CreateOS Sandbox

Self-Host the Whole Thing

Yes. You can self-host CreateOS Sandbox in your own cloud: run the control plane and bring your own object storage (S3, R2, Tigris, or MinIO), so code, snapshots, and data never leave your boundary. Regulated teams run it this way when code cannot leave their network.

The Blocker is Rarely the Isolation Model

For a regulated, security-first team, the blocker is usually the sentence: our security team will not let us send code to someone else's cloud.

Self-host answers that directly. Run the control plane and the storage yourself, and the question stops being about trust in a vendor's cloud.

How It Works

Self-host the control plane in your own environment. Storage is bring-your-own-S3, so S3, R2, Tigris, or MinIO, with FUSE mounts, live attach and detach, and local to remote sync. Point the control plane at your own buckets and the code, the snapshots, and the data stay inside your boundary.

That is real data residency, set by architecture, not a residency label on someone else's region. The deployment is region-aware: CreateOS cloud, your VPC, or on-prem.

Built for Teams Whose Code Cannot Leave

Run the control plane and the storage inside your own network and the residency question is answered by architecture rather than by a setting in someone else's console.

If you need specific controls sequenced for your environment, we work through them with you directly.

  • Self-host the control plane and the storage in your own environment.
  • Bring your own object storage: S3, R2, Tigris, or MinIO.
  • Region-aware: CreateOS cloud, your VPC, or on-prem.

Common Questions

Can I self-host a code execution sandbox in my own cloud?

Yes. Run the CreateOS Sandbox control plane in your own environment and bring your own object storage, so code, snapshots, and data never leave your boundary.

Is this a self-hosted E2B alternative?

It is a sandbox you can run entirely in your own boundary, with the control plane and storage self-hosted. The differentiator beyond self-host is the combination: per-VM kernel isolation, sandbox networking, and kernel-level egress governance together.

Where does my data live when I self-host?

In object storage you control: S3, R2, Tigris, or MinIO. The control plane points at your buckets, so the code, the snapshots, and the data stay inside your environment by architecture, not by a setting in someone else's console.

Run code you can't trust, as systems, not snippets.