Deployment
How a laboratory goes from nothing to a running Archon instance, what your infrastructure needs to provide, and what to do if your IT team wants to run the deployment itself.
How deployment works today
Archon isn’t a hosted product — there is no shared Archon cloud to sign up for. Every laboratory gets its own instance, running on infrastructure your organization controls: a server you already have, a VPS you provision, or hardware physically on-site. Quorin Technologies deploys and configures that instance for you, at no cost. What you pay for, if anything, is the software licence tier itself — see Pricing.
Because every deployment is separate, there is no central Archon login — your laboratory gets its own URL and its own accounts, entirely independent of any other laboratory running Archon. See this is a self-hosted deployment and the FAQ for what that means for where your data lives.
What your infrastructure needs to provide
These are pulled directly from what Archon actually requires to run, not rules of thumb:
| Requirement | What Archon needs |
|---|---|
| Server | Somewhere Node.js 20+ and PostgreSQL can run — Linux (the usual choice for a VPS) or Windows (for an on-premises install directly on lab hardware). Archon is a standard Node.js application; it has no platform-specific dependency. |
| Database | PostgreSQL 17. Archon is one Node.js process talking to one PostgreSQL database — no Redis, no message broker, nothing else to run alongside it. |
| Storage | Persistent disk for the PostgreSQL data directory and for uploaded attachments. Both need to survive a restart or redeploy — an ephemeral container filesystem will silently lose attachments while the database rows still point at them. |
| Inbound network | One HTTP(S) port reachable by your lab’s staff. Nothing else needs to be reachable from outside the server. |
| Outbound network | None, by default — Archon makes no outbound call unless you turn one on: SMTP for invitation and password-reset emails, or the optional AI assistant, which is off by default and stays off unless you explicitly enable it. |
There’s no published minimum CPU or RAM. A single Node process and a single Postgres database is a modest footprint, and the right size depends on how much data you store and how many people work in it at once — we size this with you as part of the deployment conversation below, rather than publishing one number that’s wrong for most labs.
From request to running instance
- Get in touch. Email quorinstartup@proton.me with a short description of your situation — a VPS you already have, hardware you want it running on, or “we don’t have anything yet, what do you recommend.”
- We confirm the details. Server access (or hosting details, if we’re provisioning it), the domain or address your staff will use, and the email address for the first owner account.
- We deploy and configure it. The database, environment configuration and initial admin account are set up on the agreed infrastructure.
- You get a URL and sign in. Your laboratory’s Archon instance is live at its own address, with an owner account ready to invite the rest of your team — see first steps in a new organization.
- A licence key is applied, if you’re on a paid tier. A signed licence key is installed once, from Settings → Billing.
How long this takes mostly depends on how quickly infrastructure details come together on your side — the deployment step itself is fast once that’s settled. Ask when you get in touch if you need a firm date to plan around.
Want your own team to deploy it?
If your laboratory’s IT team wants to run the deployment themselves rather than have Quorin do it, that’s supported. Request access to the Archon repository at quorinstartup@proton.me and mention you want to self-deploy. You get the same application every Quorin-managed instance runs.
Self-install instructions
Coming soon. The Archon repository is currently private, so step-by-step self-install instructions — prerequisites, environment configuration, running database migrations, first-run admin setup, applying a licence key, and backups — aren’t published here yet. They’ll appear on this page once the repository is public.
Until then, an IT team with repository access (see above) gets the equivalent instructions directly rather than waiting on this page.