Turning models into measurable impact.

I specialize in modeling, simulation, and optimization for real‑world systems. I enjoy moving from proof‑of‑concept to production: clean data pipelines, robust code, and clear business outcomes.

See work Download CV

Core: PDEs, inverse problems, numerical optimization
Tools: Python, C/C++, MATLAB, PyTorch/JAX
Industries: Energy, manufacturing, health tech

About

I'm an academic-turned-practitioner who likes shipping things. My research background is in [your subfields]. In industry settings I focus on creating models that are explainable, maintainable, and cost‑aware.

Recently: led a cross‑functional project reducing simulation runtime by 45% and enabling on‑device inference for field deployment.

Modeling Simulation Optimization ML Ops Numerics

Selected highlights

Cut cost by 30% via surrogate modeling

Built a physics‑informed surrogate that replaced a 6‑hour FEM solve with a 4‑minute approximation at < 2% error.

Python, PyTorch, FEM

Stabilized inverse problem in production

Regularized a noisy reconstruction pipeline; false positives dropped 18% while recall held steady.

Optimization, Bayesian methods

Real‑time control prototype

Designed an MPC demo on embedded hardware; achieved 50 Hz loop with latency budgeted for sensors.

Control, C++, Embedded

Projects

Project Name

Code

One‑sentence business outcome first. What changed? Who benefited? Then 1–2 lines of technical detail.

Impact metric (e.g., -20% cost, +12% throughput)
Stack: Python, JAX, CUDA

Project Name

Demo

Concise description focusing on constraints, tradeoffs, validation, and deployment details.

Benchmarks vs. baseline
Data, assumptions, failure modes

Publications (selected)

Lastname, Y. (2024). Title of Paper. Journal/Conference. paper · code
Lastname, Y. (2023). Another Title. Venue. paper

Tip: keep this short and link to a full list (Google Scholar/CV).

Contact

Best way to reach me is you@your-domain.tld. I'm also on LinkedIn and GitHub.