Data Center Commissioning Engineer (GPU Clusters)

Leviathan Systems; Minneapolis (MN); On-site; USD 100,000 - 150,000; Full time

Job summary

Data Center Commissioning Engineer (GPU Clusters) is a full-time on-site role with U.S. field deployment and travel requirements. You will own the acceptance of GPU clusters, from power-on walkdowns to NCCL bandwidth validation and the as-built handoff package.

Responsibilities include pre-power walkdowns, level-by-level Cx L1–L5, burn-in soak, NCCL/fabric validation, and documenting results to client sign-off. Travel and on-site leadership are required.

Qualifications

• Experience in commissioning or data center engineering.
• Familiarity with test procedures and acceptance criteria.
• Ability to document results clearly and accurately.
• Willingness to travel and lead on-site acceptance work.

Responsibilities

• Run pre-power walkdowns and level-by-level commissioning (Cx L1–L5).
• Execute acceptance test plans: burn-in soak, NCCL/fabric validation, error checks.
• Verify power, cooling, and network results against design and OEM spec.
• Drive punch-list closure with the field crew and GC.
• Assemble the as-built + test-report package for client sign-off.

Skills

Commissioning

Test procedures

Documentation

Travel readiness

On-site leadership

Job description

Data Center Commissioning Engineer (GPU Clusters)

Full-timeU.S. field deployment + travelOn-site

Own the acceptance of GPU clusters — from power-on walkdowns to NCCL bandwidth validation and the as-built handoff package.

Responsibilities

• Run pre-power walkdowns and level-by-level commissioning (Cx L1–L5)


• Execute acceptance test plans: burn-in soak, NCCL/fabric validation, error checks


• Verify power, cooling, and network results against design and OEM spec


• Drive punch-list closure with the field crew and GC


• Assemble the as-built + test-report package for client sign-off

What we're looking for

• Commissioning, test, or data center engineering background


• Comfortable with test procedures, acceptance criteria, and documentation


• Familiar with GPU cluster or high-density infrastructure a strong plus


• Rigorous — you don't sign off on 'probably fine'


• Able to travel and lead on-site acceptance work