Job summary
Seeking an experienced GPU Systems Engineer to contribute to the design, deployment, and optimization of large-scale distributed GPU compute clusters within an advanced HPC/AI research environment. The role involves performance tuning, troubleshooting, and automation across thousands of nodes in a globally distributed data center infrastructure supporting continuous global operations.
Location and work setup
- Location
- Norway
- Work setup
- On-site
Salary
USD 150000.00–300000.00 year
Responsibilities
You will design and optimize GPU compute clusters at scale, identify and resolve performance bottlenecks affecting compute, storage, and network components, and work closely with research and development teams to profile and fine-tune GPU workloads. Your duties include automating system deployment, monitoring, and troubleshooting across large node counts, managing critical infrastructure projects from concept through support, and testing and deploying new hardware and software while collaborating with vendors to address complex issues.
Qualifications
Candidates should have over five years of experience in large-scale Linux systems engineering within HPC, AI, or distributed infrastructure contexts. Expertise includes Linux system installation, performance tuning, and troubleshooting, deep knowledge of GPU workload optimization, proficiency in Python scripting and automation, and familiarity with CUDA or C/C++. Experience with NVIDIA-specific technologies such as NCCL, GPUDirect RDMA, and NVLink is desirable, along with knowledge of configuration management tools like Salt, Ansible, Puppet, or Chef. Strong diagnostic skills at hardware, OS, and network levels, coupled with excellent communication and teamwork abilities, are essential. Fluency in Norwegian is not required.