My research develops high-fidelity, data-driven models of multiphase reactive systems relevant to energy and chemical processing. A central thread is gas-solid fluidization. I study fluidized bed hydrodynamics, heat and mass transfer, and drying using coupled CFD-DEM and Eulerian simulations, and validate them with experiments such as radioactive particle tracking (RPT). I am especially interested in scientific machine learning: physics-informed neural networks (PINNs), reduced-order models, and ML surrogates that can speed up expensive simulations while respecting the governing physics. My current work extends to reacting flows, including CO₂-to-chemicals conversion and large-eddy simulation of pool and tank fires for process safety. Across these areas, I aim to combine physics-based modelling, CFD, coupled CFD-DEM simulation for multiphase reactors, experiments, and AI/ML to build fast, reliable digital tools for designing, scaling up, and optimizing industrial reactors.