Inspiration
We’ve both experienced how difficult and inefficient networking can be. Whether it’s for hackathons, startup ideas, or finding jobs, reaching out often feels random, impersonal, and time consuming. Cold messages rarely get responses, and even when they do, it can take days or weeks of back and forth to figure out if there’s a real fit. Most connections never turn into anything meaningful.
We wanted to fix that by making networking feel more intentional, structured, and fast, so people can focus less on searching and more on actually building together.
What it does
Our platform helps people find the right teammates, employers, and builders based on structured profiles and compatibility-based matching.
Users create a profile that includes their skills, interests, what they want to build, and what they’re looking for in others; whether that’s teammates, mentorship, feedback, or co-founders.
They are then shown curated profiles with a match score and a quick summary of why they are a potential fit. If both users are interested, they instantly connect and can start talking in a simple messaging space. This removes the slow, awkward back and forth of traditional networking and replaces it with fast, mutual connections.
How we built it
Our project was built with a Django backend with react and Tailwind CSS for the frontend. The matching system is based on a vector embedding architecture using Cohere's embedding v3 model. Using an asymmetric scoring scheme based on the various different aspects of each user, an aggregated score was calculated using a weighted sum. This sum then determined the percent match between two separate users.
Challenges we ran into
One major challenge was deciding how to structure matching in a way that feels meaningful rather than random. We had to think carefully about what signals actually matter in collaboration like skills, intent, goals, and complementary needs.
Another challenge was balancing simplicity with depth. We wanted onboarding to stay quick, but still collect enough information to make accurate and useful matches.
Accomplishments that we're proud of
We’re proud that we designed a system that goes beyond simple interest matching and instead focuses on real collaboration potential.
We built something we would actually use ourselves to find teammates and collaborators, which was a big goal from the start this wasn’t just a concept, but something rooted in our own real frustrations with networking.
The platform also introduces structured onboarding, match-based discovery, and a mutual acceptance flow, which together create a more intentional and efficient way for people to connect.
What we learned
We learned that matching people effectively depends heavily on understanding multiple factors like skills, goals, and intent, and how they interact together.
We also learned how small decisions in how information is collected and organized can have a big impact on the quality of connections and the overall user experience.
Log in or sign up for Devpost to join the conversation.