Senior SRE: GPU Cloud & Multi-Tenant Automation

Bitdeer (NASDAQ: BTDR); Singapore; On-site; SGD 140,000 - 190,000; Full time

Job summary

Bitdeer, headquartered in Singapore, seeks an experienced Site Reliability Engineer to own the GPU cloud service and Kubernetes platform. You will scale multi-tenant workloads, automate provisioning, and drive reliability initiatives, operational readiness, and customer-facing SLAs.

You’ll lead BMaaS workflows, implement IaC with Terraform/Helm, and integrate GitOps for rapid, safe deployments. A hands-on SRE with GPU expertise will thrive in Bitdeer’s fast-paced, cloud-centric environment.

Qualifications

• 5+ years in SRE/cloud operations with GPU workloads.

Responsibilities

• End-to-end reliability of the GPU cloud service — availability, job completion, latency, tenant experience.
• Operate production Kubernetes clusters for 100–10,000 GPUs at scale.
• Manage Nvidia GPU operator, device plugin, MIG config, time-slicing, and multi-tenant policies.
• Topology-aware scheduling using GPU locality, NVLink, and network-aware placement.
• Handle tenant onboarding, quota, isolation (namespaces, RBAC, network policies) and offboarding.
• BMaaS provisioning, tenant handoff, lifecycle management and reclamation.
• Define and monitor SLIs/SLOs/SLAs; track cluster availability and provisioning latency.
• Incident management with customer communication, runbooks, escalation, status updates.
• Maintain monitoring stack: Prometheus, Grafana, Alertmanager, PagerDuty for tenant dashboards.
• Automate GPU node failure detection, drain/cordon/taint, reschedule duties.
• Infrastructure-as-code with Terraform, Helm, and GitOps (ArgoCD/Flux).
• Create customer-facing runbooks, capacity planning, and support tiering.

Skills

Kubernetes operations

GPU workloads

Terraform

Helm

GitOps

Prometheus/Grafana

Go or Python

Incident management

Multi-tenant isolation

BMaaS provisioning

Tools

NVIDIA GPU operator

Device plugin

MIG

Time-slicing

ArgoCD/Flux

Terraform

Helm

GitOps workflows

Job description

Bitdeer, headquartered in Singapore, seeks an experienced Site Reliability Engineer to own the GPU cloud service and Kubernetes platform. You will scale multi-tenant workloads, automate provisioning, and drive reliability initiatives, operational readiness, and customer-facing SLAs.

You’ll lead BMaaS workflows, implement IaC with Terraform/Helm, and integrate GitOps for rapid, safe deployments. A hands-on SRE with GPU expertise will thrive in Bitdeer’s fast-paced, cloud-centric environment.