Senior Software Engineer, GPU Infrastructure

Job Summary

Operating the Kubernetes GPU fleet and managing multi-tenancy, the full-time Senior Software Engineer, GPU Infrastructure will focus on maintaining large-scale experiments, ensuring fault tolerance, and enhancing platform security in a remote work environment.

Key responsibilities

• Handle day-to-day operations of the Kubernetes GPU fleet, including capacity planning and node lifecycle management
• Design and implement batch scheduling and storage solutions for the GPU cluster, ensuring efficient resource allocation
• Collaborate directly with research teams to address infrastructure challenges and improve platform functionality

Required qualifications

• 3+ years of experience in systems or infrastructure engineering on production Linux, specifically with GPU or large-scale batch platforms
• Proven experience managing production Kubernetes for GPU workloads, including batch layers and resource management
• Familiarity with infrastructure as code tools such as Terraform or Ansible, and monitoring solutions like Prometheus
• Strong programming skills in languages commonly used for infrastructure, such as Python, Go, Rust, or C++
• Ability to write clear documentation for engineers, researchers, and providers, including design documents and incident reports