GPU DevOps Engineer (M/F/X)

METALINE is an IT services company specializing in outsourcing, managed services, Data Centers, and the Digital Workplace.

Founded with the ambition of supporting companies in managing and optimizing their IT infrastructures, METALINE has evolved to meet the technological challenges of today and tomorrow. As part of our growth and to meet the specific needs of one of our clients, we are looking for a GPU DevOps Engineer (M/F/X) for a strategic mission in Guyancourt (78).

You will join a team dedicated to artificial intelligence products and services, within the Data & IA domain, and contribute to the deployment, industrialization, and operation of On-Premise GPUaaS and LLMaaS platforms.

If you wish to evolve in a demanding technological environment and contribute to innovative AI projects, this mission is for you!

Main Missions

• Industrialize and automate the deployment of AI platforms in secure On-Premise environments.
• Contribute to the evolution of the architecture of GPUaaS and LLMaaS platforms.
• Participate in the implementation of managed services by integrating security, compliance, and governance constraints.
• Ensure the production deployment, documentation, and maintenance of operational and security conditions for AI platforms.
• Participate in the improvement and automation of deployment and operation processes.
• Perform presentations and demonstrations of the developed offers and services.
• Collaborate within a DevOps team in an Agile environment.
• Participate in Agile rituals, roadmap work, and backlog management.
• Participate in the technical on-call rotations of the department.

Profile Sought

Technical Skills

• Significant experience in DevOps and AI/ML platform management.
• Concrete experience in the management and orchestration of GPU resources, ideally with Kubernetes / OpenShift.
• Mastery of Kubernetes, OpenShift, and Docker.
• Good knowledge of On-Premise infrastructures and notions of Cloud environments.
• Experience with GPU orchestration solutions such as Run.ai, Slurm, ClearML, or Volcano.
• Knowledge of NVIDIA GPU Operator and MLOps environments.
• Mastery of automation and CI/CD tools: GitLab CI/CD, Jenkins.
• Good knowledge of monitoring tools: Prometheus, Grafana, ELK Stack, or Dynatrace.
• Mastery of Python and Bash scripting; Go would be a plus.
• Knowledge of AI/ML architectures and frameworks: Transformers, TensorFlow, PyTorch.
• A CKA/CKAD or OpenShift certification would be a plus.
• Experience with Kestra would be appreciated.
• Mastery of Atlassian (Jira, Wiki) tools.

Soft Skills and Interpersonal Skills

• Excellent interpersonal skills and ability to collaborate with different technical teams.
• Team spirit and a taste for working in an Agile environment.
• Proactive and able to contribute to platform evolutions.
• Rigor and organizational skills.
• Ability to communicate and present technical solutions.
• Operational English, with regular use in a professional context.

Experience and Education

• Higher education in computer science, Master's degree (Bac+5) or equivalent.
• 6 to 10 years of experience, including at least 5 years in the fields of Data and IT process automation.
• Experience in a banking environment or regulated sector would be appreciated.
• Significant experience in the operation and maintenance of production platforms is expected.

Recruitment Process

• First telephone exchange with a METALINE Recruitment Manager
• An interview with a METALINE Sales Manager
• Meeting with the client