Назад
Company hidden
11 дней назад

Senior Cluster Orchestration Engineer (AI)

Формат работы
remote (только Europe)/hybrid
Тип работы
fulltime
Грейд
senior
Английский
b2
Страна
UK/Europe/Finland
Вакансия из списка Hirify.GlobalВакансия из Hirify Global, списка международных tech-компаний
Для мэтча и отклика нужен Plus

Мэтч & Сопровод

Для мэтча с этой вакансией нужен Plus

Описание вакансии

Текст:
/
TL;DR
Senior Cluster Orchestration Engineer (AI) (Kubernetes/SLURM): Building the scheduling and control-plane layer that unifies training, inference, and batch workloads across shared GPU clusters with an accent on workload orchestration, capacity management, and platform reliability. Focus on gang scheduling, topology-aware placement, preemption, Kubernetes operators, GPU utilisation, and GitOps delivery at scale.

Location: Hybrid in London, UK or Helsinki, Finland; remote work is available elsewhere in Europe.

Company

hirify.global is building a full-stack AI cloud spanning data centers, hardware, and a cloud platform for AI teams.

What you will do

  • Design and build scheduling systems that share GPU capacity across Kubernetes and SLURM for training, inference, and batch workloads.
  • Manage perishable, short-window capacity with preemptible work and clean reclamation for higher-priority workloads.
  • Implement gang scheduling, topology-aware placement, preemption, quotas, priorities, fair-share, and queueing using or extending Volcano, KAI Scheduler, Kueue, or Slurm Bridge.
  • Develop Kubernetes operators and controllers for cluster creation, scaling, upgrades, health checks, and decommissioning.
  • Handle interactions between elastic training, checkpointing, inference autoscaling, and batch backfill competing for shared GPUs.
  • Build scheduling observability and deliver the platform through GitOps, including drift detection and safe rollout and recovery.

Requirements

  • 5+ years of experience in platform engineering, infrastructure engineering, SRE, or a related field, or equivalent expertise, with a track record of taking projects from design through deployment and maintenance.
  • Strong production software development skills in Go or Python, including Kubernetes controllers or operators.
  • Knowledge of API design, concurrency, idempotency, retries, and error recovery.
  • Deep Kubernetes knowledge covering the scheduler, scheduling framework and extension points, control plane, networking, storage, and node lifecycle.
  • Hands-on experience with batch or AI workload schedulers such as Volcano, KAI Scheduler, Kueue, or Slurm Bridge.
  • Clear understanding of gang scheduling, queueing, quotas, preemption, and fair-share.
  • Experience running GPU workloads, including device plugins, GPU-aware scheduling, multi-node training, and latency-sensitive inference.
  • Familiarity with SLURM and the differences between HPC-style and cloud-native scheduling models.

Culture & Benefits

  • Full-time, permanent employment.
  • Cash and equity compensation with local benefits.
  • Work alongside engineers, researchers, and partners across the AI ecosystem.
  • Small team with direct architectural influence and a short path from idea to production.
  • Access to state-of-the-art GPU hardware at real scale.
  • Applications must be submitted through the Careers page; applications sent by email are not accepted.

Hiring process

  • There is no artificial application deadline; hiring proceeds once a suitable candidate is found.
  • Submit the application through hirify.global's Careers page rather than by email.

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 →