Назад
Company hidden
12 дней назад

HPC / AI Specialist

40 000 - 60 000€
Формат работы
hybrid
Тип работы
fulltime
Грейд
senior
Английский
b2
Страна
Italy
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
HPC / AI Specialist (Linux/HPC/GPU): Operating and optimizing Linux-based HPC infrastructure for AI training, fine-tuning, inference, and data-intensive pipelines with an accent on workload scheduling, distributed storage, GPU memory optimization, and multi-tenant performance. Focus on supporting production AI workloads, troubleshooting compute and storage bottlenecks, and improving stable execution for engineers, researchers, and industrial clients.

Location: AI4I, OGR – Turin, Italy. Hybrid work with flexible arrangements that may be negotiated.

Salary: €40,000–€60,000 gross/year, depending on experience.

Company

AI4I is a research institute focused on application-oriented artificial intelligence, technology transfer, industrial innovation, and Industry 4.0.

What you will do

  • Operate and maintain Linux-based HPC clusters supporting AI training, fine-tuning, inference, and data-intensive pipelines.
  • Manage GPU and CPU environments, workload scheduling, resource isolation, performance tuning, and multi-tenant utilization.
  • Support distributed and software-defined storage for large-scale datasets, including VAST or equivalent systems.
  • Provide technical support for distributed training, parameter-efficient fine-tuning, checkpointing, data pipelines, runtime configuration, and GPU memory optimization.
  • Monitor system health, performance, and capacity; troubleshoot production issues and compute or storage bottlenecks.
  • Collaborate with infrastructure, engineering, research, and deployment teams to evolve system architecture and support industrial AI clients.

Requirements

  • Strong Linux system administration experience in production environments.
  • Solid understanding of CPU and GPU architectures and performance characteristics.
  • Experience operating HPC clusters or large-scale compute environments.
  • Hands-on experience with distributed and software-defined storage systems, such as VAST or equivalent.
  • Experience with workload managers and job schedulers, such as Slurm or equivalent.
  • Practical understanding of AI training and fine-tuning workloads, including GPU memory management, batch sizing, distributed execution constraints, checkpointing, and data pipeline performance.
  • Experience with parameter-efficient fine-tuning, runtime optimization, shared infrastructure, uptime responsibility, and production AI/ML training workloads.
  • Scripting and automation skills in Bash, Python, or equivalent; familiarity with containers, monitoring and observability, multi-tenant compute, and high-throughput networking.
  • Experience collaborating with engineering or deployment teams in production settings.

Culture & Benefits

  • Collaborative environment with engineers and researchers working on industrial AI deployments.
  • Direct impact on daily AI workloads and production systems.
  • Access to advanced computing infrastructure, including NVIDIA B200 accelerators and high-performance distributed storage.
  • Dedicated high-performance computing resources, state-of-the-art laboratories, and strong industrial collaborations.
  • Competitive compensation and performance-based incentives.
  • The role is shared 50% with the new Deployment Unit and works across infrastructure and client-facing teams; multiple candidates may be hired.

Hiring process

  • Submit the application exclusively through the online form.
  • Include a cover letter of no more than one page explaining how the profile fits the position, plus a CV and optional links to technical projects or operational experience.
  • Applications are accepted until December 31, 2026, or until the position is filled.

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 →