Systems, infrastructure and integration

I have moved from physical servers through virtualization and the cloud to containers and agentic systems. Every layer added tools and none retired the one before it – which is why I know what still suffices in a given environment and what no longer does.

First understand, then change. A system is not bad just because it is old – if it does its job, it can keep running. The trouble comes when an outdated design blocks scaling, flexibility or maintenance. So I start with mapping and documentation, and only then propose changes.

I do not treat security as a separate discipline bolted on at the end. It runs from everyday system administration – patching, separated privileges, backups and verifying them – to the audit and certification requirements that have to be evidenced.

I prefer efficient, stable solutions built on open source – you can look inside it and adapt it as needed, and it does not depend on a single vendor. I measure success by what remains after me: a system that runs, documentation that holds, and people who know why.