Lead Data Center Technician (Level 3)

About Corvex

Corvex is a publicly traded AI cloud company (Nasdaq: MOVE) that builds and operates GPU-accelerated infrastructure for AI workloads. We run large-scale NVIDIA GPU clusters, high-throughput storage, and confidential computing for AI companies, enterprises, and government customers, and our data center footprint is growing fast.

The Role

We're looking for a Lead Data Center Technician to be the senior on-site technical resource at our Delaware facility, a multi-megawatt NVIDIA GPU environment running InfiniBand and Ethernet fabrics. You'll start as the lead technician on site, owning day-to-day operations, doing the hands-on work, and working independently with our remote engineering, network, and NOC teams, hardware vendors, and the colocation provider. As the site grows, you'll help hire and train the team that joins you, build out the site on-call rotation, and take on more leadership of the Delaware operation.

What You'll Do

• Own day-to-day site operations, setting priorities and deciding when to escalate

• Serve as the primary on-site contact for the colocation provider, covering access, deliveries, facility tickets, and power and cooling escalations

• Work tickets with the remote NOC and engineering teams, providing clear findings, logs, and photos

• Maintain and repair GPU compute, storage, and network hardware, including drives, DIMMs, PSUs, fans, NICs/HCAs, optics, and cables

• Troubleshoot hardware faults to root cause using POST codes, BIOS, BMC, and serial console

• Diagnose physical-layer network issues, including optical levels, link errors, dirty connectors, and failed DAC or fiber cables

• Perform OEM warranty repairs under self-maintainer certification and support RMAs from diagnostics through part return

• Schedule, escort, and oversee vendor and contractor visits, and direct contract labor as needed

• Write procedures for after-hours remote hands work and verify it was completed correctly

• Receive and inventory inbound equipment, and manage the on-site spares pool

• Keep DCIM records (devices, rack elevations, serials, power, cabling) accurate and audit them regularly

• Install, relocate, and decommission equipment, including racking, cabling, power, labeling, and validation

• Maintain airflow containment, physical security, and a clean, audit-ready space

• Build and maintain site runbooks and MOPs, and follow change management and maintenance windows

• As the site grows, help hire, onboard, and train new technicians, and build out the site on-call rotation and escalation paths

What We're Looking For

• 5+ years of hands-on data center hardware experience, including time as a senior or lead technician

• Hands-on experience with NVIDIA GPU servers, including baseboard, heatsink, and power stage replacement to OEM procedure

• Field experience troubleshooting NVIDIA InfiniBand and Ethernet fabrics, including switches, optics, and cabling

• Component-level server troubleshooting (POST, BIOS, BMC, PXE)

• Solid networking fundamentals, including TCP/IP, switching, routing, and out-of-band and console access

• Structured cabling experience with copper and fiber (single mode and multimode, MPO/LC, DAC/AOC), including inspection and cleaning

• Working knowledge of three-phase power and rack PDUs

• Experience with ITSM ticketing, colocation providers, vendor dispatches, and asset and inventory records

• Experience training or mentoring other technicians

• Proven ability to work independently with minimal supervision, manage competing priorities, and make sound escalation calls

• Clear written and verbal communication, especially when documenting findings for remote teams

• Ability to obtain and maintain OEM self-maintainer certification

• Ability to pass a background check and meet colocation access requirements

• Valid driver's license and reliable transportation

• Ability to lift up to 50 lbs, work in tight spaces behind racks, and handle sustained data center noise

• Availability for on-call coverage and occasional after-hours maintenance windows and incidents

Bonus points for:

• Hands-on GPU module (SXM) replacement on NVIDIA HGX systems

• Linux command line and Python or Bash scripting for diagnostics, audits, and automation

• Experience with NetBox or a similar DCIM platform

• Familiarity with liquid-cooled environments, including CDUs, manifolds, and rear door heat exchangers

• Experience hiring technicians, building a site team, or setting up on-call rotations

• Certifications such as CompTIA Server+ or Network+, NVIDIA, or OEM hardware

Corvex is an Equal Opportunity Employer. All qualified applicants will receive consideration for employment without regard to race, color, religion, sex, national origin, sexual orientation, gender identity, disability, protected veteran status, or any other characteristic protected by law.