Engineer, Data Center

LinkedIn•Hillsboro, OR

2d•Onsite

About The Position

This role will be based in Hillsboro, OR. At LinkedIn, our approach to flexible work is centered on trust and optimized for culture, connection, clarity, and the evolving needs of our business. The work location of this role is on-site, meaning it will be performed in the specified office on a full-time basis due to the business needs of the team. As an onsite Data Center Engineer, you will be responsible for hardware deployments and acting as a point of escalation for our contracted employees responsible for addressing server break fix tickets. You will also assist Senior Data Center Engineers with the deployment of network, storage, and server nodes within your region. We are looking for candidates with server and network hardware installation experience, network cabling experience, server break fix experience, experience troubleshooting Linux Kickstart issues, and Linux/Unix troubleshooting skills.

Requirements

Bachelor's Degree or equivalent industry experience.
Relevant trade or technical certification (electrical, mechanical/HVAC, facilities, or military technical training) will also be considered.
2+ years in a data center operations, critical facilities, or comparable infrastructure role.
Hands-on server hardware break-fix at the component level — DIMMs, CPUs, fans, power supplies, HDD/SSD/NVMe, risers, and cabling.
Server and network deployment experience — racking, cabling, and initial configuration, including structured cabling and labeling standards and QC of vendor-completed installs.
Physical-layer network troubleshooting across copper and fiber — optics and transceivers, console connectivity, link and path validation.
Linux troubleshooting at the OS level (boot/POST, log review, basic system diagnosis).
Experience with imaging or provisioning workflows such as kickstart, autobuild, or PXE.
Out-of-band management experience — IPMI/BMC, iDRAC, iLO, or equivalent.
Experience with data center critical infrastructure — UPS, generator, switchgear, PDUs and busway, and chiller plant — and an understanding of redundancy topology (N+1, 2N).

Nice To Haves

Direct liquid-cooling experience — CDUs (Vertiv, Motivair, CoolIT, Boyd, nVent), rear-door heat exchangers, manifolds, quick-disconnects, and drip/leak detection.
Adjacent closed-loop experience also applies: chilled-water HVAC, industrial process cooling, semiconductor fab facilities, power generation, or marine/Navy engineering.
GPU and ML hardware exposure — NVIDIA HGX/DGX platforms, high-density rack deployments, and OEM break-fix coordination.
Water chemistry and fluid quality management — conductivity, corrosion inhibitor and biocide treatment, PG25 vs. treated water, filtration micron ratings, and sampling cadence.
MOP and SOP review experience — evaluating maintenance procedures before execution.
Vendor or data center provider coordination, including escorting, scheduling, and holding third parties to standard.
Experience with incident dashboards, DCIM, BMS, or provider portals, including the judgment to determine what to act on, escalate, or suppress.
Ticket-driven operational discipline — owning a queue against SLA expectations and coordinating resolution across teams and vendors.
Change management discipline in a live production environment, plus operational safety fundamentals (lockout/tagout, arc flash awareness, PPE, OSHA and state regulations).
Ability to read schematics, P&IDs, and technical diagrams to determine the proper sequence of operations for maintenance activities.

Responsibilities

Operate, monitor, and maintain the liquid-cooling infrastructure supporting LinkedIn's GPU and machine learning fleet, including Coolant Distribution Units (CDUs), secondary/technology cooling loops, manifolds, quick-disconnects, and rear-door heat exchangers.
Monitor loop health — supply/return temperature (ΔT), differential pressure (ΔP), flow rate, and fluid conductivity — and act on alarms before they impact GPU availability.
Perform and coordinate coolant sampling, water chemistry and fluid quality testing, filter changes, and fluid top-offs in accordance with OEM and LinkedIn standards.
On a daily basis, review LinkedIn's DCIM Incident Dashboard and act on any fatal or warning incidents, escalating to your regional Site Supervisor or Senior Critical Infrastructure Engineer as necessary.
Triage warning notifications to determine how they should be acted upon or suppressed.
Monitor environmentals for all data center suites and ensure they remain within ASHRAE standards as defined in our leases with providers.
Utilize internal tools (DCIM) and external tools (provider portals, BMS) to monitor power load, capacity, and environmental data across all data center suites.
Review all MOPs, SOPs, and maintenance advisories provided by the facility operator; flag gaps, sequencing risk, or deviations from LinkedIn standards before work begins.
Ensure major maintenance activities are captured in LinkedIn's internal change management process with appropriate risk classification, notification, and rollback plans.
Maintain the regional Critical Infrastructure maintenance calendar and keep stakeholders informed of upcoming high-risk activities.
Coordinate and schedule supplier/vendor onsite activities, including access, escort, and safety briefings.
Exercise stop-work authority — recognize when vendor work is unsafe, out of sequence, or out of compliance, and halt the work to address concerns.
In collaboration with the provider's site engineering team, support annual and quarterly preventive maintenance on all critical infrastructure, verifying that LinkedIn standards and industry best practices are followed.
Support server, storage, and network deployments within your region — racking, cabling, initial configuration, and quality control of vendor-completed installs against LinkedIn cabling and labeling standards.
Perform and support component-level hardware break-fix (DIMMs, CPUs, fans, power supplies, HDD/SSD/NVMe, risers) and serve as an escalation point for managed-services technicians.
Troubleshoot physical-layer network issues across copper and fiber, including optics, console connectivity, and link and path validation.