Toronto // Sysop // BBS Era // Founded 2002
Founded in 2002, back when the network still felt like something you could build, run, and understand yourself. A SysOp mindset, carried into the AI era — keep the personal layer yours.
01 // Origin

The early internet taught a simple idea: the machine in front of you could be your own. Run the server, keep the files, know the system, decide who gets access.
The technology changed. The question didn’t: what parts of your digital life should actually belong to you?
CETI is how we build it: a local, sovereign model that uses data — self-improving, learning. Data that’s signed, so you can verify what it learned from and that it hasn’t been tampered with. Memory, identity, and permissions stay on the machine you control. Models can change. Your personal layer doesn’t have to.
02 // Your node
A local-first personal system, built as CETI. Run the core on your machine. CETI learns from your data, improves itself over time, and keeps what it learns in signed data you can verify. Memory stays portable, inspectable, and under your control.

03 // Services
Darknet Consulting LLC. We work on complex problems and out-of-the-box ideas — the kind that don’t have a playbook yet, and where the obvious answer isn’t the right one.
The tangled ones. The ones that cross systems, teams, and assumptions. We untangle them — patiently, and all the way through.
Not louder versions of the same idea. Different ideas. The kind that only show up when you step outside the usual frame.
Ideas are only half of it. Work that still makes sense when the moment, the market, or the model changes.
04 // CETI — Proof of work
CETI is a local, sovereign model that uses data — self-improving, learning. It gets better from what it works on, and the data it learns from is signed, so it can be verified. No landlord. No rented memory. It runs on your iron.
Runs locally, on your machine. Your data stays yours.
Learns from use. What works gets kept. What fails becomes something to learn from.
Data that’s signed, so you can verify what it came from and that it hasn’t been changed.
05 // Signal
Your AI shouldn’t become a rented archive of your life. Memory should be portable, inspectable, and under your control.
Use a local model for routine work, or route difficult tasks to the model or provider that makes sense.
Hardware can evolve. The processor can change. The node remains where your identity and memory live.

06 // Evidence
Toronto / After dark — local models, old machines, new ideas, and a view of home.