personality, social preferences, and values.
By wrapping it in a nostalgic newspaper aesthetic, we pay homage to the rich tradition of campus journalism while delivering modern, data-driven insights. It is a labor of love, code, and a lot of caffeine.
The journey began with Research Gathering, where we tapped into the pulse of student life. We posted questions to Freedom Walls and communities across the Big 4 universities (ADMU, DLSU, UP, UST) to collect raw, unfiltered experiences about culture, academics, and social life. These insights were compiled into comprehensive research documents that served as the foundation for everything that followed.
For Quiz Development, we fed these research PDFs to AI to identify distinct university characteristics. We generated initial questions based on these patterns and refined them multiple times to strip away generic phrasing. Each question was carefully mapped to personality traits that align with specific campus cultures—like QPI pressure tolerance for ADMU or the fast-paced trimestral life of DLSU.
We then built a robust Scoring System that moved beyond simple binary results. We developed a percentage-based tier system (from 0% to 100%) to represent nuance. To bring these scores to life, we studied the engaging writing style of 16personalities, iterating heavily to find a direct, personal voice. We crafted detailed descriptions for every compatibility tier, ensuring that whether you matched 5% or 95%, the result felt accurate to the ground-truth research.
Finally, during Refinement, we meticulously removed AI-sounding phrases and ensured every description was cited back to our initial research. We balanced honesty with appeal, highlighting both the struggles and triumphs of each university. The result is a system built on a key principle: Raw student voices → AI analysis → Human refinement → Research validation.
This project is open-source and community-driven. We need your help to ensure our "sorting hat" remains accurate and our campus lore stays up-to-date.
Fork the repo on GitHub
Contribute Now