Remuneration
Not specified
Location
Lyon, ARA, France
Visa sponsorship
Not specified
Job summary
Lead a team of three engineers building and operating an internal GPU/HPC platform for AI training and inference. Provide data science, ML engineering, and product teams with secure, automated self-service infrastructure, while owning platform architecture, data governance, service-level objectives, capacity, GPU budgets, incident escalation, and cost and energy efficiency. The environment includes Kubernetes or Slurm, Terraform, GitOps, high-performance networking, and distributed storage.
Benefits
Hybrid remote-work policyEmployee share ownership planTenure recognition programVacation and sports subsidiesCompany childcare or crèche, depending on siteOnline training and certification platformDigital medical and social support for employees and their familiesWell-equipped officesMulticultural teams
Qualifications
- Strong experience designing and operating large-scale GPU/HPC or cloud infrastructure.
- Proficiency with orchestration tools such as Kubernetes or Slurm, Infrastructure as Code tools such as Terraform or Ansible, and CI/CD or GitOps tools such as ArgoCD.
- Strong knowledge of high-performance networking, including InfiniBand or RoCE, and distributed storage, including Ceph, Lustre, or S3.
- Proficiency in Python, Go, and Bash.
- Ability to work autonomously and collaboratively with engineers and internal service users.
- Strong service orientation and ability to adapt to changing contexts.
- Comfort working in an international environment in English.
Responsibilities
- Manage a team of three engineers.
- Design, deploy, monitor, maintain, and evolve AI infrastructure for training and inference.
- Serve as the primary contact for internal users, gather requirements, define service-level objectives, and communicate service commitments.
- Own platform data governance, including lifecycle, classification, access, and compliance with GDPR, ISO 27001, and SecNumCloud.
- Define platform architecture and lead architecture decision records and code reviews.
- Manage service-level objectives, capacity, GPU budgets, and cost and energy-efficiency metrics.
- Lead Infrastructure as Code, GitOps, CI/CD, observability, and high-availability practices.
- Coach engineers in their technical and personal development.
- Participate in on-call duty and act as an escalation contact during major incidents.
- Apply hardware, software, and data security throughout the platform.
- Collaborate with internal teams in France and internationally.
Skills
AnsibleArgo CDBashCephGoKubernetesPrometheusPythonS3TerraformOVH
Languages
EnglishPythonGoBash
Work schedule
On-call duty
Industry
Cloud computingArtificial intelligenceHigh-performance computing
Relocation
No