Member Technical staff-Compute Infra HPC
Core
Build compute infra software for frontier AI compute clusters at global scale, ensuring health, automation, and secure scheduling of large GPU systems.
Role type
Senior IC distributed systems engineer (AI compute infrastructure)
Builds
Secure, observable, schedulable AI compute clusters and fleet lifecycle management systems
Domain
AI supercomputing, distributed systems, cloud infrastructure
Required skills
distributed systems, Kubernetes, Linux, infrastructure as code, control plane design, automation, telemetry, debugging complex cross-layer issues, technical leadership
Preferred skills
hyperscale compute infrastructure, cluster lifecycle systems, Kubernetes internals, cloud foundations, low-level systems (kernels, drivers, BMC), GPU/accelerator systems, high-performance interconnects (InfiniBand, RDMA), production reliability ownership
Technologies
Go, Rust, C++, C#, Java, Python, Kubernetes, Terraform, Linux, InfiniBand, RoCE, RDMA, NVLink, NCCL
Responsibilities
Design distributed services and control planes for cluster lifecycle; build primitives for cluster bootstrap and networking; automate rack/node qualification and fleet transition; develop node/fleet health platforms; build policy-driven automation for maintenance and rollouts; translate incidents into durable software; partner with hardware and datacenter teams; provide technical leadership and mentorship
Seniority
Senior, hands-on IC with leadership responsibilities