SRE, HPC AI Infrastructure (m/f/d)

Job description

Join your #OneTeam

Join the AI Infrastructure & Data team in the AI Business Unit. You will manage a team of three engineers designing, deploying, monitoring, maintaining, and evolving internal AI training and inference infrastructure. You will provide Data Science, ML Engineering, and internal product teams with a high-performance, available, secure platform that is efficient in cost and energy, using our data centers and industrial expertise. These teams will use our GPU, storage, and network infrastructure through a highly automated self-service platform, from dataset to production model.

Key responsibilities

• Manage three engineers responsible for designing, deploying, monitoring, maintaining, and evolving AI infrastructure.
• Serve as the single point of contact for internal client teams: gather needs, define service-level objectives (SLOs), and communicate.
• Own platform data governance, including lifecycle, classification, GDPR compliance, ISO 27001, SecNumCloud, and access.
• Define and maintain the target platform architecture, lead key architecture decisions (ADRs), and conduct code reviews.
• Manage SLOs, capacity, GPU budgets, and platform cost and energy-efficiency metrics.
• Provide experienced technical leadership in Infrastructure as Code, GitOps, CI/CD, observability, and high availability.
• Coach engineers through tailored personal and technical development.
• Participate in on-call rotations and act as escalation point and client contact during major incidents.
• Continually challenge the status quo and explore changes needed to meet internal needs.
• Apply hardware, software, and data security throughout the value chain.
• Work closely with OVHcloud teams in France and internationally.

Your impact after six months

You will understand the AI Infrastructure & Data team, IT department, existing infrastructure, current AI workload challenges, and internal teams' needs. You will have contributed GPU/HPC expertise and ideas to technical improvements and may have begun prototyping or testing orchestration or storage solutions. You will have established relationships with internal users such as Data Scientists and ML Engineers, gathered their needs and feedback, and created feedback loops for continuous improvement. You will have explored market solutions and best practices for AI infrastructure and recommended suitable approaches for OVHcloud.

Your impact after one year

You will have played a key role in deploying and automating new AI infrastructure, implementing tools and processes such as IaC and GitOps to ensure operations, security, and scalability. You will have shared your HPC/AI infrastructure expertise through internal training, presentations, or workshops and may have contributed to events or conferences. You will have engaged with the developer community, sharing knowledge, asking questions, and contributing to open-source projects.

Required skills

• Strong technical experience designing and operating GPU/HPC or large-scale cloud infrastructure.
• Proficiency with orchestrators such as Kubernetes and Slurm; IaC such as Terraform and Ansible; and CI/CD/GitOps tools such as ArgoCD and Prometheus.
• Strong knowledge of high-performance networking (InfiniBand, RoCE) and distributed storage (Ceph, Lustre, S3).
• Scripting/development skills in languages such as Python, Go, and Bash.
• Independent and able to work closely with your team and users of the services you provide.
• Able to focus on complex technical work and adapt to changing circumstances.
• Service-oriented, with a desire to have a positive impact on internal partners.
• Comfortable working internationally in English.