2 days ago
San Jose, CA, USAMid Level
Base Salary
$201k - $301k/yr
Responsibilities
- Deploy, configure, maintain, and troubleshoot Linux-based CPU and GPU compute clusters for HPC, distributed AI training, and inference workloads.
- Administer Slurm and Kubernetes environments, including scheduling, resource allocation, access, containers, GPU runtimes, and workload deployments.
- Automate infrastructure provisioning, configuration, and lifecycle management using scripting, configuration management, and infrastructure-as-code tools.
- Validate compute hardware, BIOS, firmware, GPU drivers, and software compatibility, and support platform upgrades with testing, rollout, and rollback plans.
- Configure and troubleshoot host networking, high-speed Ethernet, RoCEv2, InfiniBand, RDMA interfaces, VLANs, MTU, routing, connectivity, congestion, and end-to-end performance.
- Troubleshoot issues across compute, GPU, networking, storage, virtualization, containers, and cloud environments.
- Perform cluster acceptance testing and workload benchmarking to improve performance, utilization, scalability, and reliability.
- Develop monitoring, dashboards, alerting, documentation, operating procedures, deployment guidance, and troubleshooting runbooks.
- Contribute to incident resolution, root-cause analysis, vulnerability remediation, and preventive reliability improvements.
- Collaborate with infrastructure teams, cloud providers, network engineers, researchers, and hardware and technology partners to introduce and support new compute capabilities.
Requirements
- Hands-on experience with Linux systems engineering, HPC infrastructure, distributed computing, GPU infrastructure, or large-scale compute environments.
- Experience deploying, administering, or supporting production HPC or GPU compute clusters.
- Strong Linux administration and troubleshooting skills across operating systems, networking, storage, and system performance.
- Experience with Slurm, Kubernetes, or comparable scheduling and orchestration platforms.
- Experience with GPU drivers, containerized GPU workloads, and accelerator software stacks; AMD Instinct GPUs, ROCm, or RCCL experience is particularly relevant.
- Knowledge of MPI, collective communication, and multi-node AI training concepts.
- Experience with high-speed Ethernet or InfiniBand networking, including 100/200/400 Gb/s environments.
- Knowledge of RDMA technologies such as RoCEv2 or InfiniBand, including queue pairs, PFC, and ECN.
- Familiarity with leaf-spine networking, ECMP, oversubscription, latency, bandwidth, and congestion.
- Familiarity with Kubernetes networking for GPU workloads, including CNI plugins, SR-IOV, or RDMA device plugins.
- Experience with shared or parallel storage such as NFS, Lustre, BeeGFS, IBM Storage Scale, or Weka.
- Proficiency with Python or Bash and automation or infrastructure tools such as Ansible, SaltStack, and Terraform.
- Experience with Git-based development workflows and monitoring tools such as Prometheus and Grafana.
- Experience with cloud infrastructure such as OCI, Azure, AWS, or specialized GPU cloud providers.
- Familiarity with KVM virtualization, PyTorch, and AI inference platforms such as vLLM or SGLang.
- A bachelor's or master's degree in Computer Science, Computer Engineering, or a related technical field is preferred.
Categories
About AMD
AMD designs and sells CPUs, GPUs, and adaptive/embedded computing products for PCs, data centers, gaming, and edge devices. Its portfolio includes Ryzen and EPYC processors, Radeon and Instinct graphics, and adaptive SoCs from its Xilinx acquisition, sold to OEMs, cloud providers, and device makers. Founded in 1969 and headquartered in Santa Clara, it is a public company on NASDAQ and supplies semi-custom chips for major game consoles.
