5+ years in infrastructure or platform engineering, SRE, architecture, or field and solutions engineering roles
Must have Unix & Linux administrator experience
Deep Kubernetes expertise (CKA/CKS or equivalent), grounded in having run clusters in production rather than in a lab
Strong command of enterprise storage: Fibre Channel SAN (zoning, WWPNs, fabric login and path behavior), iSCSI, NVMe/TCP, NAS protocols, dm-multipath, LVM, and array-side provisioning, masking, and mapping
Real understanding of compute and server hardware: bare metal provisioning and lifecycle, BIOS, firmware and driver management, HBA and NIC selection, NUMA and CPU/memory topology, local NVMe layout and endurance
Production-grade virtualization knowledge
Plus an understanding of KubeVirt as a migration target
Practical data center networking: VLANs and trunking, MTU and jumbo frames, bonding and LACP, L2 versus L3 boundaries, routing, load balancer integration, and how partitions, congestion, and asymmetric paths present in a distributed storage system
Solid Linux systems background extending to the block and device layers: udev, kernel and sysctl tuning, systemd, I/O scheduling, and performance triage with tools like iostat, blktrace, and perf
Working understanding of customer development environments (CI/CD, application deployment patterns, dev and test workflows) sufficient to reason about platform needs beyond just the storage layer
Strong technical writing skills.