Site Reliability Engineer – GPU/HPC Infrastructure (Remote – MENA) بدوام كامل في الإمارات العربية المتحدة
التخصص
متفرقات
نوع الدوام
دوام كامل
الخبرة
أي مستوى
الموقع
الإمارات العربية المتحدة
وصف الوظيفة
This is a fully remote role open to candidates across the MENA regionAbout Saturn CloudSaturn Cloud builds infrastructure for running AI, machine learning, and data workloads at scale. Our platform helps teams develop, deploy, and operate compute-intensive workloads across modern cloud and GPU infrastructure.We’re looking for a GPU/HPC-focused Site Reliability Engineer based in the MENA region to help operate and troubleshoot the large-scale GPU infrastructure supporting Saturn Cloud Token Factory.This role is focused on the infrastructure below and around the Kubernetes layer. You’ll serve as a technical escalation point for GPU health, NVIDIA software, high-performance networking, topology, and distributed GPU performance issues affecting production inference workloads.The ideal candidate comes from GPU infrastructure, HPC, AI infrastructure, neocloud, hyperscaler, or large-scale ML platform operations rather than traditional application support.We value engineers who actively use modern AI coding agents to improve the speed and quality of their engineering and operational work.What You’ll DoDiagnose and resolve production issues affecting large-scale GPU inference infrastructureTroubleshoot NVIDIA datacenter GPUs, drivers, CUDA compatibility, and GPU container runtimesInvestigate GPU health issues, including Xid errors and hardware or driver failure modesDiagnose PCIe, NUMA, GPU placement, NVLink, and NVSwitch issuesTroubleshoot multi-GPU and multi-node workloadsInvestigate high-performance networking and distributed communication failuresDistinguish application and inference-runtime issues from GPU, fabric, topology, node, driver, or hardware failuresWork with Kubernetes-based GPU workloads and NVIDIA GPU Operator/device pluginsUse production observability and GPU metrics to diagnose reliability and performance issuesWork directly with GPU-cloud and infrastructure providers when incidents require hardware or fabric investigationProduce clear technical evidence showing where the failure occurs and what the appropriate infrastructure team needs to investigateWhat We’re Looking ForCore SkillsHands-on experience using AI coding agents and agentic development tools to accelerate engineering, debugging, automation, and operational workflowsDeep Linux systems debugging experienceNVIDIA datacenter GPU administrationNVIDIA driver installation, upgrades, and troubleshootingStrong understanding of CUDA and driver compatibilityNVIDIA Container Toolkit/runtimeExperience with NVML and nvidia-smiGPU health diagnostics, including Xid errors and common hardware/driver failure modesUnderstanding of PCIe topology, NUMA, and GPU placementUnderstanding of NVLink and NVSwitch fundamentalsFamiliarity with Kubernetes GPU Operator and device pluginsExperience running containerized GPU workloadsHigh-Performance NetworkingYou should have strong experience with several of the following:InfiniBandRDMARoCENCCLGPUDirect RDMANIC/GPU topologyNCCL testing and distributed workload diagnosticsBandwidth and latency troubleshootingMulti-node GPU communication failuresYou should be able to determine whether a production issue originates in the application, GPU, network fabric, topology, node, or another underlying infrastructure layer.Inference InfrastructureYou don’t need to be an ML researcher, but you should understand how modern inference workloads exercise GPU infrastructure.Relevant experience includes:vLLM, NVIDIA Dynamo, Triton, or comparable inference runtimesTensor and pipeline parallelismModel loading and GPU memory consumptionKV cacheContinuous batchingGPU and memory utilizationOut-of-memory diagnosisMulti-GPU and multi-node inferenceBasic inference performance analysis, including throughput, latency, and GPU saturationAdditional SkillsKubernetes troubleshooting sufficient to independently investigate GPU workloads inside a clusterPrometheus/Grafana and NVIDIA DCGM metricsBash and Pythoncontainerd/DockerExperience with bare-metal GPU clusters or GPU cloud infrastructureFamiliarity with B200/B300/H200/H100-class systems is highly desirableStrong PlusesNVIDIA DCGMNVIDIA DynamoKAI Scheduler or GroveSpectrum-XMellanox/ConnectX networkingOFED/DOCASlurm or HPC cluster administrationKubernetes-based GPU cloudsLarge GPU fleet operationsGPU burn-in, qualification, and health-check toolingWhat Success Looks LikeWhen an inference workload becomes unavailable or significantly slower, you can determine whether the problem is caused by the inference runtime, GPU memory pressure, a failed GPU, a driver/kernel interaction, PCIe/NVLink/NVSwitch topology, NCCL, RDMA/InfiniBand fabric, or the underlying node.You can collect enough evidence to distinguish a Saturn Cloud software issue from an operator infrastructure or hardware problem and communicate precisely what an infrastructure provider needs to investigate.You’re the person the team turns to when Kubernetes looks healthy, but the GPU workload isn’t.Why Saturn CloudRemote-first culture with a high-trust, high-ownership environmentWork directly with cutting-edge GPU and AI infrastructureSolve complex production problems spanning hardware, networking, Kubernetes, and modern inference systemsHelp shape the reliability and operational practices behind large-scale AI workloadsCompensation & BenefitsCompetitive salaryFlexible PTOFully remote position
التعليقات
لا توجد تعليقات منشورة حتى الآن.