Senior Architect & Lead Engineer — Server Management Platform

Principal / Lead Go Systems Engineer — Bare-Metal & GPU Infrastructure Project Overview: We are building a greenfield, high-scale server management platform for a GPU server manufacturer. The platform will manage Compal BMC hardware via Redfish/IPMI, monitor next-gen NVIDIA (B200/B300) and AMD (MI300/MI350) GPUs, and deliver a white-labeled management console shipped with every server. This is an intensive, 18–20 week project requiring a senior-level systems expert who can achieve immediate technical autonomy from week one. What You Will Own: • BMC adapter development — Compal/AMI MegaRAC SP-X Redfish quirks and IPMI fallback • Two-tier architecture — control plane + zone workers designed for 25,000 device scale • GPU monitoring design — NVIDIA DCGM + AMD ROCm SMI integration architecture • Code review — every critical PR, same-day turnaround • Phase 0 solo — BMC characterization report delivered in week 1 • Client-facing — architecture decisions and technical escalations Requirements (Must Have) • 5+ years Go — systems-level, not just APIs • Redfish / IPMI / BMC — gofish, bmclib hands-on production experience • Distributed systems — NATS JetStream, Temporal, PostgreSQL at scale • GPU monitoring — NVIDIA DCGM or AMD ROCm SMI experience • Kubernetes + Helm — production deployments • Available immediately — no exceptions Tech Stack: Go 1.22+ · Redfish · IPMI · bmclib · gofish · DCGM · ROCm SMI · Temporal · NATS JetStream · PostgreSQL 16 · VictoriaMetrics · Kubernetes · Helm This is the critical path role. The entire team depends on this person. Only apply if you have direct BMC protocol experience (gofish or bmclib). Generic proposals will not be reviewed. Best of Luck