I built a collaborator discovery network because cold-emailing strangers in 2026 is still somehow the best option — Synapse

Nobody tells you that finding a collaborator in academia is a part-time job.
You start with cold emails that disappear into the void. Then you upgrade to LinkedIn Premium — because apparently paying to message strangers is just the cost of doing science now. Then you spend hours crawling through institution websites one by one, trying to figure out who's working on what, whether their research aligns with yours, whether the page is even current — half of them aren't, some are in languages you don't speak, a few look like they haven't been updated since 2009.

All of that, before a single conversation happens.

And that's if you already know how to play the game. For a researcher fresh out of university, the whole thing is a wall. Nobody taught you the unwritten rules. You don't know how to reach out without it feeling awkward. So you don't. And your entire research network stays exactly as large as your advisor's rolodex.
But strip away the lab coats and it's the same story everywhere. The indie developer who needs a designer. The engineer who needs a domain expert. The consultant who needs a technical co-founder. Everyone is one great collaborator away from their best work — and the tools we have weren't built for finding that person.

I watched this happen enough times that I stopped complaining and built something instead.

What I built:
Synapse (synapsebase.ca) — pure collaborator discovery. You describe your project and what skills you're missing. You find people doing complementary work that your existing network would never surface. For new researchers and builders, joining a project is the ice-breaker — you contribute first, build confidence, and grow from there.
Stack: React + Supabase + Cloudflare Pages.
Where it's at: Live, real users, projects on the board.

Two things I'm genuinely wrestling with and would love honesty on:

1. Positioning — when you read "find collaborators beyond your network," who do you immediately picture using this? I want to know if my mental image of the user matches yours.
2. The cold start problem — every discovery platform lives or dies on having enough people on both sides. What's the one thing you've seen actually work for cracking that in the early days — not the theory, the thing that actually moved the needle?

Happy to answer anything about the build, the problem, or what it's like solo building a product around a full-time research job.