Hardware Specialist Remote

Nscale is the GPU cloud engineered for AI. We provide cost-effective, high-performance infrastructure for AI start-ups and large enterprise customers. Nscale enables AI-focused companies to achieve superior results by reducing the complexity of AI development. Our GPU cloud bolsters technical capabilities and directly supports strategic business outcomes, including cost management, rapid innovation, and environmental responsibility.

At Nscale, our Engineering team plays a critical role in driving the deployment and then subsequent management of our infrastructure and software platforms..

We thrive on a culture of relentless innovation, ownership, and accountability, where every team member takes pride in their work and drives it with excellence and urgency. As an Nscaler, you’ll build trust through openness and transparency, where everyone is inspired to do their best work. If you join our team, you’ll be contributing to building the technology that powers the future.

We’re looking for a Hardware Specialist to own the lifecycle of our GPU compute platforms from selecting and configuring the right systems with OEMs (Dell, Lenovo, etc.) to getting them powered, cooled, racked, networked, and kept current in production. You’ll be our primary liaison to vendors and the internal go‑to for NVIDIA GPU servers and their surrounding ecosystem.

• Liaise with hardware vendors (e.g., Dell, Lenovo; plus channel/SIs) to translate workload needs into server configurations and bills of materials (BOMs).
• Maintain a current view of the NVIDIA data center GPU portfolio and adjacent components (chassis/HGX assemblies, NVLink/NVSwitch options, NICs/DPUs, storage, optics).

• Produce accurate node and rack power budgets (typical/maximum), accounting for redundancy (A/B feeds), PSU efficiency, and power capping.
• Specify and coordinate PDUs (single/three‑phase, breaker sizing), UPS integration, and downstream distribution with facilities.
• Validate power draw during burn‑in and acceptance; track capacity vs. plan.

• Define and validate cooling strategies for high‑density GPU racks (air‑cooled and liquid‑ready): hot/cold aisle, ΔT targets, airflow direction, blanking, containment.
• Where applicable, plan for rear‑door heat exchangers (RDHx) or direct‑to‑chip (D2C) liquid cooling: CDUs, facility water parameters, quick‑disconnects, leak detection, and maintenance procedures.
• Monitor inlet/outlet temps and set alert thresholds tied to environmental telemetry.

• Establish and maintain golden images/profiles for BIOS/UEFI, BMC/iDRAC/iLO/XCC, NIC, NVMe, GPU firmware, and switch OS—plus rollback and emergency procedures.
• Automate configuration and updates using Redfish/IPMI/CLI/Ansible (or similar), and track CVEs/compliance baselines.
• Tune BIOS for performance/power (NUMA, C‑states, PCIe bifurcation/Above 4G, SR‑IOV) aligned to GPU workloads.

• Design and integrate host networking for AI/HPC clusters:
  Ethernet (incl. RoCEv2/DCB/PFC/ECN) and/or Infini Band (spine‑leaf, fabric bring‑up, and validation).
• Specify NICs, optics/transceivers, and cabling; implement LAG/LACP/MLAG/EVPN‑VXLAN as appropriate.

• Define hardware acceptance/burn‑in (power/thermal soak, memory/disk tests, GPU diagnostics).
• Track inventory, spares strategy, failure rates, and vendor SLA performance.
• Produce runbooks, rack elevations, wiring diagrams, firmware matrices, and change records consumable by on‑call and data center staff.
• Partner with Security on secure boot/TPM, firmware signing, and chain‑of‑custody; own data‑erasure standards for decommissioning.

• 5+ years building and operating enterprise/x86 server hardware in data centers, including hands‑on experience with GPU‑accelerated platforms.
• Proven vendor management with at least one major OEM (Dell Power Edge, Lenovo Think System, etc.).
• Demonstrable expertise in all of the following:

Power: per‑node/rack budgeting, A/B feeds, PDU selection, and validation.

Cooling: air and liquid‑ready designs for high‑density…