Research & Academic Work
I am drawn to problems that sit at the boundary between mathematical rigour and empirical discovery — where physics, statistics, and machine learning meet.
Research Interests
The domains I am actively reading, building in, and hoping to contribute to.
Physics-Informed ML
Embedding physical laws as inductive biases in neural architectures — PINNs, symmetry-aware models, and conservation law constraints.
Computer Vision
Object detection, image segmentation, and visual representation learning with applications in scientific imaging.
Natural Language Processing
Information retrieval, semantic similarity, and document analysis — applied through projects like PlagCheck.
Data Science & Statistics
Statistical learning theory, uncertainty quantification, and rigorous experimental analysis with real-world datasets.
Computational Physics
Monte Carlo simulation, numerical methods, and detector modelling as used in high-energy physics experiments.
ML for Science
Applying modern ML methods to accelerate scientific discovery — from parameter estimation to surrogate modelling.
Academic Activity
Programs applied to, collaborations, and academic engagements.
SOKENDAI KEK Tsukuba / J-PARC Summer Student Program
KEK High Energy Accelerator Research Organization, Japan · 2026
Applied for the prestigious summer student program at KEK Tsukuba and J-PARC, Japan's leading particle physics facilities. Program involves hands-on research in experimental high-energy physics. Despite a strong application, I was not selected for the program, but it provided valuable experience in preparing research proposals and understanding the competitive nature of international research opportunities.
Learn moreSoftec 2026 — NU FAST
FAST National University of Computer and Emerging Sciences · 2026
Participated in Softec 2026, one of Pakistan's largest technology and computing competitions hosted by FAST-NUCES.
Learn morePaper Reading List
Papers I have read, am reading, or return to frequently — signals of genuine engagement, not a CV line.
Physics-informed neural networks: A deep learning framework for solving forward and inverse problems involving nonlinear partial differential equations
Raissi, Perdikaris & Karniadakis — Journal of Computational Physics, 2019
Foundational paper on PINNs — directly motivates my interest in physics-constrained ML.
An Introduction to Monte Carlo Methods in Statistical Physics
Newman & Barkema — Oxford University Press
Grounding in the statistical foundations used in my radioactive decay simulation.
Attention Is All You Need
Vaswani et al. — NeurIPS, 2017
Core transformer architecture — essential reading for understanding modern NLP and vision models.
Machine Learning for High Energy Physics
Radovic et al. — Nature, 2018
Survey of ML applications at particle physics experiments — directly relevant to J-PARC/KEK work.
A Statistical Learning Theory Perspective on Generalization in Deep Learning
Bartlett, Montanari & Rakhlin — Proceedings of the National Academy of Sciences, 2021
Understanding why deep nets generalise — connecting theory to empirical practice.