I've been doing this since 2014.

I started in Linux systems administration — LDAP in high availability, MySQL clustering, load balancers, the unglamorous infrastructure that everything else sat on. That grounding turns out to matter more than I expected. A lot of cloud-native engineering is the same problems with better tooling, and knowing the problems underneath makes the tooling make sense.

Since then I've worked across telecom, SaaS and enterprise environments, on systems where availability carried regulatory weight and maintenance windows didn't really exist. Most of my work now is Kubernetes platform architecture — the networking, the tenancy model, the delivery pipeline, the guardrails — plus the migrations that get workloads there and the cost work that keeps them affordable.

I'm based in Stockholm, Sweden and work with clients across Europe and the US. I take on architecture engagements, fixed-scope builds, and fractional platform lead work for teams that need architect-level judgement without a full-time hire.

The thing I'm actually selling isn't Kubernetes. Plenty of people know Kubernetes. It's judgement about which of forty possible answers fits a particular team's size, skills and deadline — and the willingness to tell you when the answer is “don't do this at all.”