growing brain systems that learn
we're a biocomputing company growing large-scale neuronal tissue, living hardware with memory baked in at the cellular level.
backed by Y Combinator · S26latest
early rollouts: our neurones learning to play Frogger
our pursuit
we're growing scalable brain systems — the next-gen hardware layer for biocomputing.
biological neural tissue is already being trained — on games, on next-token prediction — but wetware has been stuck at the ~1M neurone scale due to the vascularisation problem: past a geometric limit, cells sit too far from an oxygen supply and die. we've developed a tissue-culture approach that breaks that limit.
that lets us grow far larger biological neural nets beyond model sizes that would be traditionally reasonable for GPU-based training — with continual learning baked into the substrate, the sample efficiency of living neurones, and a training platform that sits an abstraction layer above backprop. we provide the substrate, interfacing, and cloud infrastructure for third parties to train on our systems.
careers
we're a small team based in Cambridge, UK, and we're hiring exceptional engineers and researchers across neurobiology, electrophysiology, electrical engineering, and reinforcement learning.
if our work sounds interesting to you, write to us at [email protected] with relevant information on your background.