Why I build this

I got my first computer at six, a Windows 98 machine, and I never really stopped taking things apart to see how they work underneath. That is not a founding story. It is just why I like this work.

The founding story is more ordinary. I spent years advising organisations on getting their applications under control: universities, public authorities, mid-sized companies. And at the end of nearly every project, the same question was left standing in the room: so who operates this now?

Usually there was no good answer. Either it landed on one person in-house who kept it running on the side, or it went to a large provider where, eventually, nobody knew who to call. I have watched both fail. The second one fails more quietly.

What I built out of that

gh0stcloud is an attempt to answer that question without dodging it.

The platform runs the things nobody builds for fun: clusters, network rules, TLS, secrets, backups, monitoring. On top of it run either applications our customers deploy themselves, or applications we operate end to end. It is the same platform. The only difference is how much of the responsibility sits with us.

Both exist side by side because both are needed. A developer who wants to put something online in the evening does not need a sales call. A university having an examination system operated does not need self-service. It needs someone who picks up the phone.

What I do not claim

We are a small company. That is a real objection, and I think it is a fair one, which is why it is here and not in the fine print.

What I put against it: gh0stcloud follows one principle, full automation. The platform repairs itself first, and people step in where automation should not decide on its own. Those people have names: Linus Stiehm and me, with experienced operators from our network when a situation calls for more depth. How that works in detail is described in how operations keep running.

Changes go through Git, so it stays traceable who wanted what and when. Costs are measured by the hour rather than estimated. And the documentation is public, so nobody depends on one person being the only one who knows how it works. A provider you cannot leave is not a good provider, which is why leaving is documented too.

I do not promise availability I do not measure, and I do not promise response times that are not written down. For self-service they are on this site, for managed operations in your service description. And if gh0stcloud is not the right fit for what you are doing, I would rather say so in the first conversation than in the third month.

Why Germany

The data centres are in Germany, and that is not a marketing decision. Some of our customers are not permitted to do it any other way. For everyone else, it is simply one fewer question to resolve later.

If you are reading this

If you are considering handing us your operations: write to me directly. I have these conversations myself, and the most honest version of one is where we are both allowed to conclude that it does not fit.

If you have just built something and want it online: just start. You do not have to talk to anyone for that. That is the whole point.

Falk Hähner Founder, gh0stservice GmbH