from 14 €/mo
Cloud servers S/M/L
Compute in Germany, backup and monitoring included
AI operations
Monitoring that does not only alert — it repairs
Incident detected, ticket opened, standard action carried out, human informed. Around the clock, from €49 per server and month.
- List of permitted actions in the contract
- Critical incidents always go to a human
- Monthly report
AI operations is an operating layer that watches your systems around the clock and responds to incidents itself. Monitoring detects the deviation, an agent diagnoses it, opens a ticket and carries out a standard action permitted in writing beforehand — restarting a service or rotating logs, for example. The agent then verifies whether the incident is gone and documents everything. Critical incidents always go to a human. There are three levels: watch, act, run.
Why “alerting faster” is not enough
Most night-time incidents are not mysteries but repetitions: a service hangs, a disk fills up, a certificate expires. They do not need analysis, they need an action — immediately. That is exactly what the agreed list of permitted interventions is for.
Three levels
| Level | What the agent does | Price per server |
|---|---|---|
| Watch | monitors, opens tickets, notifies, produces the monthly report | from €49/month |
| Act | in addition: carries out permitted standard actions at once; at night our technical support responds | from €129/month |
| Run | in addition: runs operations, plans capacity, proposes changes for approval | from €249/month |
What the agent is allowed to do
The list is part of the contract; you can read it and change it.
- Restart a service
- Rotate and archive logs
- Delete temporary files from a fixed list
- Renew a certificate
- Extend a disk within the agreed limit
- Clear a queue
- Start an out-of-schedule backup
What the agent never does: delete data · change configuration permanently · touch user permissions · shut systems down · trigger costs. A critical incident is never handled without a human.
What you receive every month
A report with three numbers: how often something deviated from the norm, what share was resolved automatically, how available the systems were. Plus the list of actions carried out.
Frequent questions
What if the agent gets it wrong?
Every action is traceable and limited to standard interventions; actions that risk data are not on the list. The agent checks the state before acting and the result afterwards.
Does the AI see our data?
No. It works with operational metrics — load, logs, service state. The contents of your applications are not part of monitoring.
Does this replace our service provider?
For the night-time routine, yes. Planning, changes and root-cause analysis stay with humans, deliberately.
Does it run on our own servers?
Yes, provided we have access. The details are settled during the inventory.
Work out the level
Tell us how many servers you have and what is critical at night. We propose the level and show the list of actions.
Book a callRelated services
from 59 €/mo
IT administration per seat
Users, devices, updates, first line for your staff
from 4 900 € one-off
On-premise AI server
Ready machine with a local model — your documents stay inside