Prakash Verma

I am a data scientist in training, most of my work involved developing and benchmarking computational tools to compute molecular properties accurately and efficiently. My interest in data science grew stronger when I realized that my biggest contribution to science was not due to fancy mathematical derivation or efficient programming but it was because of finding hidden patterns in data of electron behavior. Using simple linear regression, I was able to construct an effective one-electron potential that provides an excellent approximation to correlated electrons. Some of my notable contributions in computational chemistry: Developed analytical gradient method for non-variational density functional theory. Developed scalar relativistic module in PSI4 using C++ and Python to understand the role relativistic effects have on NEXAFS. Extended the gold-standard method of quantum chemistry (i.e. coupled cluster) to calculate electrical and magnetic properties of larger molecules by implementing the massively parallel linear density response module in ACES3 using Fortran 77 and domain-specific language SIP/ SIAl. Developed computational tools in NWchem to calculate magnetic properties (Electron spin resonance parameter viz. A- and g- tensor) of heavy elements using 2-component quasi-relativistic ZORA-DFT approaches using global array toolkit and Fortran 77. My Google Scholar page can be found at http://bit.ly/2sHK3C3