Your AI agents need their own server
Running AI coding agents on your laptop works fine for one-off tasks. But the moment you want agents running overnight, processing a backlog of issues, or executing multi-step workflows while you sleep, your laptop becomes the bottleneck. Close the lid and everything stops.
You could spin up a cloud server and configure it yourself. SSH hardening, firewall rules, Node.js, Docker, swap space, security updates. That’s a full afternoon before you even install your first agent. And you’ll forget something. Most teams do.
We kept doing this setup for every new client engagement and every new project server. Same steps, same mistakes, same afternoon gone. I built a tool to stop repeating myself.
What Dodo VPS does
One command provisions a hardened Hetzner server with your choice of AI coding agents pre-installed. Interactive wizard, about 10 minutes from start to SSH.
Also works on existing Ubuntu servers with the --on-server flag.
What’s included
Agents
- Claude Code
- Codex
- Gemini CLI
- OpenCode
Runtime
- Node.js 22
- Homebrew
- Bun
- Docker (optional)
Security
- SSH hardening
- UFW firewall
- Fail2ban
- Auto security updates
- Swap
Networking
- Optional Tailscale VPN for private access
How I use it
Bug triage on autopilot
A Linear issue gets created, an agent on the VPS reproduces the bug, opens a PR with a fix, and tags the right reviewer. I wake up to a PR, not a bug report.
Code review before humans touch it
Push to a branch and agents run security scans, pattern checks, and architectural review. By the time a developer opens the PR, the easy stuff is already caught.
Long-running knowledge work
Agents process incoming emails, update CRMs, generate weekly reports, and flag anything that needs a human decision. These processes run for hours. They can’t live on a laptop.
The VPS is the foundation. The value is what runs on it.
We do half-day and full-day sessions where I set up the VPS, configure your agent infrastructure, and build out the hooks, skills, and configs that sit on top. By end of day, you have agents running autonomously on your own server.
$2,500 to $10,000 depending on team size and scope. Most teams land around $5K.
Book a sessionGetting started
On an existing Ubuntu server:
Server pricing
Hetzner Cloud. Billed directly to you. No markup.
| Tier | Specs | Price |
|---|---|---|
| Small | 2 GB / 2 CPU | ~$6/mo |
| Medium | 4 GB / 3 CPU | ~$11/mo |
| Large | 8 GB / 4 CPU | ~$19/mo |
| XL | 16 GB / 8 CPU | ~$34/mo |
Not sure where to start?
I’ve configured this infrastructure for dev teams across SaaS companies, agencies, and startups. If your engineers are using AI tools but the output is inconsistent or the setup feels fragile, that’s exactly what these sessions fix. Happy to jump on a 20-minute call first. No pitch.
Schedule a call