We built the tool we wanted for our own clusters
Standing up Talos on Hetzner by hand means the same dozen steps every time: image, load balancer, machine configs, servers, bootstrap, kubeconfig. A missed step is hard to spot, and a half-finished cluster is hard to clean up.
Beyond Cloud turns those steps into a workflow engine that records every step, retries what it can, and recovers runs that stall. It puts a workspace around that for credentials and team access. We run our own Talos clusters on Hetzner and manage them with it.
We are a small independent team and the product is in early access. You will talk to the engineers who build it.
How we work
- Show the work
- Every cluster is created by a workflow whose steps, timings and logs you can read. If something fails, you should see where and why without opening a support ticket.
- Your account, your resources
- We orchestrate infrastructure in your Hetzner project instead of reselling compute. You can always see and keep what we create.
- Standard parts
- Talos, Kubernetes and a normal kubeconfig. There is no proprietary runtime to migrate away from.
- Say what is real
- We list what works and what does not. We don’t show customer logos, uptime numbers or certifications we can’t back up.
What it's made of
- API and workflow engine. A single Go service that runs the HTTP API, WebSocket updates and the workflow engine.
- Provisioning. Hetzner Cloud API for servers, networks and load balancers. Talos for the operating system and Kubernetes bootstrap.
- Dashboard. A React app at app.usebey.com, plus an Android preview app.
- Secrets. AES-256-GCM encryption at rest, scoped to a workspace.
Talk to us
Questions, feature requests or a cluster you want to move: write to [email protected].