IT Infrastructure Engineer - RDT Hosting and Infrastructure

About the role

Roche's Scientific Computing and Data Infrastructure team builds computing environments and shared data services for its global R&D organizations. As an IT Infrastructure Engineer for RDT Hosting and Infrastructure, you will work inside cross-functional product teams, partnering with scientists and external providers to deliver scientific computing platforms that combine modern DevOps practices with research needs.

Key responsibilities

• Support day-to-day operations and maintenance of High Performance Storage (HPS) platforms and parallel file systems, keeping systems available, monitoring performance and storage efficiency, and assisting with structured change management.
• Apply a DevOps mindset using automation tools, basic infrastructure-as-code and CI/CD pipelines to build, update and maintain research environments with senior engineers.
• Work with internal customers and stakeholders across the service lifecycle, from architecture design and deployment to steady-state operation and refinement.
• Collaborate with researchers to optimize scientific workflows and batch job patterns, improving cluster utilization and throughput.
• Identify and implement improvements to technical systems, operational workflows and team collaboration.

Required profile

• 1-3 years of experience, ideally in a multinational or fast-paced setting; healthcare or life sciences experience is a plus.
• Bachelor's degree in Computer Science, IT, Engineering or a related field.
• Foundational knowledge of distributed storage and parallel file systems (e.g. GPFS, Lustre, BeeGFS).
• Working knowledge of Linux distributions (RHEL, CentOS, Rocky, Ubuntu) and command-line troubleshooting.
• English fluency.

Required skills

• Ability to write Bash scripts and basic Python for automation; routine use of Git.
• Hands-on experience with at least one configuration or automation tool (e.g. Ansible, Terraform) and basic CI/CD concepts.
• Familiarity with HPC environments, batch jobs or workload schedulers (e.g. Slurm, LSF).
• Nice to have: monitoring tools such as Prometheus, Grafana, ELK or Nagios.