Technical Support Engineer

You will be the first technical contact for customers experiencing GPU cluster issues. Own tickets from initial response through resolution, manage SLA commitments, reproduce problems, collect relevant logs and configuration, and determine whether faults originate with the platform or a provider. Work daily in Linux and the command line while troubleshooting SSH access, NFS mounts, storage, quotas, security groups, containers, drivers, billing, and account issues. Execute and improve runbooks, maintain customer documentation, communicate clearly under pressure, escalate with complete context, and participate in an on-call rotation across time zones.

Responsibilities

• Own customer tickets from first response through resolution
• Track SLA commitments and classify issue severity
• Reproduce problems and collect relevant logs and configuration
• Troubleshoot SSH access, NFS mounts, storage, quotas, security groups, containers, drivers, billing, and account issues
• Determine whether faults originate with the platform or a provider
• Execute and author runbooks
• Maintain customer-facing documentation and internal knowledge bases
• Escalate issues with complete technical context
• Participate in an on-call rotation for critical issues

Requirements

• Very strong Linux and command-line experience
• Experience owning infrastructure or cloud support tickets against response SLAs
• Solid networking and storage fundamentals including SSH, NFS, mounts, DNS, firewalls, and security groups
• Working familiarity with GPU workloads including nvidia-smi, drivers, CUDA, and containers
• Clear and fast written communication under time pressure
• Good judgment about knowledge limits and when to escalate
• Comfort working across time zones and participating in on-call rotations
• Experience with ticketing and on-call tools such as Zendesk, Linear, or PagerDuty is preferred
• Bash or Python scripting experience is preferred
• Exposure to Slurm, Kubernetes, or Docker in multi-tenant environments is preferred
• Background in GPU cloud, HPC, or hardware-adjacent support is preferred