Tarak Karmakar
IIT Delhi
Tarak Karmakar is an Associate Professor in the Department of Chemistry at the Indian Institute of Technology Delhi. He joined IIT Delhi as an Assistant Professor in June 2021, following a postdoctoral fellowship (2017– 2021) in the group of Prof. Michele Parrinello at ETH Zurich, USI Lugano, Switzerland, and IIT Genova, Italy. He obtained his Ph.D. from Jawaharlal Nehru Centre for Advanced Scientific Research (JNCASR), Bangalore, in 2016. His research group at IIT Delhi works on developing advanced simulation methodologies, machine learning interatomic potentials (MLIPs) for materials, AI-driven applications in chemistry, and multiscale modeling of nano–bio systems. He has been awarded the Early Career Researchers prize from Molecular Physics (Taylor & Francis) in 2022, received Teaching Excellence Award from IIT Delhi in 2024, and elected into the Indian Academy of Sciences Associateship with effect from the year 2023.
Session 2C: Lectures by Fellows/Associates
Chairperson: K. N. Raghavan, Krea University, Sri City
From atoms to algorithms: Computer simulations in the age of AI
Computer simulations have long served as essential tools for investigating the behaviour of molecules, materials, and biological systems at the microscopic and dynamic levels. Traditionally, computer simulations, particularly molecular dynamics simulations, have relied on well-defined physical models and rigorous numerical methods, with their scope often constrained by computational resources. Rapid development of artificial intelligence (AI) and advances in computer technology are now transforming the execution, acceleration, and conceptualization of simulations. AI is fundamentally reshaping computer simulations by enabling the modeling of molecular interactions in condensed phase systems with improved speed and accuracy, facilitating the exploration of rare events, allowing the calculation of thermodynamic and kinetic properties, and advancing the design of new materials and pharmaceuticals. The integration of physics-based models with data-driven approaches marks the beginning of a new era in which simulations function not only as analytical tools, but also as drivers of scientific and technological innovation.