Member of Technical Staff Research Infrastructure Engineer

You will maintain and scale research infrastructure for large-scale training workloads. You will optimize application and infrastructure performance, work with research teams on cost-efficient solutions, diagnose distributed-system bottlenecks, build telemetry and monitoring, and participate in on-call incident response.

Responsibilities

• Maintain and optimize research infrastructure
• Scale infrastructure while maintaining reliability and performance
• Design cost-efficient infrastructure solutions with research teams
• Identify and resolve distributed-system performance bottlenecks
• Build and evolve telemetry and monitoring systems
• Participate in on-call rotations and incident response

Requirements

• Experience building or operating large-scale training platforms
• Experience with large-scale GPU compute clusters
• Ability to debug performance and reliability issues across distributed fleets
• Knowledge of Kubernetes, Infrastructure as Code, AWS, and GCP
• Experience with SLURM
• Knowledge of Python, Bash, Go, NVIDIA GPU drivers and operators, OpenTelemetry, and Prometheus

Benefits

• Equity
• Reasonable travel costs covered
• Relocation encouraged but not required