Senior SRE - Customer-Facing GPU Cloud Reliability

Bitdeer; San Jose (CA); On-site; USD 180,000 - 260,000; Full time

Job summary

Bitdeer Technologies Group is building an AI-operated GPU cloud with a customer- facing service where reliability is the product. You will own end-to-end reliability from onboarding to workload execution, incident response, and post-incident recovery.

You will design observability and automation to make a 10,000-GPU cloud dependable for tenants who rely on it. You will manage production Kubernetes clusters optimized for GPU workloads, implement multi-tenant isolation, and drive SLOs/SLAs with

Qualifications

• 5+ years in SRE / cloud operations, with GPU workloads experience.
• Deep Kubernetes operations knowledge for GPU workloads.
• Experience with multi-tenant cloud platforms and strong isolation.
• Customer-facing service ownership with SLA/SLO awareness.
• Proficiency in Terraform, Helm, and GitOps workflows.

Responsibilities

• Own reliability of the customer-facing GPU cloud service end-to-end.
• Manage production Kubernetes clusters (100–10,000 GPUs).
• Configure Nvidia GPU operator, device plugin, MIG, and multi-tenant policies.
• Develop topology-aware scheduling and tenant isolation mechanisms.
• Lead incident response, runbooks, and post-incident reviews.
• Build self-service observability and capacity planning tooling.

Skills

SRE

Kubernetes

GPU workloads

Terraform

Go/Python

Observability

Incident management

Multi-tenant security

Tools

Terraform

Helm

GitOps (ArgoCD/Flux)

Prometheus

Grafana

Job description

Bitdeer Technologies Group is building an AI-operated GPU cloud with a customer- facing service where reliability is the product. You will own end-to-end reliability from onboarding to workload execution, incident response, and post-incident recovery.

You will design observability and automation to make a 10,000-GPU cloud dependable for tenants who rely on it. You will manage production Kubernetes clusters optimized for GPU workloads, implement multi-tenant isolation, and drive SLOs/SLAs with