Senior GPU Compute Infrastructure Engineer (Remote)

Boundless; Northern (KY); Hybrid; USD 175,000 - 250,000; Full time

Benefits offered by this job

Equity allocation

Remote-first with off-sites

Health, dental, vision

Job summary

Boundless is coordinating GPU compute at scale as it becomes a leader in AI. As a Senior Infrastructure Engineer (GPU Compute), you'll build and operate the compute fabric powering our AI inference workloads across a large, heterogeneous GPU fleet including consumer RTX 5090 and datacenter hardware.

You will optimize scheduling, ensure always-on availability, drive down $/GPU-hour, and work with a remote-first, globally distributed team with a bias for action.

Qualifications

• 5+ years of infrastructure/DevOps experience operating large-scale production systems
• Deep expertise in Kubernetes, Docker, and container orchestration at scale
• Strong Linux systems administration skills
• Proficiency in infrastructure-as-code tools (Terraform, Ansible, Pulumi)
• Track record of managing mission-critical, high-throughput systems
• Strong IaC background in heterogeneous environments
• Proficiency in at least one scripting/programming language (Python, Bash, TypeScript, Go)
• Comfort navigating ambiguity with a bias for action

Responsibilities

• GPU Fleet Orchestration: operate a heterogeneous, multi-region GPU fleet using SkyPilot, Kubernetes/k3s, and cloud+on-prem providers
• Compute Scheduling & Utilization: maximize GPU utilization across inference workloads and manage placement across spot/on-prem/cloud
• Bare-Metal & GPU Optimization: optimize PCIe, ReBAR, NUMA, CUDA/driver tuning, memory config, and network topology

Job description

Boundless is coordinating GPU compute at scale as it becomes a leader in AI. As a Senior Infrastructure Engineer (GPU Compute), you'll build and operate the compute fabric powering our AI inference workloads across a large, heterogeneous GPU fleet including consumer RTX 5090 and datacenter hardware.

You will optimize scheduling, ensure always-on availability, drive down $/GPU-hour, and work with a remote-first, globally distributed team with a bias for action.