ailinuxinfrastructure as codecudagpuhpcnvidiaslurminfiniband
Вакансия из Hirify Global, списка международных tech-компаний Для мэтча и отклика нужен Plus
Пожаловаться
Зарплата и рынок
ЗП не указанаразвернуть
На похожих ролях по рынку
$89к$101кмедиана$146к
32 в выборке
На СНГ рынке: $3.6к/мес ($2.5к - $4.3к)
88
Отличная вакансия
развернуть
Чёткая роль senior-уровня в быстро растущей сфере AI-инфраструктуры с большой технической ответственностью; при этом стоит учитывать дежурства и ограничение по локации Великобританией.
Дежурства
Современные технологии
Ответственность за результат
Сфера AI
Мэтч & Сопровод
Для мэтча с этой вакансией нужен Plus
Подключить Plus
Описание вакансии
Текст:
Hirify AI/Оригинал
TL;DR
HPC Engineer (AI Infrastructure): Validating, tuning, and operating large-scale NVIDIA GPU clusters for AI and HPC workloads with an accent on InfiniBand fabrics, distributed performance, and production reliability. Focus on automating cluster validation, diagnosing GPU and fabric faults, optimizing NCCL and storage performance, and building monitoring for customer-facing infrastructure.
Location: Remote, based in the UK, with occasional travel to data centre sites. London, England, United Kingdom are listed as role locations.
Company
hirify.global operates Hyperstack, a self-serve GPU cloud and AI infrastructure platform for high-performance AI, HPC, and compute-heavy workloads, powered by renewable energy and NVIDIA architecture.
What you will do
Validate NVIDIA GPU clusters before customer handover through burn-in, health checks, NCCL, HPL, and storage benchmarks; resolve underperformance.
Operate production HPC clusters, including Slurm, GPU drivers, CUDA, InfiniBand fabrics, and parallel storage.
Tune clusters for AI training and HPC workloads using NCCL settings, GPUDirect RDMA, topology-aware scheduling, and storage optimization.
Build GPU, fabric, and job-level monitoring; lead incident response and root-cause analysis for GPU and fabric failures.
Automate provisioning, configuration, and validation with infrastructure-as-code and Git-based workflows.
Collaborate with solution architects, network, validation, support, NVIDIA, and OEM teams; turn incident learning into checks, runbooks, and documentation.
Requirements
Hands-on experience operating NVIDIA GPU infrastructure at scale, including drivers, CUDA, DCGM, and GPU hardware fault diagnosis.
Production HPC cluster operations experience with a workload manager such as Slurm.
Strong practical InfiniBand experience, including subnet management, RDMA, fabric monitoring, and fault-finding.
Experience benchmarking and tuning distributed workloads with NCCL tests, HPL, or similar tools.
Strong Linux systems administration skills covering OS-level networking and storage, infrastructure automation, and Git-based workflows.
Nice to have
Experience with large NVIDIA platforms such as HGX H100/H200/B200 or GB200 rack-scale systems, NVIDIA UFM, and GPUDirect RDMA.
Experience with parallel file systems such as WEKA, VAST, DDN/Lustre, or GPFS.
Experience running GPU workloads on Kubernetes or integrating HPC clusters with Kubernetes or OpenStack.
Background supporting AI/ML training, MPI-based HPC applications, research computing, or open-source HPC projects.
Culture & Benefits
Permanent remote role based in the UK with an on-call rota and occasional data centre travel.
Flexible working arrangements, remote or hybrid depending on role and location.
Competitive salary, annual discretionary bonus scheme, and employee wellbeing benefits.
25 days of holiday plus public holidays, private medical insurance, and enhanced pension scheme.
Ownership, autonomy, career progression, and a collaborative international culture focused on trust, transparency, and impact.
Будьте осторожны
Если работодатель просит войти в их систему, используя iCloud/Google, прислать код/пароль, запустить код/ПО, не делайте этого. Это мошенники. Обязательно жмите «Пожаловаться» или пишите в поддержку. Подробнее в гайде →