You found one strong R&D candidate and they’re not moving. Now what? In academia, the people someone publishes with are the closest read you’ll get on adjacent talent — same lab, same methods, often the same building. This tool turns one name into a ranked shortlist of those collaborators, so you stop sourcing one person at a time.
Type a researcher’s name — say Andrej Karpathy — and hit Find co-authors. We pull their last 50 papers from Arxiv, count everyone they’ve published with, and rank the top 15 by how many papers they share. Each row shows the collaborator and their paper count (12 papers), and clicks through to that person’s own Arxiv profile, so you can repeat the move and walk the network outward. Frequent collaborators bubble to the top because co-authorship is a real signal: senior PIs publish with their lab’s students, junior researchers publish with their advisors.
One honest limit: this only sees Arxiv. Fields that don’t preprint heavily — much of biology, chemistry, anything behind closed publishers — will look thin or empty here. And a name like “J. Smith” or a researcher who shares a name with someone else will give you noise; the more distinctive the name, the cleaner the result. Treat the ranking as a lead list, not a verdict. The drill-down link is where the actual sourcing starts.