Hardware Operations Technician (IT Systems Technician)

Quick Facts

• Role: Hardware Operations Technician (IT Systems Technician)

• Focus: End-to-end operation and lifecycle management of a distributed Linux GPU cluster

• Work mode: Responsible role within an international, research-oriented team

Description

You will ensure the reliable hardware operation of a large distributed Linux GPU cluster across its full lifecycle—from planning and upgrades to component replacement. The role includes building/refreshing compute nodes (including GPU upgrades), evaluating new hardware with benchmarks to inform purchasing, and performing systematic hardware fault analysis and repair. You will also verify electrical safety for the cluster hardware and portable devices according to DGUV.

Responsibilities

• Own reliable cluster hardware operation from planning through exchange

• Build compute nodes, assemble single components, and upgrade existing systems with new GPUs

• Maximize simulation performance per invested euro

• Evaluate new hardware using benchmarks and derive procurement decisions

• Analyze hardware faults systematically and replace defective components

• Coordinate with suppliers and collaborate with internal technical teams and external service providers

• Perform DGUV electrical safety inspections for cluster hardware and portable devices

Requirements

• Completed training or degree (e.g., Electrical Engineering, Information Technology) or comparable practical expertise

• Preferred: certified electrician/electrical expert

• Very good knowledge and experience administering Linux clusters

• Advantage: GPU-computing experience

• Advantage: Infiniband/Ethernet knowledge and experience

• Advantage: Monitoring/automation experience

• Advantage: Knowledge of infrastructure for climate and power supply

• Very good German and good English

• Valid driver’s license class B

• Ability to use pragmatic, unconventional solutions

Benefits

• Responsible work with significant room for shaping your work

• Modern, high-performance hardware and software

• Supportive, low-hierarchy, research-focused culture with short decision paths

• Opportunities for professional and career development and support of initiative and creativity

• Position is initially limited to 2 years, with an unbefixed employment option not excluded

• Pay according to TVöD Bund based on qualifications and experience, plus public-service benefits