Applied R&D
I am an applied mathematician and engineer specializing in data-driven simulation of complex physical systems, scientific software, and high-performance computing. My work combines reduced-order models, data assimilation, and machine learning with physics-based solvers to produce tools that are fast enough to deploy, interpretable enough to trust, and rigorous enough to use in environmental and energy settings.
Capabilities
The focus is on turning strong research into tools that can support engineering decisions, forecasting workflows, and operational modeling.
Simulation Systems
From finite element solvers to deployable low-dimensional approximations, I build simulation pipelines that preserve physical structure while reducing turnaround time enough for calibration, design, and forecasting.
Forecasting Pipelines
I use Kalman-type methods, variational ideas, and multi-fidelity algorithms to combine sparse observations with physical models and produce calibrated predictions under uncertainty.
Scientific Software
I treat software architecture, testing, profiling, documentation, and reproducibility as first-class work, not afterthoughts. That is what makes methods usable outside a paper.
Compute Strategy
My workflow includes MPI, GPU-aware design, distributed snapshot generation, and Slurm-based deployment, with a constant focus on throughput and maintainability together.
Current work
Current work focuses on scalable multiphysics and uncertainty pipelines that support environmental risk assessment and robust design decisions.
Selected Work
GridapROMs.jl
Built to make high-performance reduced models usable in real computational workflows, including nonlinear, transient, and multi-field systems.
MeteoModels.jl
Designed to support scalable forecasting workflows where observations, model error, and uncertainty all need to be handled without excessive complexity.
Highlights
Developing digital-twin and large-scale multiphysics frameworks for offshore floating solar systems, with forecasting and UQ workflows for reliable design and environmental risk assessment.
Built reduced-order methods, tensor compression workflows, unfitted finite element techniques, and scientific software for parameterized PDEs and computational fluid dynamics.
Worked on topology optimization for compliant aerospace mechanisms under manufacturing and stress constraints, connecting numerical methods to engineering requirements.
Notes
Project note
On why interfaces, tests, benchmarking, and documentation are not polish work but part of the technical result.
Modeling note
A short view on mismatch, regime coverage, observability, and the difference between compression and reliable prediction.
Forecasting note
How filtering pipelines become operational components when uncertainty needs to drive decisions rather than just estimates.
Practice
This section is meant for brief, high-signal notes about simulation systems, software architecture, and forecasting workflows.