About

Soumyadipta Sengupta is a Data Science Researcher at Shell, with over 4 years of relevant experience in the field. He has a PhD in Applied Physics from Eindhoven University of Technology, a Master of Science in Petroleum Engineering from Texas A&M University, and a Bachelor of Technology in Petroleum Engineering from IIT Dhanbad. At Shell, Soumyadipta has been involved in various projects, including devising and implementing techniques for run length estimation of certain units on an offshore asset using historical PI tag and lab time series data, improving forecast of an exogenous variable in a project using a boosting model, testing and analyzing results of an anomaly detection tool using probability theory, performing extreme value analysis to help decision making in corrosion inspections, researching Physics Inspired Neural Nets for foaming and comparing with conventional numerical methods, and leading a project related to Anomaly and trend detection in vibration and lube oil system data. Soumyadipta has also worked as a PhD Researcher at NWO (Dutch Research Council), where he studied the morphology and dynamics of polyelectrolyte membranes (PEM) used in fuel cells/flow batteries. He used atomistic (MD) and dissipative particle dynamics (DPD) simulations to simulate newer class of PEMs with variations in side chain groups, nanofiller content, and type. He analyzed trajectory data generated by LAMMPS using custom codes in MATLAB and Python, and published 4 first author papers in reputed journals. At Texas A&M University, Soumyadipta worked as a Graduate Research Assistant, where he worked on various data science topics, including non-parametric regression to predict petrophysical properties from log data, Kriging, Sequential Gaussian simulation, and Bayesian updation to include secondary data in constructing subsurface realizations. He also built a 3D reservoir simulator using streamline and finite difference techniques and optimized waterflooding using gradient-based techniques. Soumyadipta's technical skills include Python, Keras, Azure, Tensorflow, CNN, Deep Learning, ML, SQL, Seaborn, AI, RNN, Scikit, Pandas, Neural Networks, UI, Frontend, and Backend. He has also received recognition in the form of SRA award for his work as a technical lead on a project related to Anomaly and trend detection in vibration and lube oil system data.