Orchestration Workload Engineer - ACE - AI Factory

Roche is seeking a Workload Orchestration Engineer for its Accelerated Compute Engineering (ACE) team. Roche focuses on preventing, stopping and curing diseases while ensuring long‑term access to healthcare worldwide. The ACE team acts as a centre of excellence for high‑performance compute and AI infrastructure, supporting researchers, data scientists and engineers across the enterprise. In this position you will be the recognised internal specialist for workload orchestration, responsible for advancing and maintaining the scheduler stack on Roche's high‑performance computing (HPC) platforms. Your daily work will involve designing, scaling and fine‑tuning SLURM configurations, integrating custom plugins, managing topology‑aware scheduling and GPU resources, and bridging SLURM with cloud‑native tools such as Kubernetes. You will also embed container standards (Singularity/Apptainer) into SLURM, troubleshoot multi‑tenant bottlenecks, and collaborate with observability engineers to build telemetry dashboards that monitor job efficiency, queue times and hardware utilisation. The role requires a solid academic background and deep technical expertise. Candidates must hold a bachelor's or higher degree in computer science, applied mathematics, computational engineering or a related field, and possess extensive systems‑engineering experience focused on workload scheduling and SLURM administration. Proven mastery of SLURM architecture, including scaling, upgrading, policy design and GPU resource modelling, is essential, as is strong knowledge of SLURM operations, accounting and observability. Hands‑on experience with Kubernetes and HPC container runtimes such as Singularity or Apptainer is also required, together with a background in life‑science, pharmaceutical R&D or high‑performance scientific research environments. Demonstrated ability to lead complex technical initiatives and mentor engineering peers is a further must‑have. The ACE team operates globally, delivering mission‑critical on‑premises and cloud infrastructure that underpins Roche's digital products. You will work on cross‑functional projects that set standards and policies for compute environments across the company, gaining exposure to a broad range of scientific and AI workloads. While the description does not specify location benefits, the role offers the chance to influence a worldwide compute platform and to grow within a collaborative, inclusive culture that values personal expression and open dialogue. What the role asks for: - Bachelor's or advanced degree in CS, Applied Math, Computational Engineering - Extensive systems engineering experience in workload scheduling and SLURM administration - Proven expertise in SLURM architecture, scaling, tuning, and GPU resource management - Deep knowledge of SLURM operations, accounting, and observability tooling - Hands‑on experience with Kubernetes and HPC container runtimes (Singularity, Apptainer) - Experience in life sciences, pharma R&D, or high‑performance scientific research - Demonstrated leadership of complex technical initiatives and engineering mentorship - Nice‑to‑have: Experience integrating SLURM with Kubernetes (e.g., Slinky, Run:ai)