PhD Research
PhD Research
Xanadu
Xanadu
Xanadu
This project explores how quantum photonic processors can accelerate flood modeling at high resolution, a task that traditionally requires massive HPC clusters. By offloading linear subproblems of the Shallow Water Equations to Xanadu’s photonic quantum hardware, the system demonstrates how hybrid quantum–classical methods can support environmental simulations.
Overview
Overview
I developed a Django-based prototype that generates synthetic terrains, runs a classical shallow-water solver, and optionally calls a quantum routine via PennyLane and Strawberry Fields. Results showed that the quantum-assisted solver converges to classical benchmarks for small systems, with accuracy maintained within a 5% error margin. Although current quantum devices remain limited in scale, this work demonstrates a pathway to more scalable, real-time flood forecasting as photonic hardware advances.
The project highlights how near-term quantum devices can integrate into scientific modeling pipelines, offering early proof-of-concept for hybrid workflows that combine HPC and quantum resources. A web-based dashboard presents simulation outputs in interactive 3D, enabling users to toggle between classical and quantum solver modes and compare performance side-by-side.
DrMaz.ai
Overview
I developed a Django-based prototype that generates synthetic terrains, runs a classical shallow-water solver, and optionally calls a quantum routine via PennyLane and Strawberry Fields. Results showed that the quantum-assisted solver converges to classical benchmarks for small systems, with accuracy maintained within a 5% error margin. Although current quantum devices remain limited in scale, this work demonstrates a pathway to more scalable, real-time flood forecasting as photonic hardware advances.
The project highlights how near-term quantum devices can integrate into scientific modeling pipelines, offering early proof-of-concept for hybrid workflows that combine HPC and quantum resources. A web-based dashboard presents simulation outputs in interactive 3D, enabling users to toggle between classical and quantum solver modes and compare performance side-by-side.