About
A physicist, mostly.
I work on light, geometry, and the numerical habits that keep the two honest. Currently an MSc student at Western University, after a BSc at Xiamen University Malaysia.
I am a physics graduate of Xiamen University Malaysia, now pursuing an MSc at Western University in London, Ontario. My work sits where transformation optics meets the bench: geodesic lenses, conformal mapping, and single-photon interference.
The through-line is an old idea — that a medium can be described by a geometry — but I am interested in the places where the equivalence becomes a practical constraint. A surface that exists in a metric may not exist as a machined object. A cloak that is exact in the ray limit may be useless at the bandwidth you need. The gap between those statements is where the work is.
How I work
Derive it, then integrate it.
A closed-form result is the best check on a numerical one. If the two disagree, the disagreement is the result.
Make the figure reproducible.
Every plot on this site has parameters, a model and, where it matters, code. A figure that cannot be regenerated is a slide, not evidence.
State the assumption that could break.
Geometric optics, scalar fields, exact rotational symmetry. The interesting failure is usually the one the model was designed to ignore.
This site
The site is hand-written HTML, CSS and a little JavaScript. No framework, no analytics, no trackers. The typefaces — Newsreader and IBM Plex — are self-hosted. The interactive figures on the research page are computed in the browser: the Luneburg rays are integrated with RK4, the cloak grid is a live coordinate map, and the single-photon pattern is sampled from the interference law.
The source is public at github.com/hello-zxs/zhengxs. If something is wrong — a derivation, a citation, a browser bug — I would rather hear about it than leave it.
Contact
For research conversations, corrections, or a question about the thesis, email is best. I am happy to send a PDF, a data file, or the code behind a figure.