DevOps Engineer for HEP GPU Benchmarks (IT-CE-LCG-2026-168-GRAP)

The Worldwide LHC Computing Grid (WLCG) project is a global collaboration of around 170 computing centres in more than 40 countries, linking up national and international grid infrastructures. As computing evolves for the High-Luminosity LHC (HL-LHC) era, WLCG faces a rapidly growing computing demand alongside increasingly heterogeneous hardware environments (CPUs, GPUs, Cloud, and HPC).
To guide procurement, resource pledging, performance studies and accounting, the High Energy Physics (HEP) community relies on domain-specific benchmarking tools to accurately measure and compare computing resources.
As a Graduate / Applied Engineer (QUEST), you will join our team in CERN's IT Department to support the development, operation, and user support for WLCG HEPScore benchmarking services. You will contribute to key activities, helping to build and maintain the next-generation benchmarking software stack for the WLCG community.
Your responsibilities
Software Development: Actively contribute to the development, packaging, and distribution of the HEP benchmark software stack. Integrate, profile, and validate production-level GPU application workloads from LHC experiment software (e.g., event generation, simulation, fast inference, and HLT).
Containerisation: Build, test, and maintain container images of HEP workloads using automated GitLab CI/CD pipelines.
Service Maintenance & Enhancement: Ensure the smooth operation of the core benchmarking infrastructure and data ingestion services using Puppet, GitLab CI/CD, ActiveMQ, Logstash, OpenSearch, Kibana, Apache Spark and Grafana.
Performance Analysis: Collect, process, and analyze grid-wide benchmark data to evaluate performance scaling across diverse CPU/GPU architectures and server configurations.
Technical Support: Provide technical support and documentation to the WLCG user community.
Your profile
Advanced Python Development: Solid experience in Python programming,including modular code structure,packaging, distribution, and testing.
Containerisation: Solid understanding of Docker and Apptainer/Singularity, along with familiarity with GPU container execution runtime environments.
DevOps & Infrastructure Automation: Proficiency with GitLab CI/CD pipelines, container registries, configuration management (Puppet), and deployment methodologies.
Linux Systems & Performance Profiling: Strong Linux administration and shell scripting skills. Experience in system metrics tracking (e.g., CPU/GPU utilisation, hardware performance metrics, power monitoring).
Data Analytics Pipelines & Visualisation: Experience with monitoring and data analytics stacks such as OpenSearch/Elasticsearch, Logstash, Kibana, Grafana, or Apache Spark.
Communication & Collaboration: Good technical writing and communication skills to interact with domain experts across the global WLCG collaboration.
Skills
Programming & Scripting: Python (code, packaging, testing), Bash, Git.
Containerisation: Docker, Apptainer/Singularity.
GPU Computing: NVIDIA Container Toolkit, CUDA, nvidia-smi, AMD ROCm, rocm-smi, GPU profiling.
DevOps: GitLab CI/CD, Puppet, Linux administration.
Monitoring & Data Analytics: OpenSearch/Elasticsearch, Logstash, Kibana, Grafana, Apache Spark, ActiveMQ.
Spoken and written English, with a commitment to learn French.
Eligibility criteria:
You are a national of a CERN Member or Associate Member State.
You have a professional background in Applied Physics, Software Engineering (or a related field) and have either:
a Master's degree with 2 to 6 years of post-graduation professional experience;
or a PhD with no more than 3 years of post-graduation professional experience.
You have never had a CERN fellow or graduate contract before.

Skills
Python, Git, Linux, Docker, CI/CD, Elasticsearch, Data Analysis, Apache Spark, Technical Writing, Accounting

Employment type
Full-time

Work arrangement
No