HPC & AI Workload Orchestration Engineer

Roche & Company; Madrid; Presencial; EUR 90.000 - 120.000; Jornada completa

Descripción de la vacante

Roche seeks a Workload Orchestration Engineer to advance our HPC and AI compute platforms within the ACE team. You will design and deploy orchestration stacks, optimize schedules, and ensure efficient GPU and CPU utilization across multi-tenant environments.

You will work with Run:ai, SLURM, and container runtimes to enable scalable AI model training and high-performance simulations, driving reliability and throughput for Roche researchers and engineers.

Formación

• Bachelor’s degree or higher in Computer Science, Applied Mathematics, Computational Engineering, or a related technical field.
• 5+ years of systems engineering experience focusing on workload scheduling, resource management, and multi-tenant cluster optimization.
• Deep familiarity with Enterprise Linux and distributed systems architecture.
• Expert-level SLURM administration including partitions, accounting, and plugins; strong Singularity experience.
• Hands-on AI orchestration with Run:ai, Kubernetes, and containerized GPU scheduling.
• Understanding of high-speed interconnects (InfiniBand, RoCE) and MPI/NCCL multi-node communication.
• Automation skills for scheduler configurations and telemetry gathering.

Responsabilidades

• Design, deploy, and govern SLURM-based workload orchestration across HPC and AI platforms.
• Deploy Run:ai as core orchestration and virtualization layer for AI workloads with fractional GPU allocation.
• Integrate SLURM with Kubernetes and ensure secure, reproducible environments using Singularity/Apptainer.
• Define topology-aware scheduling and QoS policies to optimize multi-tenant usage.
• Profile and tune queues, fairness, and resource distribution; collaborate on telemetry dashboards.
• Troubleshoot distributed training, MPI synchronization, and driver issues; maintain IaC for scheduling policies.

Conocimientos

Leadership

Collaboration

Agile mindset

Problem solving

Educación

Bachelor’s or higher in CS/Math/Engineering

Herramientas

SLURM

Run:ai

Kubernetes

Singularity/Apptainer

Enroot

MPI

NCCL

InfiniBand

RoCE

Descripción del empleo

Roche seeks a Workload Orchestration Engineer to advance our HPC and AI compute platforms within the ACE team. You will design and deploy orchestration stacks, optimize schedules, and ensure efficient GPU and CPU utilization across multi-tenant environments.

You will work with Run:ai, SLURM, and container runtimes to enable scalable AI model training and high-performance simulations, driving reliability and throughput for Roche researchers and engineers.