ML Ops Engineer

Join our Team

About this opportunity:

We're seeking an MLOps Engineer to support Ericsson's AI infrastructure operations on the Sferical.ai platform. You'll be the technical bridge between our AI teams and infrastructure, ensuring 216+ GPUs are used efficiently, securely and reliably. This role focuses on operational excellence, troubleshooting complex systems and enabling smooth AI workload execution in a governed enterprise environment.

What you will do:

• Operate and monitor GPU workload orchestration using Run:ai, managing projects, queues, resource allocation and job submissions


• Troubleshoot Kubernetes clusters, including pod scheduling, storage connectivity, deployments and runtime issues


• Support data movement across storage systems (S3, NFS) and manage access controls through enterprise identity systems


• Monitor infrastructure health, track GPU utilization and produce capacity reports for planning and governance


• Coordinate with users, internal teams and external support vendors to resolve incidents and process change requests


• Prepare infrastructure for future scaling, including Kubernetes upgrades, inference enablement and security integrations


• Create operational guides, dashboards and documentation to help teams navigate the platform

The skills you bring:

• Strong hands-on experience with Kubernetes, pod scheduling, RBAC, storage management and API-based troubleshooting


• Working knowledge of GPU scheduling platforms, distributed training and AI workload patterns


• Practical experience with object storage (S3), NFS, data pipelines and large-scale data movement


• Understanding of enterprise security, access controls, compliance workflows and governed network environments


• Experience troubleshooting across infrastructure layers—networking, storage, containers and applications


• Structured, security-conscious mindset comfortable in restrictive and approval-driven environments


• Experience with VAST Data, Run:ai, NVIDIA platforms, Slurm or distributed training frameworks is a plus