Customer Reliability Engineer
Core
Triage, debug, and resolve customer issues for large AI training and inference workloads running on GPU clusters sourced from global providers.
Role type
Senior IC Customer Reliability Engineer (Linux/Kubernetes)
Builds
Monitoring scripts, dashboards, alerts, and runbooks for cluster and node health.
Domain
AI Infrastructure / GPU Compute / Cloud Operations
Deliverable
production ML models
Required skills
Linux troubleshooting, Kubernetes, Python scripting, Bash scripting, incident management, technical writing, GPU failure analysis
Preferred skills
NVIDIA GPU production experience, InfiniBand/RoCE networking, HPC/batch schedulers, Prometheus/Grafana/Datadog, Terraform/Ansible/Helm
Technologies
Linux, Kubernetes, Python, Bash, NVIDIA GPUs, InfiniBand, RoCE, Slurm, Prometheus, Grafana, Datadog, Terraform, Ansible, Helm
Responsibilities
Triage and fix customer issues from reproduction to closure; Debug Linux layer problems (processes, networking, storage, kernel logs); Investigate Kubernetes issues (pods, nodes, scheduling); Handle 24/7 on-call incidents and update customers; Push external providers to resolve faults; Build automation scripts and monitoring tooling.
Seniority
Senior, hands-on IC
