Senior SRE & Automation Engineer — GPU Cloud Reliability

Bitdeer Group; San Jose, Austin (CA, TX); Hybrid; USD 180,000 - 240,000; Full time

Job summary

Bitdeer Technologies Group seeks an experienced SRE to own the reliability of our customer-facing GPU cloud. You’ll ensure availability, low provisioning latency, and strong tenant experience, managing end-to-end lifecycle from onboarding to post-incident recovery.

You will work with Kubernetes at scale (100–10,000 GPUs), implement BMaaS provisioning, and set up tenant isolation, SLOs, and observability spanning Prometheus, Grafana, and alerting.

Qualifications

• 5+ years in SRE/cloud operations with GPU workloads.
• Deep Kubernetes operations and multi-tenant GPU configuration.
• Experience with Prometheus, Grafana, and scalable alerting.
• Hands-on Terraform, Helm, and GitOps (ArgoCD/Flux).
• Customer-facing cloud service experience and SLA management.
• Strong incident response and runbook automation capabilities.

Responsibilities

• Own end-to-end reliability of the GPU cloud service for tenants.
• Manage onboarding, provisioning, and tenancy isolation for workloads.
• Operate BMaaS provisioning and lifecycle workflows.
• Maintain SLI/SLO/SLAs and dashboards for customer visibility.
• Lead incident response with customer communication and post-incident reviews.
• Automate control-plane actions and self-service observability for tenants.

Skills

SRE experience

Cloud operations

Incident management

Capacity planning

Multi-tenant isolation

Automation and runbooks

Tools

Kubernetes

Prometheus

Grafana

Terraform

Helm

GitOps (ArgoCD/Flux)

Ironic/MAAS

Job description

Bitdeer Technologies Group seeks an experienced SRE to own the reliability of our customer-facing GPU cloud. You’ll ensure availability, low provisioning latency, and strong tenant experience, managing end-to-end lifecycle from onboarding to post-incident recovery.

You will work with Kubernetes at scale (100–10,000 GPUs), implement BMaaS provisioning, and set up tenant isolation, SLOs, and observability spanning Prometheus, Grafana, and alerting.