Senior MLOps Engineer

Overview

In this role you will provision and maintain multi-rack GPU server clusters for AI inference and training workloads. You will work hands-on within the MLOps and System Administration team to support research and development infrastructure. You will tackle advanced networking, Linux administration, and IaC-driven deployment to enable scalable, reliable workloads. This is a pivotal, visible role that shapes the core platform enabling researchers to pursue cutting-edge work.

Responsibilities

• Provision and maintain multi-rack GPU server clusters for inference and training
• Hands-on hardware and software orchestration across complex infrastructures
• Manage Linux servers across distributions with hypervisors/VMs
• Implement and manage IaC (Terraform, OpenTofu) for infrastructure deployment
• Design and configure network, switches, routers, and firewalls for high-performance clusters
• Collaborate in a small team to support R&D infrastructure and reporting
• Support storage and HPC-related components as part of the cluster ecosystem

Key requirements

• Significant engineering experience in System Administration, HPC Engineering, or MLOps Engineering
• Modern GPU server deployment, tuning and management
• Advanced networking technologies (Infiniband, RDMA, RoCE)
• Advanced Linux skills with servers running hypervisors/VMs and multiple distributions (e.g. Ubuntu, RHEL)
• Expertise in source code management (Git)
• IaC tools (Terraform, OpenTofu)
• Strong understanding of network design and switch/router/firewall configuration

• hands-on
• teamwork
• proactive problem-solving
• GPU server deployment and management
• Infiniband, RDMA, RoCE networking
• Linux with hypervisors/VMs

…