I am a postdoctoral researcher at the University of Zurich with Prof. Dr. Nicola Serra. Being from computer science background, I conduct research at the intersection of particle physics and machine learning. Currently working on using generative modles for faster simulation of physics experiments. And second, working on machine learning for the design of physics experiments and machines.

In my latest work, I presented a case for the use of reinforcement learning for physics instrument design. Find the preprint here (sorry, arxiv delays):

PDF: Physics Instrument Design with Reinforcement Learning

Besides that, I am also working on machine learning in the field of epidemiology.

Before that, I did my Ph.D. at CERN as a doctoral student and was advised by Dr. Maurizio Pierini and Dr. Jan Kieseler. In my Ph.D, I worked on graph neural networks (GNNs) for multi-particle reconstruction. Utilizing point-cloud based dynamic GNNs, I proposed the first-ever machine learning based end-to-end reconstruction method. While I focused on the High Granularity Calorimeter (HGCAL) at the Compact Muon Solenoid (CMS) experiment, the techniques studied were general and now are also being used for other reconstruction tasks such as track reconstruction.

Please consult my Google scholar profile for the rest of my work.