Technical Support Engineer

About the role

We are building our level 2 technical support team from the ground up, and we are looking for a Technical Support Engineer to join it. Our customers are not filing tickets about forgotten passwords; they are AI companies running massive GPU training jobs and production inference at scale, and when they contact us, something real is wrong.

Responsibilities

• Own level 2 tickets escalated from the global operations center, driving them to resolution or to a clean, well-evidenced handoff to engineering.
• Diagnose issues across the full stack, Linux hosts, GPU health, Kubernetes workloads, Slurm scheduling, Weka and Vast storage, and high-speed networking.
• Investigate degraded and failed training and inference workloads using logs, metrics, and cluster telemetry, and give customers a clear account of what happened and why.
• Triage GPU and node hardware faults, including nodes presenting fewer than 8 GPUs, RMA candidates, and thermal or power events, and coordinate with data center operations on remediation.
• Work directly with customer engineering teams in writing and on calls, matching their technical level and keeping them informed while an investigation is open.
• Write and maintain runbooks that let the global operations center resolve recurring issues at level 1 and flag the gaps where no runbook exists yet.
• Build diagnostic scripts and small tools that shorten investigations and reduce repeat work.
• Contribute to the customer-facing knowledge base and documentation, so recurring questions stop becoming tickets.
• Partner with technical account managers on account health, giving them the technical read behind utilization dips and ticket trends.
• Participate in an on-call rotation backing the global operations center outside business hours and help shape the teams move toward full 24-7 coverage.
• Feed repeated patterns back to engineering and product as evidence, not anecdote.

Requirements

• 3+ years in Linux systems administration, site reliability engineering, infrastructure operations, or a technical support engineering role with real diagnostic ownership.
• Strong Linux troubleshooting depth: system, networking, file systems, storage, process and resource investigation, log analysis, and comfort at the kernel boundary when applications go quiet.
• Working knowledge of Kubernetes, reading cluster and pod state, understanding scheduling and placement, and diagnosing why a workload will not run.
• Experience with a batch scheduler in a shared compute environment, ideally Slurm, including job submission, queue behavior, and node state.
• Scripting ability in Python or Bash, sufficient to automate diagnosis.
• Experience working tickets in a structured incident and tracking system such as JIRA, PagerDuty, or equivalent.
• Excellent written communication, specifically the ability to explain a technical finding clearly to an engineer who is frustrated and under time pressure.
• Willingness to participate in an on-call rotation.

Preferred Qualifications

• Background in GPU cloud, HPC, or AI/ML infrastructure operations.
• Familiarity with AMD Instinct platforms, MI300X, MI325X, MI355X, and the ROCm software ecosystem, or transferable depth with NVIDIA/CUDA.
• Experience diagnosing high-speed networking issues, RDMA, link performance, and the class of problems where the GPUs look healthy but throughput has collapsed.
• Exposure to Weka, Vast, or other high-performance storage environments.
• Container runtimes in HPC contexts, Docker, Enroot, Pixis, or Aptainer.
• Hands-on use of observability platforms such as Grafana and Prometheus.
• Understanding of distributed training frameworks, PyTorch, JAX, and inference-serving patterns, enough to recognize when a failure is the customer's code rather than our infrastructure.
• Experience with ITIL, or another structured incident and change management framework.