Machine Learning Engineer with HPC background

The Swiss National Supercomputing Centre (CSCS) develops and operates cutting-edge, high-performance computing (HPC) systems as an essential service facility for science. The centre enables world-class research through its scientific user lab, which is available to domestic and international researchers in academia, industry, and the business sector. The centre is operated by ETH Zurich and has offices in Lugano (headquarters) and Zurich.

CSCS is finalizing the implementation of its new “Alps” research infrastructure. As a world-class HPC system, it will also cater to large scale Machine Learning (ML) users who require an optimally configured environment for performance and ease-of-use.

To expand our team and capabilities, CSCS has open positions at our offices in Lugano and Zurich for ML engineers with HPC backgrounds who will work at the intersection between ML, software, and system engineering.

Initially you will be involved in the following activities:

• Collaborative benchmarking of our new machine to optimize throughput for very large-scale ML training — this effort will not only yield specialized support, advisory, and debugging capabilities to serve our most demanding ML users, but it will also drive system and software engineering enhancements across the entire training stack;
• Contributing to the definition and implementation of dedicated service platforms to meet the specific needs of our ML users;
• Participating in multi-disciplinary collaborations dedicated to exploring data-centric innovations for addressing both emerging and established scientific challenges.

As an ML Engineer, you will play a crucial role in enabling large-scale ML projects to run efficiently on our infrastructure. At the same time, you will have the opportunity to join world-class, cross-functional teams in the execution of ML research projects.

Your responsibilities:

• Design and maintain our environment for optimal performance and ease-of-use for ML training projects;
• Provide expert support, guidance, and advice to our ML users, thus enabling them to extract maximum value from our cutting-edge infrastructure;
• Actively contribute to cross-functional collaborations involving data-centric innovations and large scale ML training;
• Envision, develop, and operate supporting services for common ML use cases;
• Engage with the ML community and stay up to date with the latest advancements in relevant technologies.

Expected qualifications:

• Master's degree in computer science or a related field, with a focus on ML;
• Experience in developing and managing complex ML projects in a scientific context;
• Extensive experience with Python and established deep learning libraries;
• Experience with distributed ML training involving large-scale jobs in HPC or cloud environments;
• Experience in performance tuning;
• Experience with collective communication libraries.

Experience with the following is preferred, though there will be ample opportunities to learn and gain experience on the job:

• CUDA/C/C++ programming;
• Natural language processing (NLP), in particular large language models (LLMs) and concepts, as well as training and fine-tuning with tooling ecosystems;
• Reproducible MLOps pipelines.

In an organization that values autonomy, ownership and continuous learning, the successful candidate will have the opportunity to gain further specialized hands-on experience through activities characterized by the challenging aspects typical of the HPC field, joining compelling world-class ML collaborations involving partner institutions, and operating on cutting-edge hardware.

• ETH Zurich is a family-friendly employer with excellent and flexible working conditions.
• You can look forward to an exciting working environment, cultural diversity and attractive offers and benefits.
• We value the diversity of our team and, to further enhance the diversity of our workforce, we particularly encourage women to apply.