Вакансия из Hirify Global, списка международных tech-компаний Для мэтча и отклика нужен Plus
Пожаловаться
Зарплата и рынок
развернуть
€40к-60к/год
€49к€55кмедиана€67к
По похожим ролям (8) ~ €55к/год
На СНГ рынке: $3.4к/мес ($2.7к - $4.2к)
86
Хорошая вакансия
развернуть
Чётко очерченная роль в эксплуатации HPC даёт практическую работу с современной AI-инфраструктурой и промышленными нагрузками; при этом задачи охватывают и инфраструктуру, и поддержку AI-процессов.
?Широкий круг задач
Современная инфраструктура
AI для промышленности
Мэтч & Сопровод
Для мэтча с этой вакансией нужен Plus
Подключить Plus
Описание вакансии
Текст:
Hirify AI/Оригинал
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 →