Senior Slurm HPC & Kubernetes Scheduler Engineer

Bitdeer Technologies Group; San Jose (CA); On-site; USD 190,000 - 270,000; Full time

Job summary

Bitdeer Technologies Group seeks a Staff Slurm Cluster & HPC Scheduling Engineer to own Slurm across GPU fleets, both bare metal and VM-based nodes, and lead the adoption of the Slinky operator stack for unified GPU resource management.

The role is highly hands-on, customer-facing during onboarding and escalations, with responsibility for multi-tenant scheduling, reliability, and platform-level work across Kubernetes integration and Terraform-driven automation.

Qualifications

• 8+ years in HPC, systems, or cloud infra engineering.
• 4+ years operating production Slurm clusters at 100+ GPU-node scale.
• Deep hands-on Slurm expertise with slurm.conf, slurmdbd, MUNGE and JWT authentication.
• Strong GPU and fabric fundamentals: NVIDIA drivers, InfiniBand/RoCE, DCGM.
• Production Kubernetes experience and Slurm-on-Kubernetes exposure.
• Go/Python scripting and automation experience.
• Mature English written and verbal communication.

Responsibilities

• Design, deploy, and operate production Slurm clusters on bare metal and VM-based GPUs.
• Lead topology-aware scheduling for GPU fabrics and validate NCCL bandwidth.
• Own multi-tenant scheduling policy, including partitions, QOS, and reservations.
• Drive Slinky on Kubernetes and evaluate slurm-bridge for co-scheduling.
• Manage elastic capacity between Slurm and Kubernetes based on demand.
• Operate Pyxis/Enroot and container runtimes with correct GPU constraints.
• Write runbooks, onboard enterprise customers, and mentor engineers.

Skills

HPC engineering

Slurm expertise

Kubernetes experience

Communication in English

Tools

Python

Bash

Go

Terraform

Ansible

Kubernetes

Docker

NVIDIA drivers

DCGM

InfiniBand/RoCE

NCCL

Slurm

Slinky

Job description

Bitdeer Technologies Group seeks a Staff Slurm Cluster & HPC Scheduling Engineer to own Slurm across GPU fleets, both bare metal and VM-based nodes, and lead the adoption of the Slinky operator stack for unified GPU resource management.

The role is highly hands-on, customer-facing during onboarding and escalations, with responsibility for multi-tenant scheduling, reliability, and platform-level work across Kubernetes integration and Terraform-driven automation.