Harikrishna Sahu
Research Scientist II, Georgia Tech, USA
Dr. Sahu joined Georgia Tech as a postdoc in September 2019. Prior to joining Georgia Tech, he was a postdoc fellow at Nanjing University, China. He received his M.Sc. degree in Chemistry at National Institute of Technology Rourkela and a Ph.D. degree in computational chemistry at Indian Institute of Technology Guwahati. His research focuses on developing and applying computational and machine learning tools to accelerate materials discovery.
Notable Contributions
My research mostly contributes to the development of organic materials for next-generation electronic devices.
-
polyT5: A Polymer-Focused Language Model
polyT5 is a polymer-specific encoder–decoder language model designed to understand polymer chemistry, predict key properties, and generate new polymers tailored to targeted performance. It enables direct design of synthesizable polymers without exhaustive enumeration. Applied to dielectric polymer discovery, polyT5 proposed over 20,000 promising candidates, one of which was experimentally validated, establishing a fast AI-driven pathway for accelerated polymer innovation.
-
Polymer Structure Predictor
PSP is an autonomous polymer model generator, to build a hierarchy of polymer models, ranging from oligomers to infinite chains to crystals to amorphous models, using simplified molecular-input line-entry system (SMILES) strings of polymers. Being a first of its kind, it will facilitate automation in polymer property prediction and design.
-
Organic Solar Cell
High performing machine learning models are developed to predict the PCE, Voc, Jsc, and FF, using relevant quantum chemical properties of donor/acceptor materials in organic solar cells. This study demonstrates how to design organic materials for a specific photovoltaic application.
Ramprasad Group
Currently, I am working as a Research Scientist in the Ramprasad group, Georgia Tech, USA. Professor Ramprasad is an eminent scientist in the field of polymer informatics. Ramprasad Group