Menu
User documentation
Publications|Stories|2026

Alps Empowers Scientists: Ana Klimovic

They design AI systems that make the most of our world-class supercomputer. They see the Alps as a platform for collaboration and a foundation for sovereign, independent AI. They use it as a mathematical microscope to explore the complexity of molecular interactions and to make quantum effects visible. They are our users. In this series, we introduce some of the people behind these projects and explore how they are using Alps to advance science and benefit society.

July 27, 2026 - by CSCS

Name
Ana Klimovic 

Position
Associate Professor in the Department of Computer Science at ETH Zurich. 

Area of research
Computer systems, cloud computing, systems for AI 

My focus
My research explores how to design high-performance and energy-efficient software systems for large-scale applications like AI and cloud data analytics. I am interested in designing declarative software interfaces that abstract the underlying hardware infrastructure, making the cloud easy to use while enabling the computing platform to optimize energy efficiency and performance under the hood.

    “My goal is to design software systems that make the most out of this cutting-edge hardware” - Ana Klimovic. (Image: Alessandro Della Bella, CSCS)

    What supercomputing and Alps mean for me
    Supercomputing facilities like Alps enable researchers across a variety of fields to make new discoveries and advance science. My goal is to design software systems that enable researchers to make the most out of this cutting-edge hardware by efficiently scheduling jobs and making them run as fast as possible. 

    What challenges do I face
    Software like the operating system and cluster scheduling framework sits between user applications (e.g., AI model training) and the hardware (e.g., GPU cluster). Applications and hardware are both evolving rapidly, putting pressure on the system software layer in between to address the new needs of applications while making best use of the new features that modern hardware offers. 

    What I like most about my work
    I enjoy working at the cutting edge of computing technology and building systems that have real impact. I find it fascinating and rewarding to analyse where inefficiencies in current software/hardware systems come from and work together with my students and collaborators on solutions to improve efficiency. 

    My favourite recent project using Alps
    We are using the Alps supercomputer to train and serve large-scale fully transparent and open-source AI models in the Swiss AI initiative, including the Apertus multi-lingual large language model (LLM). I am especially excited about our elastic, distributed inference service, which gives researchers access to state-of-the-art open models while we take care of optimizing GPU scheduling, request batching, and other system optimizations to maximize performance and energy efficiency. We are in the process of extending the inference service to a model fine-tuning service, so that researchers can customize models for specialized tasks. We are also working together with CSCS to make this inference and fine-tuning service available to Swiss companies who are exploring sovereign AI deployments for innovation projects. 

    Career background 

    • 2026  – Present Associate Professor in the Department of Computer Science at ETH Zurich
    • 2020 – 2026 Assistant Professor in Computer Science at ETH Zurich 
    • 2019 – 2020 Research Scientist at Google Brain, Mountain View 
    • 2013 – 2019 PhD in Electrical Engineering at Stanford University 
    • 2013 – 2015 Masters in Electrical Engineering at Stanford University 
    • 2009 – 2013 Bachelors in Applied Science and Engineering in Engineering Science at the University of Toronto