Infrastructure Engineer

About the Role

As an Infrastructure Engineer, you will be working closely with the SRE and ML teams to deliver fast, reliable on-prem inference. This will involve partnering with the ML and MLOps teams on on-prem inference serving, including model deployment, GPU scheduling, and performance tuning. Your work will have a direct impact on critical missions across private and public-sector customers.

The role is based in Germany, and the company offers a remote-first approach with occasional team events in Berlin. This provides a great opportunity for those looking to work remotely while still being part of a collaborative team.

What You Will Do

• Design, size, provision, and operate bare-metal GPU server fleets across on-prem and air-gapped environments
• Own the NVIDIA GPU stack end to end, including drivers, CUDA, GPU Operator, Container Toolkit, MIG, and DCGM
• Build the bare-metal substrate Kubernetes runs on, including node lifecycle, container runtime, GPU device plugins, node feature discovery, and kernel/NUMA tuning
• Engineer data-center networking and resilient storage, including VLANs/switching, RDMA, Ceph/ZFS/NVMe
• Partner with ML and MLOps on on-prem inference serving, including model deployment, GPU scheduling, and performance tuning
• Plan and run on-site build-outs, including rack integration, power/UPS and cooling sizing, commissioning, capacity planning, runbooks, and operator handover
• Operate in air-gapped or on-prem environments and travel to customer sites for builds and deployments
• Document your work, focusing on methodical and calm approaches during hardware incidents
• Collaborate with the SRE and ML teams to deliver fast, reliable on-prem inference

What We Are Looking For

• 5+ years of experience in bare-metal, HPC/GPU, data-center, or systems infrastructure engineering
• Hands-on ownership of physical and compute infrastructure, including firmware, BMC, PXE, kernel, and storage tuning
• Strong bare-metal Linux skills, including RHEL/Rocky/Ubuntu, and solid networking and storage fundamentals
• Real experience with the NVIDIA GPU stack, including drivers, CUDA, GPU Operator, MIG, and DCGM
• Experience serving GPU models in production, including model deployment, GPU scheduling, and performance tuning
• Comfortable operating in air-gapped or on-prem environments and traveling to customer sites for builds and deployments
• Eligible to work in Germany

Nice to Have

• German language skills (B1+)
• NVIDIA DGX/HGX or Slurm experience
• InfiniBand/RDMA fabrics experience
• Inference optimization experience, including TensorRT-LLM, vLLM, and quantization
• Certifications such as NVIDIA NCP-AIO, Red Hat RHCSA/RHCE, or CKA/CKS
• Field-engineering experience and familiarity with secure or regulated deployment environments

Benefits and Perks

• Modern architecture and stack
• Remote-first approach in Germany with occasional team events in Berlin
• Home office budget and great equipment
• 30 days vacation
• Direct impact on critical missions across private and public-sector customers
• Opportunity to work on a next-generation data intelligence platform using cutting-edge technologies
• Collaborative team environment with a focus on remote work
• Professional development opportunities, including training and certifications
• Access to the latest technologies and tools, including NVIDIA GPU stacks and Kubernetes

• Tip: Make sure you have hands-on experience with the NVIDIA GPU stack, including drivers, CUDA, GPU Operator, MIG, and DCGM, as this is a crucial part of the role.
• Tip: Highlight your experience with bare-metal Linux, including RHEL/Rocky/Ubuntu, and your understanding of networking and storage fundamentals.
• Tip: Be prepared to discuss your experience with on-prem inference serving, including model deployment, GPU scheduling, and performance tuning.
• Tip: Emphasize your ability to work in air-gapped or on-prem environments and your willingness to travel to customer sites for builds and deployments.
• Tip: Show your understanding of the company's technology stack and your enthusiasm for working on a next-generation data intelligence platform.
• Tip: Be prepared to provide examples of your experience with documentation, focusing on methodical and calm approaches during hardware incidents.
• Tip: Research the company culture and values, and be prepared to discuss how you can contribute to the team's collaborative environment.