I help engineering teams turn fragile Kubernetes clusters into infrastructure that upgrades, scales, and heals itself — whatever hardware it has to run on.
I close the gap between a cluster's desired state and what's actually running — building the operators, admission controls, and multi-cluster tooling that let a platform look after itself.
In practice: CRD-based operators that encode real operational judgment. Multi-cluster orchestration that behaves the same whether the target is a hyperscaler region or a rack in a data center. Helm charts and CLIs a team can actually run, day to day, long after rollout.
This is the short version. Current role and full background live on LinkedIn.