You will develop and operate CERN’s next-generation machine learning platform, enabling scientists and engineers to train, deploy and manage AI workloads at scale. Working with cutting-edge technologies including Kubernetes, GPUs and cloud-native tools, you will help build the infrastructure supporting machine learning for the High-Luminosity LHC.

Your responsibilities

• Develop and operate the NGT machine learning platform, covering model training, hyper-parameter optimisation and serving, targeting optimal usage of detector resources.
• Manage cluster resources, integrate new resources, on-premises or via external resources (such as clouds or HPC centers), and optimise resource usage (e.g. via CPU/GPU/RAM affinity layouts) and the existing MLOps workflows to cover deployment, observability and lifecycle management & versioning of models.
• Extend the cloud native infrastructure to support advanced workload scheduling, system monitoring & accounting, as well as prompt cost control.
• Produce and present reference architectures, documentation and best practices to support CERN users and use cases.
• Ensure efficient collaboration with other groups in `IT and in other departments, as well as constant alignment with the evolving landscape of external research and industry organisations.