> Markdown version of [/jobs/ext/4806947-ai-data-center-infrastructure-architect](https://www.wearedevelopers.com/jobs/ext/4806947-ai-data-center-infrastructure-architect). Every page supports `.md` or `Accept: text/markdown`. Links point to the HTML versions so they work for humans too. Agent guide: [/agents.md](https://www.wearedevelopers.com/agents.md). --- # AI Data Center Infrastructure Architect - **Company:** RAM QS LLC - **Location:** Dallas, TX, United States - **Contract:** Permanent contract - **Skills:** Artificial Intelligence, Cloud Computing, Data Centers, Data Center Infrastructure Management (CIM), Network Security, Network Connections, AI Infrastructure, Computer Network Operations - **Published:** November 20, 2026 - **Apply:** https://www.careerbuilder.com/job-details/ai-data-center-infrastructure-architect-dallas-tx--01b9d8bc-f9d6-428b-b113-baaaab568d0e ## About the Role 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. ## 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.