Senior GPU Cloud Platform Engineer

Bitdeer; San Jose (CA); On-site; USD 180,000 - 230,000; Full time

Job summary

Bitdeer, a world-leading AI and Bitcoin mining infra company, seeks an experienced Kubernetes SRE to build and operate the GPU-focused control plane for NeoCloud. You will design deployments, automate remediation, and ensure scalable, fault-tolerant workloads across a multi-tenant cluster.

The role emphasizes topology-aware scheduling, BMaaS provisioning, and incident-driven automation. You will collaborate with infra and platform teams to deliver reliable GPU workloads at scale in a

Qualifications

• 5+ years in Kubernetes operations.
• At least 2 years managing GPU workloads on K8S.
• Deep understanding of Nvidia GPU operator and device plugin.
• Experience with multi-tenant K8S platforms and isolation guarantees.
• Hands-on experience with bare-metal provisioning and lifecycle automation (Ironic, MAAS).

Responsibilities

• Manage production Kubernetes clusters for GPUs at scale (100–10,000 GPUs).
• Operate Nvidia GPU operator, device plugin, MIG, and time-slicing.
• Enforce topology-aware scheduling and NVLink awareness.
• Create CRDs for GPU workload lifecycle management.
• Integrate AI frameworks: Slurm, Ray, Kubeflow on K8S.
• Ensure multi-tenant isolation with namespaces and RBAC.
• Manage BMaaS provisioning and tenant onboarding workflows.
• Maintain IaC with Terraform providers and modules.
• Define SLIs/SLOs and monitor availability; manage incidents.
• Lead runbook automation and post-incident reviews.

Skills

Kubernetes operations

GPU workloads on Kubernetes

NVIDIA GPU operator

Topology-aware scheduling

Multi-tenant Kubernetes

Bare-metal provisioning

Terraform

Helm

GitOps workflows

Go or Python

SRE fundamentals

Prometheus/Grafana

RBAC & isolation

CI/CD tooling

Tools

ArgoCD

Flux

Prometheus

Grafana

Ironic/MAAS

Job description

Bitdeer, a world-leading AI and Bitcoin mining infra company, seeks an experienced Kubernetes SRE to build and operate the GPU-focused control plane for NeoCloud. You will design deployments, automate remediation, and ensure scalable, fault-tolerant workloads across a multi-tenant cluster.

The role emphasizes topology-aware scheduling, BMaaS provisioning, and incident-driven automation. You will collaborate with infra and platform teams to deliver reliable GPU workloads at scale in a