Роль описана четко и сочетает руководство командой с практической работой в быстрорастущей AI-инфраструктуре, но регулярные дежурства, внерабочие эскалации и поездки добавляют заметную нагрузку.
Дежурства и эскалации
Современная инфраструктура
Масштабирование команды
Мэтч & Сопровод
Для мэтча с этой вакансией нужен Plus
Подключить Plus
Описание вакансии
Текст:
Hirify AI/Оригинал
TL;DR
AI Reliability Engineering Manager (AI Infrastructure/GPU Operations): Running an AI Reliability Engineering team of eight to ten engineers across EMEA and operating an estate of approximately 29,000 GPUs with an accent on queue ownership, SLA performance, incident escalation, and operational automation. Focus on leading P1 response, improving SOPs, diagnosing Linux and high-performance fabric issues, and scaling follow-the-sun operations toward a one-million-GPU planning target.
Location: UK, London; the role covers EMEA and requires out-of-hours escalations, on-call participation, and travel to data centers and customer sites.
Company
Nscale operates AI infrastructure and reliability operations for large-scale GPU estates.
What you will do
Lead a team of eight to ten AI Reliability Engineers through 1:1s, objectives, reviews, probation, development, and hiring.
Own the operational queue, including prioritization, SLA adherence, MTTR, backlog, near-breach ownership, and handoffs to Engineering and vendors.
Run the team rota and on-call coverage within the EMEA model and global follow-the-sun handover.
Act as the first management escalation point for P1 incidents and customer escalations; lead post-incident reviews and convert recurring incidents into problem records and durable fixes.
Own pillar SOPs and drive automation that reduces manual remediation time.
Coordinate with Data Center Operations, Network Operations, Engineering, fleet tooling, and automation teams; report weekly operational metrics and explain changes.
Requirements
6+ years in infrastructure, operations, or support engineering in production environments, including 2+ years working with GPU, HPC, or large data center estates.
3+ years managing engineers in an operational support function, including reviews, development plans, and performance management through to outcomes.
Experience running a queue against SLAs and accountability for operational metrics.
Clear, specific, and concise written and verbal communication.
Decisive and pragmatic approach to making calls with incomplete information, improving processes without slowing delivery, and committing to decisions.
Solid production Linux and networking skills, working knowledge of GPU fleet failures and repairs, ITIL and SRE practices, and sufficient scripting ability to review engineering work.
Comfort with out-of-hours escalations, on-call work, and travel to data centers and customer sites.
Nice to have
NVIDIA GPU platform experience with nvidia-smi, DCGM, XID interpretation, and RMA; AMD Instinct experience is beneficial.
InfiniBand or RoCE diagnostics, NCCL, NVLink, NVSwitch, and Slurm with Pyxis and Enroot.
Rack-scale systems such as GB200 and GB300 NVL72.
BMC and Redfish, MAAS, NetBox, OpenStack, Kubernetes, and Ansible or Terraform.
High-performance storage such as VAST, Ceph, or NFS at scale.
Follow-the-sun operations and ITIL, Linux, networking, Kubernetes, or NVIDIA certifications.
Culture & Benefits
Operational coverage follows an EMEA model with global follow-the-sun handovers.
The role includes on-call responsibilities, out-of-hours escalations, and travel to data centers and customer sites.
The current infrastructure estate is approximately 29,000 GPUs, with a planning target of one million GPUs by the end of 2027.
Hiring process
Nscale does not accept unsolicited candidate submissions from recruitment agencies.
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 →