About
Shared Computer Network is a collective that pools member resources to fund, build and operate locally-run AI models, hosted on hardware installed next to the communities using it.
It is as much a shared learning environment as it is a service. Members contribute to a common infrastructure and, in return, get hands-on access to local AI systems and a growing body of shared knowledge about running them.
From there we are exploring applications that are approachable and useful for any community, such as group-based shared knowledge bases, where what individuals gather is pooled so everyone benefits from the collective experience of the community.
What it looks like
A small cluster of consumer-grade PCs sits behind a secure gateway. Members hold keys to a number of models configured on it, and reach them two ways: a chat window in the browser, and an API key for everything else.
The API is the same one commercial providers speak, so most tools need only a base URL and a key pointed somewhere new. This is called the "Bring Your Own Key" (BYOK) model that can plug community scale AI models into any application: a chat app on your phone, a coding assistant, an agent you are building.
Where your data goes
Nowhere. Every request is served by machines in the room; your data does not leave Canada, or even Vancouver. That is the point of running it here rather than renting it.
Shared discovery
The collective is organised around finding things out together. The expectation is that members experiment, document what they learn, and make their results openly available to one another.
One of the central purposes of the project is to advance every member's technical understanding of local AI infrastructure faster than any of them could alone. The code is public for the same reason: the whole stack is on GitHub and/or Tangled.
Getting involved
We are in an early prototype phase with a select group of early access contributors. If you are an individual or an aligned organisation, get in touch at hello@sharedcomputer.network.
If you run a coworking space, an educational institution, or similar community space, especially in Canada, and want a setup like this one, we would like to hear from you. Working out how to make it repeatable is part of the project.
Read on: the system that runs it, and the team that builds it.