AI agents for your team.
Managed.

Deploy a private Centaur instance in 60 seconds. Your tools, your credentials, your Slack.

Deploy

60 seconds to running

Pick a tier, choose a region, click create. The provisioner handles k3s, Helm, TLS, and NetworkPolicy automatically.

Isolate

Your infrastructure

Each deployment runs in its own Kubernetes namespace with dedicated Postgres, iron-proxy credential injection, and network isolation.

Save

7× cheaper than AWS

Runs on Hetzner Cloud. A dev instance starts at $9/mo. The equivalent on EKS costs $560/mo.

How it works

1

Create a deployment

Name it, pick dev / small / production, choose a region. The dashboard creates the record and the provisioner takes over.

2

Add your credentials

OpenAI key, Slack bot token, signing secret. The provisioner pushes them to K8s automatically — you never touch kubectl.

3

Talk to your agent

Mention the bot in Slack. It runs in a sandboxed environment with your tools, your repos, and your credentials — isolated from every other tenant.

Pricing

No per-seat fees. No usage surprises. Infrastructure cost only.

Dev
$9/mo
Single node. Good for testing and low-traffic use. 2 vCPU, 4 GB RAM.
Small Production
$82/mo
3-node HA with dedicated Postgres. Survives single-node failure. 12 vCPU, 24 GB total.
Production
$136/mo
3-node HA with larger nodes and burst capacity. 24 vCPU, 48 GB total.

Ready to deploy?

Your first deployment is free for 7 days.

Built on Centaur by Paradigm

trycentaur.dev