Mridul Kumar, Ph.D | Postdoc
Background: I hold a Bachelor's degree in both Mathematics and Physics, and a Master's degree in Physics with a specialization in Electronics. During my Master’s program, I was introduced to data-driven techniques such as Machine Learning and Deep Learning, which sparked a deep interest in the field. To apply these techniques effectively, I taught myself Python programming, which proved instrumental in my Master's thesis. In this work, I developed a model to forecast the probability of pest infestation in crop fields using machine learning approaches.
Building upon this foundation, my doctoral research further explored plant stress detection in agricultural settings. I developed a mathematical model to represent how plants absorb nutrients from the soil over time and how this absorption is affected under stress conditions. To enable real-time stress detection, I applied anomaly detection algorithms. This research led to a granted patent and enabled me to design and implement custom systems for data acquisition, analysis, visualization, and machine learning-based predictions.
Alongside my research, I have served in multiple faculty roles, teaching undergraduate-level courses in subjects including Data Visualization and Management with Python, Machine Learning, Deep Learning, and Applied Physics. I have nearly four years of teaching experience.
As a side project, I developed CFUcalc, a mobile application for automated bacterial colony counting on petri dishes. This app has been copyrighted and is actively used at my former institution. It is available on the Google Play Store as well.
Currently, I am a postdoctoral researcher at Ben-Gurion University of the Negev, where I have shifted my focus to applying data-driven methodologies to 4D Scanning Transmission Electron Microscopy (4DSTEM). My goal is to enhance material property characterization using advanced machine learning techniques.
Scientific interests : My main interest lays in the area of designing and studying novel nano-materials towards applications, urgent to solve problems facing nowadays humanity. I focus on two-dimensional materials, especially MXenes, as they provide wide room for the property tuning. I am fascinated about how the same material can be promising for energy harvesting and elector-chemical storage, molecular-level sensing, environmental remediation, biomedical applications, etc. Another area of interest for me is how MXenes can be combined with other nano-materials (for example, magnetic and plasmon nano-particles) to form multi-functional nano-composites. I enjoy working in a lab, performing both synthesis and complex characterization of materials, I am keen of finding new approaches and optimizing the existing ones to rationalize the output of my experiments. Recently I also strated studying modeling techniques and machine learning methods in order to proceed to computer-assisted materials screening and development of materials with per-assigned features. My research interests lie at the intersection of instrumentation and systems design, data-driven scientific discovery, mathematical modelling of physical systems, and the application of machine learning to scientific imaging and sensing.
Hobbies: Outside my academic and research pursuits, I enjoy exploring a wide range of non-fiction books and audiobooks, often centered around science, technology, and human behavior. I'm also a fan of cinema and television, with a particular passion for science fiction and action—Interstellar remains one of my all-time favorites. In my leisure time, I’m an avid gamer; titles like God of War (2018 and Ragnarök), Red Dead Redemption 2, and the Spider-Man series are among my favorites for their immersive storytelling and gameplay.
I also like to play Cricket, Badminton, and Chess.
External Links: