Sr SRE & Automation Engineer (Customer Facing)
Core
Own end-to-end reliability of a customer-facing AI GPU cloud service, managing 10,000-GPU clusters, tenant lifecycle, and incident response to ensure SLA compliance.
Role type
Senior Site Reliability Engineer (SRE) & Automation Engineer
Builds
AI-operated GPU cloud platform for tenant training, fine-tuning, and inference workloads
Domain
AI infrastructure, GPU cloud computing, Kubernetes operations
Deliverable
production ML models
Required skills
Kubernetes operations, GPU workload management (Nvidia operator, MIG, time-slicing), topology-aware scheduling, multi-tenant isolation, Terraform, Helm, GitOps, Go/Python programming, AIOps, runbook-as-code
Preferred skills
Bare-metal provisioning automation, customer-facing SLA management, self-service observability design
Technologies
Kubernetes, Nvidia GPU operator, Prometheus, Grafana, Alertmanager, PagerDuty, Terraform, Helm, ArgoCD, Flux, Go, Python
Responsibilities
Manage production Kubernetes clusters for 100–10,000 GPUs, handle tenant onboarding/offboarding and quota management, execute incident response and post-incident recovery, build automated remediation workflows, maintain monitoring and observability stacks, define and meet customer SLIs/SLOs/SLAs
Seniority
Senior, hands-on IC