Operate and scale large GPU infrastructure across Linux, bare-metal, cloud, storage, networking, and Kubernetes environments. Build provisioning and automation systems, deploy platform improvements, troubleshoot incidents, support customer infrastructure, improve observability, and participate in an on-call rotation. Partner with infrastructure, networking, software platform, and customer success teams to improve reliability and reduce operational toil.

Support ML engineers operating production training and inference workloads across Kubernetes, cloud, and GPU infrastructure. Diagnose distributed PyTorch, CUDA, NCCL, networking, storage, scheduling, and performance issues; analyze observability data; advise customers during incidents; and improve reliability through automation, tooling, documentation, runbooks, and operational processes. The role is remote in the Philippines or Singapore and follows a Sunday-Wednesday APAC shift.