Big Techday 25: Proxima Fusion's ConStellaration challenge - Veronika Siska & Santiago Cadena

Proxima Fusion's ConStellaration challenge: Datasets to drive the future of fusion energy

Europe's fastest-growing fusion company, Proxima Fusion, is working to build the world's first stellarator fusion power plant, delivering virtually limitless clean energy. Getting there requires tackling one of the world’s most complex technical challenges: Quasi-isodynamic (QI) stellarator optimization. Proxima believes machine learning has an outsized role to play in solving that challenge, which is why they've teamed up with Hugging Face to launch a collaborative challenge inviting the global machine learning community to help solve three different stellarator design problems. Proxima engineers Santiago Cadena and Veronika Siska present an open dataset of diverse QI-like stellarator plasma boundary shapes and introduce three optimization benchmarks of increasing complexity to show how learned models trained on Proxima's dataset can lower the entry barrier for optimization and machine learning researchers to engage in stellarator design and to accelerate cross-disciplinary progress toward bringing commercial fusion energy to the grid.


About the speakers:
Dr. Veronika Siska: Dr. Veronika Siska is a Research Software Engineer at Proxima Fusion, where she builds data pipelines and full-stack systems to support simulation and machine learning for stellarator optimization. She previously worked at TNG Technology Consulting on large-scale data-driven solutions across industries, helping to establish the company’s Hungarian branch, and at the Austrian Institute of Technology on data sharing technologies for public safety and climate applications. Veronika studied Physics and Applied Mathematics before completing a PhD in Computational Biology at the University of Cambridge, and her career spans data science, software engineering, and applied research. She enjoys bridging the gap between research and application by turning innovative methods and prototypes into reliable, production-ready systems.

Dr. Santiago Cadena: Dr. Santiago Cadena is Machine Learning Researcher and Engineer at Proxima Fusion, where he works on data-driven approaches for stellarator optimization. Previously, he has applied Machine Learning to real-world products around mapmaking at Lyft, and neuromotor interfaces at Meta. He conducted graduate studies in Computational Neuroscience and Machine Learning at the University of Tübingen and the Max Planck Institute for Intelligent Systems in Germany where he used deep learning modeling and large‑scale single‑cell neural recordings to study visual processing in the brain. He holds bachelors degrees in Electrical and Biomedical engineering from Universidad de los Andes. Passionate about cross‑disciplinary innovation, he thrives on applying AI/ML to novel scientific and engineering challenges.