Tharindu Madusanka recently completed his PhD at the University of Manchester and is now acting as an Applied Scientist at Amazon. He is part of Amazon’s AGI Audio Foundational Modelling team, which is responsible for developing advanced audio models, including the Nova Sonic model class.

In his doctoral research, Tharindu explores the formal reasoning capabilities of transformer-based language models under the supervision of Dr. Ian Pratt-Hartmann and Dr. Riza Batista-Navarro. During his PhD, he also completed a research internship at Amazon. Prior to commencing his doctoral studies, he held positions as a Research Scientist at the University of Moratuwa and as a Data Scientist at Turing.

Tharindu's research has been featured at leading AI conferences, where it has received notable recognition, including the Best Paper Award at ACL and the Outstanding Paper Award at EACL.