Systems Engineer / Site Reliability Engineer - Los Angeles - Onsite
Role Summary: Systems Engineer / SRE
We are seeking a Systems Engineer / Site Reliability Engineer (SRE) to design, automate, and operate the core operating platform supporting our high-density AI inference infrastructure. In this role, you will 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. You will 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. By establishing proactive SLIs/SLOs, telemetry pipelines, and automated self-healing mechanisms, you will ensure high availability and deterministic performance across tens of thousands of distributed accelerator nodes.
Key Responsibilities & Technical Qualifications
The ideal candidate brings 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, Google Cloud Platform). You possess 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). You are proficient in infrastructure as code and configuration management (Terraform, Ansible, SaltStack) and write clean, production-grade automation in Python or Go.
Key responsibilities include managing Kubernetes clusters at scale, operating robust observability frameworks (Prometheus, VictoriaMetrics, Grafana, OpenTelemetry), troubleshooting distributed GPU and network bottlenecks (RoCEv2/InfiniBand), and driving blameless root-cause investigations to systematically engineer away manual operational toil.
