Recently postedRemoteTechnology

Senior Staff Bare-Metal Provisioning Software Engineer - REMOTE

Synersys TechnologiesUnited States🇺🇸United States

Quick overview

SeniorityMid Senior

Work modeRemote

LocationUnited States

MicroservicesAnsibleDNSHTTPPythonVaultKubernetesLinux

Job description

Senior / Staff Bare-Metal Provisioning Software Engineer - REMOTE

We are seeking a software-first Bare-Metal Provisioning Engineer to architect and build the zero-touch hardware lifecycle control plane for our AI Inference Cloud (NVIDIA GPU, Intel x86, and SambaNova clusters over a Juniper fabric). Modeled after bare-metal cloud architectures (e.g., Equinix Metal / Tinkerbell), you will write Go and Python microservices that take raw servers from rack power-on to immutable Linux and Kubernetes readiness without human intervention.

What You’ll Own

OOB Hardware & Firmware Control Plane: Build and maintain Go/Python controllers (using bmclib, Redfish, IPMI, or metal-toolbox) to automate power states, BIOS/UEFI settings (PCIe bifurcation, SR-IOV, NUMA, C-states), and firmware updates across multi-vendor GPU/ASIC servers.

Zero-Touch Network Boot & OS Lifecycle: Architect automated iPXE/PXE provisioning pipelines (Tinkerbell, Matchbox, Ironic, or Metal3) to stream and boot immutable Linux and bare-metal Kubernetes (Talos, RKE2, Kubeadm).

Inventory & DDI Reconciliation: Automate hardware discovery, burn-in attestation, and state sync against NetBox (DCIM/IPAM plan of record) to dynamically drive local Kea DHCP (Option 66/67) and Split-Horizon CoreDNS.

Air-Gap-Survivable GitOps: Implement pull-only provisioning workflows (Argo CD, Flux, ansible-pull) with local GPG/Cosign signature verification, local Harbor OCI registry mirrors, and TPM 2.0 / Vault at-boot secret injection.

Must-Have Stack & Experience

5+ years building bare-metal provisioning software, hardware lifecycle engines, or fleet automation at a bare-metal cloud (Equinix Metal/Packet, CoreWeave, Crusoe), hyperscaler, or large-scale HPC/AI fleet.

Production Go & Python: Proven track record writing custom provisioning daemons, BMC/Redfish API clients, and declarative state machines in Go and Python.

Deep Boot & OOB Internals: Mastery of UEFI/BIOS, iPXE/PXE, HTTP/TFTP boot, Ignition/cloud-init, IPMI, Redfish, and Linux kernel/NUMA/PCIe tuning.

Autonomous Local Infra: Hands-on experience integrating NetBox APIs, local DHCP/DNS, and bare-metal Kubernetes without relying on public cloud managed service