Clink!

Inspiration and How it works

Imagine you're at a hackathon and meet someone you're sure you know from somewhere. Maybe it's a mutual friend or an old colleague, but you can't place it. That's where Clink! comes in. Bring your phones together and Clink! shows you the people you both follow or know on Instagram and LinkedIn. It also shows how your paths have crossed: "You both went to UIUC at the same time" or "Both in Champaign 2021–2024."

How we built it

We built it in Swift UI with a server in FastAPI and several auth providers managing connections between your social platforms. We used Meta's AIs and Graph API especially.

Our story

Initially we started by working on a cool piece of tech more than the problem itself. We thought by, using a combination of Meta's Graph API and Serp's search endpoints, we could find where the mutual relationships between people without an explicit "LinkedIn connection". For example, I shared many classes in high school with my friend bob who also attended my high school at the same time as me. What if, I was able to find who bob is and who he also went to high school and college with, weigh that on a graph and see how relevant or close his friends are in relation to my friends.

We realized together that this tech we were working with has a huge application in finding our mutual friends.

Just today, at the hackathon, I met someone who I thought I knew and we had to open Instagram and find each other and check our mutuals, then check LinkedIn after that to see if we knew anyone in common. This should be a solved problem!

The idea behind our app is that you can simple hold your phone next to someone else with the app open and you'll instantly connect / follow with them on all social platforms and find each other's mutuals, including that which aren't easily findable - beyond just your connections and followings. This is only possible with the use of agents to web scrape and find mutual people :)