Systems Engineer/Site Reliability Engineer (SRE)

Tasks

• design, automate, and operate the core operating platform supporting our high-density AI inference infrastructure.
• bridge the gap between bare-metal accelerator hardware and upper-layer customer inference workloads, owning the reliability, scalability, and lifecycle automation of our compute fleet.
• architect automated provisioning pipelines (PXE/iPXE, ZTP), manage container orchestration layers (Kubernetes), and optimize Linux kernel/driver performance specifically tuned for high-throughput GPU execution.
• establishing proactive SLIs/SLOs, telemetry pipelines, and automated self-healing mechanisms
• managing Kubernetes clusters at scale, operating robust observability frameworks (Prometheus, Victoria Metrics, Grafana, Open Telemetry), troubleshooting distributed GPU and network bottlenecks (RoCEv2/InfiniBand), and driving blameless root-cause investigations to systematica

Requirements

• extensive systems engineering or SRE experience within hyperscale environments, large-scale distributed systems, or specialized GPU neocloud operators (e.g., CoreWeave, Nebius, Crusoe Cloud, AWS, GCP).
• deep Linux internals expertise including kernel tuning, systemd, NUMA topologies, PCIe subsystem debugging, and storage I/O optimization, alongside hands-on mastery of accelerator stacks (NVIDIA drivers, CUDA, fabric managers, NCCL).
• proficient in infrastructure as code and configuration management (Terraform, Ansible, Salt Stack) and write clean, production-grade automation in Python or Go.

Schedule

• Duration: 6-12 Months

Location

• San Jose, CA / Los Angeles, CA