<aside>
Applied Sciences
Physicist & Machine Learning Researcher. University of Cambridge
</aside>
Sam Leeney is a physicist, machine learning researcher and entrepreneur whose work bridges precision cosmological inference, GPU-accelerated computation and machine learning. He is based at the University of Cambridge's Cavendish Laboratory, where his research focuses on detecting faint signals from the early universe using novel machine learning and Bayesian statistical methods. His doctoral work centres on 21cm cosmology, where he develops physics-informed neural networks and Bayesian frameworks to extract the signature of neutral hydrogen from the cosmic dawn, a period when the first stars and galaxies began to form. Sam has built a distinctive record of cross-domain impact across five years of applied machine learning research. His methods have transferred across four scientific domains: from 21cm cosmology and machine learning to transient science and medical physics. In transient astronomy, his work extends to the detection and classification of fast radio bursts, pulsars and Type Ia supernovae, applying Bayesian anomaly detection and GPU-native differentiable spectrophotometry to problems ranging from radio time-frequency data to supernova distance-ladder analyses relevant to the Hubble tension.
Alongside his research, Sam founded C3 (Cambridge Compute Company), a GPU compute marketplace and orchestration platform designed to address the compute access inequality holding back academic research. C3 has secured endorsements from the University of Cambridge, University of Oxford, MIT Media Lab and Imperial College London.
Within FTL, Sam leads external research initiatives - bringing deep expertise in scalable Bayesian computation, applied machine learning and research infrastructure, alongside direct experience translating frontier academic work into commercially viable platforms. His work reflects FTL's mission to bridge interdisciplinary research and industrial collaboration for the advancement of global science.
Cambridge Profile | LinkedIn | GitHub | Google Scholar | Cambridge Compute Company