I develop numerical frameworks for quantum field theories and complex systems—connecting rigorous physics with scalable, uncertainty-aware computation.
I develop high-precision multiscale methods for parameter estimation in large stochastic field models. Treating gradient flow as explicit renormalization-group evolution enables continuous β-function frameworks and robust extrapolations with controlled uncertainty.
I build scalable simulation and analysis code for non-perturbative lattice formulations, validate it against analytic benchmarks, and generate large ensembles for propagator measurements across parameter space.
Convolutional networks identify regime changes without direct access to governing equations. Histogram reweighting improves prediction and transfer learning tests generalization across related interacting systems.
I modeled hard boundary constraints with MIT Bag–type conditions, reproduced continuum reference values, and resolved apparent universality violations in unstable discretisations through numerical regularisation.