Nature, 2016
Help make sure no mind is lost.
We are a small founding team building the memory layer of the human brain. The work is early and hard. It matters.
See open rolesIn the research
Built on decades of open memory research
Science, 2013
Creating a false memory in the hippocampus
Read paperJournal of Neural Engineering, 2018
Developing a hippocampal neural prosthetic to facilitate human memory encoding and recall
Read paperWhy
Our Mission
Make sure no human mind is lost: to disease, to injury, to the repair of the brain itself, or to time.
A person is, in a very literal sense, their memory: their relationships, skills, language and sense of who they are. Yet memory is the one part of us with no backup and no repair. Dementia, brain injury, stroke and aging erase it, and no approved therapy restores what is lost.
Chiron Discovery is building the memory layer of the human brain, and in time a cloud for human memory. Today that means models that read, verify and help restore memory from human neural recordings. One day it means preserving it.
Our principles
How we work
A technology that reads memory is among the most sensitive ever built. These principles shape what we build and what we refuse to build.
The person comes first
Every model we train is about someone's memory. Memory belongs to the person.
Evidence over hype
We claim what we have shown, nothing more. Verification comes before claims.
Own the whole problem
Small team, wide scope, real responsibility.
Honest about stage
We are early. You will help decide what this company becomes.
Restore, never implant
We help bring back what a person already remembers. We never write false memories.
Open science where it does not harm people
Methods and validation are published. Individual data never is.
Day to day
A small team, close to the science
In person in San Francisco
We work together in person, with flexibility for exceptional people. Hard problems move faster at a whiteboard than in a thread.
Methods in the open
We publish how we work and we cite what we build on. Our analyses are meant to be checked, rerun and argued with.
Deep work, few meetings
Decoding the brain rewards long, uninterrupted attention. We protect it, and we keep the calendar deliberately quiet.
Scientists and engineers together
Neuroscientists, machine learning engineers and clinicians sit on the same problems, not in separate handoffs.
Open roles
Help us build the memory layer
No open roles right now.
You do not need to be a neuroscientist to work here. You need to care about getting memory right. If you can move this mission forward, tell us how.
Get in touch