Postdoctoral Researcher · University of the Witwatersrand, Johannesburg, SA · ATLAS Collaboration, CERN
I am currently working as a postdoctoral researcher at the University of the Witwatersrand,
Johannesburg, South Africa. I work on the ATLAS experiment at CERN. My current research
focuses on Dark Sector searches and the Calorimeter Fast Simulation pipeline for the ATLAS detector.
I obtained my Ph.D. in Experimental High Energy Physics from IISER Pune, India, under the supervision
of Prof. Sourabh Dube. My doctoral
research focused on searches for physics beyond the Standard Model (BSM) in
multilepton final states using at the CMS experiment at CERN.
In addition to physics analyses, I contributed to the development and improvement
of CMS tracking software in preparation for the High-Luminosity LHC upgrade.
Outside physics, I am an amateur photographer, a keen motorbike rider, and a
passionate football fan of Argentina and Barcelona.
Arnab Laha · Wits University
Click any image to enlarge
Searches for BSM phenomena in multilepton final states, model-independently and/or targeting specific models: Vector-like leptons, seesaw mechanisms, and leptoquarks. Machine learning to enhance search sensitivity. Currently working on searches exploring Dark Sector models in the ATLAS experiment. Designing a cut-based strategy for emerging jets search using Full Run 3 ATLAS dataset.
Implementation of the Phase 2 tracker geometry and tracking in the CMS Fast Simulation package for the HL-LHC upgrade, enabling faster Monte Carlo event generation for HL-LHC physics studies.
Working on ATLAS Fast Calorimeter simulation software (AF3) to implement fast reconstruction of long-lived particles in the ATLAS calorimeter.
Investigating hidden patterns in multidimensional HEP data using PCA, UMAP, and Autoencoders for lepton classification on CERN Open Data. Also uses GANs to speed up W+jets and Z+jets event simulation at LHC.
Reviewing heavy flavor tagging usage in CMS EXOTICA analyses and supporting analyzers with b-tagging tool implementation in liaison with the CMS b-tagging object group. Over 40 analyses reviewed.