Infrastructure Engineer (Compute) - Remote

Overview

Fluidstack is the AI Cloud Platform. We build GPU supercomputers for top AI labs, governments, and enterprises. Our customers include Mistral, Poolside, Black Forest Labs, Meta, and more.

In Short

• Design and implement GPU/ASIC infrastructure at the server, rack, and system level.
• Troubleshoot complex GPU and compute system related failures.
• Develop and maintain hardware/firmware management services.
• Automate all aspects of the server lifecycle.
• Own end-to-end compute lifecycle, including partnering with vendors on RMAs.
• Serve as the main point of contact for hardware escalation and troubleshooting.
• Monitor system performance, identifying and resolving bottlenecks.
• Automate deployment and management tasks to improve efficiency.
• Collaborate with storage and network teams to ensure cohesive infrastructure operations.

Requirements

• 5+ years of experience in compute infrastructure engineering.
• Strong knowledge of Linux systems administration and performance tuning.
• Experience with bare metal provisioning tools (MaaS, Metal3, Tinkerbell, or other).
• Familiarity with GPU hardware and workload optimization, especially kernel and driver level requirements.
• Proficiency in automation tools (e.g., Ansible, Terraform).
• Experience operating Kubernetes and SLURM clusters.

Benefits

• Competitive total compensation package (salary + equity).
• Retirement or pension plan, in line with local norms.
• Health, dental, and vision insurance.
• Generous PTO policy, in line with local norms.
• Fluidstack is remote first, but has offices in key hubs. For all other locations, we provide access to WeWork.