AI Data Center Infrastructure Architect

RAM QS LLC
Dallas, TX, United States
1 day ago
Apply on www.careerbuilder.com
Prepare application

Role details

Contract type
Permanent contract
Employment type
Full-time (> 32 hours)
Working hours
Regular working hours

Tech stack

Artificial Intelligence Cloud Computing Data Centers Data Center Infrastructure Management (CIM) Network Security Network Connections AI Infrastructure Computer Network Operations

Job description

You will assess readiness in four domains and score each against NVIDIA’s site readiness criteria.

  1. Power: rack density at 190 to 230 kW per rack, distribution including 480 VAC feeds, A/B redundancy at the rack, and the effect of NVIDIA power capping on provisioned capacity.
  2. Liquid cooling: coolant distribution units, the secondary (technology cooling system) loop at 45C supply, rack manifolds, water quality to ASHRAE guidelines, and leak detection with its interplay with fire suppression.
  3. Space: floor loading, rack layout, cable pathways, and the delivery path for 48U racks (dock, doors, freight elevator).
  4. Network: connectivity and the exact demarcation points where the Customer’s network and security hand off.

You will also:

  • Translate Vera Rubin NVL72 Reference Architecture requirements into facility requirements and review the facility’s design inputs against them.
  • Work with NVIDIA’s deployment team on planning inputs and site readiness criteria.
  • Define ownership boundaries, including the facility water loop versus the rack cooling loop.
  • Confirm how rack telemetry integrates with BMS and DCIM, and who signs off each commissioning level.
  • Identify the architecture decisions needed for deployment, including headroom for future expansion.
  • Document constraints, open decisions and dependencies.

Key Work Products

  • Initial Technical Readiness Assessment: red, amber or green score per domain against NVIDIA criteria, with the gap behind each score.
  • Architecture Decision and Dependency Log: master record of decisions, owners, dependencies and constraints.
  • Deployment Prerequisite Summary: what the facility and each party must provide before hardware arrives.
  • Open Customer Decision List: customer facing extract of the open items from the log, with owners and due dates.

Requirements

Network Operations Center, Artificial Intelligence (AI), Artificial Intelligence (AI), Consulting, Water Quality Testing, Fire Suppression/Control, Cargo/Freight, Network Connectivity, Network Security, Telemetry, * 12+ years in data center design or AI infrastructure, with recent work at Principal or Lead Architect level.

  • Personally led site readiness for at least one NVIDIA GB200 or GB300 NVL72 deployment.
  • Working command of facility power and cooling at the level needed to challenge MEP engineers: electrical one lines, CDU sizing, secondary loop design, floor loading.
  • Hands on knowledge of NVIDIA rack scale reference architectures and the DGX SuperPOD or NVIDIA Cloud Partner deployment model.
  • Experience writing readiness assessments and decision logs that a customer signs off on.
  • Comfortable working directly with NVIDIA deployment teams, facility owners and customer network and security teams.

Preferred Qualifications

  • Vera Rubin platform exposure (also called Vera Rubin NVL144 in earlier material).
  • Hyperscale or neocloud data center design background.
  • Direct liquid cooling commissioning experience, Level 1 through Level 5.
  • Familiarity with NVIDIA MGX rack specifications and OCP rack standards.
  • Professional Engineer license or equivalent is a plus, not required.

Apply for this position

This job is hosted externally. Click below to view the full posting and apply.

Apply on www.careerbuilder.com
Prepare application

Good distractions

Loading talks and stories from around this role…