Every time we go to hackathons, meetups or other events we get a rare opportunity to spend a few hours with hundreds, often thousands, of interesting people. Yet in those few precious hours we are lucky if we get to talk to one or two likeminded individuals or people knowledgable about topics we want to explore next.

PeopleHunt gathers publicly available information about all attendees from a list you provide and shows you those who share your interests, needs, and network of friends or colleagues. Registration is not required, unless you want to pull in your LinkedIn or GitHub contacts to discover people you may already know who are attending the event. The more attendees use People Hunt, the better the data and matching gets, but unlike most registration-driven systems, PeopleHunt us instantly useful to even the first user at any event. We use machine learning to calculate close correlations between topics and people and have a perpetually increasing data set for future matches.

Sifting through thousands of online profiles is impossibly time consuming, but relationships created at the right place and time can be incredibly valuable. Long-term we can charge heavy users a subscription fee, or a use fee per number of fetched requests and monetize free searches with other relevant recommendations.

TEAM: Diana Zink (DoerHub, NBA, CNN); Jing Wang (Google Ads Machine Learning); Manny Karyampudi (Fullstack Mobile Software Developer); Yingpeng Xu (Backend Performance Software Engineer);

Share this project: