Inspiration
Before starting my first semester at Mahindra University, I wanted to explore the clubs and communities on campus. I expected to find everything in one place, but the information was scattered across the university website, social media, and old event pages.
I spent more time searching for information than actually learning about the clubs.
So before I even stepped onto campus, I built Sortling.
What it does
Sortling is a Discord bot that helps students discover clubs and communities at Mahindra University.
Students can browse clubs, search using different names or abbreviations, and get recommendations based on their interests through the /sort quiz. Instead of jumping between multiple websites, they can quickly figure out which clubs might be a good fit and where to find them.
Behind the scenes, Sortling maintains a verified club directory built from official university pages and public club sources.
The Mahindra University '26 Discord community is already using Sortling to explore campus life before classes begin. Since I'm an incoming fresher too, it's been fun seeing other students use something I originally built for myself.
How I built it
I built Sortling with Python, Nextcord, SQLAlchemy, and Render.
The recommendation engine compares a student's interests with club profiles and decides which question to ask next based on previous answers, so students usually get recommendations after just a few questions instead of filling out a long form.
A lot of the work went into the data. I built a pipeline that collects information from university pages, processes it into a structured format, detects duplicate club names and aliases, and prepares everything for review before it's added to the directory.
Challenges I ran into
The hardest part wasn't building the bot. It was building the data behind it.
There wasn't a single place that had complete and up-to-date information about every club. Some clubs only had social media pages, some appeared under different names, and some information was simply outdated. Verifying everything took much longer than I expected.
Getting the recommendations to feel useful without asking students too many questions also took a lot of experimenting.
Accomplishments that I'm proud of
I'm happy that Sortling solves the exact problem that made me build it.
It now has a verified club directory, supports aliases and fuzzy search, recommends clubs based on student interests, and is already being used by students in the Mahindra University '26 Discord community before we've even started college.
What I learned
Going into this project, I thought most of the work would be writing code.
It turned out that building reliable data was just as important. Spending time verifying information, organizing it properly, and deciding what not to include made a much bigger difference than I expected.
What's next for Sortling
Right now, Sortling is built for Mahindra University.
The next step is adding campus events, hackathons, workshops, internships, research opportunities, and other things students usually end up searching for during the semester.
After that, I'd like to bring Sortling to other universities across India. Every campus has the same problem in different forms, with information spread across websites, social media, and student communities.
If that goes well, I'd love to eventually support universities outside India too. I hope future students can start college with Sortling already waiting for them.
Built With
- beautifulsoup4
- discord-api
- git
- github
- machine-learning
- nextcord
- python
- render
- scikit-learn
- sqlalchemy
- sqlite
Log in or sign up for Devpost to join the conversation.