// computational researcher · physics × quantitative finance
I build mathematical models of complex, uncertain systems — and validate them against noisy, real-world data.
A NASA Fellow and PhD candidate probing the equation of state of neutron stars, applying the same statistical toolkit to quantitative finance.
Jump to Astrophysics Quantitative finance
I'm a NASA Fellow and PhD candidate in Astrophysics at Southern Methodist University, advised by Dr. Fred Olness and still collaborating with Dr. William Newton at East Texas A&M. I bring 5+ years of Bayesian/MCMC modeling, Monte Carlo simulation, and Python data pipelines built on large, noisy, multi-source datasets.
My research infers neutron star structure from multi-messenger data — fusing heterogeneous streams, stress-testing models under uncertainty, and turning complex statistical output into insight. I've published in Physical Review Letters and Physics Letters B, trained at MIT's National Nuclear Physics Summer School, and built an equation-of-state pipeline now part of the NSF-funded MUSES project.
The same toolkit drives my quantitative-finance work with Traders@SMU's Quantitative Strategies group — market microstructure, backtesting, and risk-aware strategy. And as an adjunct professor at SMU, I teach honors physics built around creative, real-world themes.
Two parallel tracks, one methodology: build a principled model, validate it without mercy, and let the data overturn it. Switch between them below.
I'm drawn to just about everything more than 200 miles above sea level — my cheeky shorthand for “legally, space.” In practice, that's dense-matter nuclear astrophysics: connecting laboratory nuclear physics to the interiors of neutron stars. I'm fascinated by combining scales — and problems — that rarely meet: quantum objects on classical scales, which takes a problem-solving intuition that spans all of physics. My focus has evolved a good deal — from undergraduate neutron-skin work to the machine-learning and cooling projects I'm on now.
Concurrent projects sit side by side — select any node to expand it.
Systematic quantitative research: kernel methods, regime-aware features, and honest out-of-sample validation for options and cross-asset markets. The physics carries over directly:
End-of-day options trading pipeline — signal ingestion, feature engineering, trade selection, and IBKR execution — with a 10k-simulation block-bootstrap backtester, regime-aware features, and disciplined position sizing.
A Lorentz-invariant kernel for RKHS models, capturing scale-dependent nonlinear interactions in financial time series — physics intuition, transplanted.
A seven-model systematic framework (trend, mean-reversion, dealer-flow, cross-asset carry, long-vol, equity factors, inflation-regime) under TSMOM-grade validation: walk-forward, stationary block bootstrap, look-ahead audits, and pre-registered acceptance bars. Three models paper-deployed on IBKR with a Bayesian meta-allocation layer for volatility and correlation budgeting.
Traders@SMU is SMU's premier trading and markets club — structured like a small multi-strategy hedge fund. In its Quantitative Strategies sector I build systematic, model-driven research in Python; I joined as an analyst, promoted to Co-Sector Head within a single semester, and built the team's full backtesting infrastructure.
I help teach members quantitative strategy and book development hands-on: small, shippable projects on real data that build toward research-grade models. This year's work runs on two tracks.
A custom Reproducing Kernel Hilbert Space (RKHS) model that finds the non-linear, regime-dependent structure linear factor models (CAPM, Fama–French) miss: the same instrument can trend in one volatility regime and mean-revert in another. On crude-oil futures, it blends fast microstructure signals — order-book shape, order-flow toxicity, price impact, options hedging flows — with slower ones from the volatility risk premium, cross-asset momentum, and news sentiment, staying tractable across tens of millions of trades.
What finally worked came from several method changes I pushed: two-stage estimation, a Kalman filter that treats the kernels as regime detectors rather than return predictors, and inverse-volatility position sizing. Everything runs through out-of-sample validation — purged walk-forward, block-bootstrap Monte Carlo, and historical stress scenarios — to guard against overfitting and look-ahead.
This semester
The on-ramp. The RKHS model is a high barrier to entry, so new analysts start here — learning options pricing, volatility, and microstructure through small, shippable projects pointed at a real question: how and why emerging markets misprice risk relative to the US, where inefficient or crowded retail flow can stretch prices in ways you don't see at home. It's also a feeder — vol-surface, factor, and microstructure work maps directly onto the RKHS model's Kslow and KGEX thinking, so strong analysts graduate into Track 1.
Goals
Actively exploring quantitative research, data analytics, and risk-modeling roles in finance — particularly private equity, hedge funds, and investment banking.
As an adjunct professor at SMU, I teach introductory honors physics — building real physical intuition, and a bridge from the classroom into research.
I use creative, real-world themes to build genuine physical intuition — not just problem-solving reflexes. In one lab, each group got the raw materials for a working headphone — paper cups, magnets, and wire — and no instructions; the first team to build one and name the song on my laptop won Starbucks gift cards. Over two semesters I've doubled the course's enrollment — a real win for the department.
Each semester, students also build toward a real research poster on the course's theme and present it to the department and the school's deans — a genuine research capstone inside an intro class.
To deepen that pull, I bring graduate students and faculty in to present their research, showing students what physics looks like beyond the textbook — and drawing more of them toward both research and further coursework. And I remind them: you don't have to be a physics major to do physics research — it's a way to build skills and strengthen a CV.
In my first semester of grad school, the undergraduate quantum-mechanics professor fell ill and the department asked me to take over the course — my first time leading a university classroom.
Physical Review Letters 130, 112701
Much effort is devoted to measuring the nuclear symmetry energy through neutron star (NS) and nuclear observables. Since matter in the NS core may be non-hadronic, observables like radii and tidal deformability may not provide reliable constraints on properties of nucleonic matter. We demonstrate that coincident timing of a resonant shattering flare (RSF) and gravitational wave signal during binary NS inspiral probes the crust–core transition region and provides constraints on the symmetry energy comparable to terrestrial nuclear experiments. We show that nuclear masses, RSFs and measurements of NS radii and tidal deformabilities constrain different density ranges of the EOS, providing complementary probes.
Physics Letters B 834, 137481
We present the first Bayesian inference of neutron star crust properties to incorporate neutron skin data, including the recent PREX measurement of the neutron skin of 208Pb, combined with recent chiral effective field theory predictions of pure neutron matter with statistical errors. Using a compressible liquid drop model with an extended Skyrme energy-density functional, we obtain the most stringent constraints to date on the transition pressure Pcc = 0.33 ± 0.07 MeV fm−3 and chemical potential μcc = 12.6 (+1.8/−1.9) MeV (which control the mass, moment of inertia and thickness of a neutron star crust), the proton fractions that bracket the pasta phases yp = 0.115 (+0.016/−0.017) and ycc = 0.041 (+0.007/−0.006), as well as the relative mass and moment of inertia ΔMp/ΔMc ≈ ΔIp/ΔIc = 0.54 (+0.05/−0.09) and thickness ΔRp/ΔRc = 0.129 (+0.019/−0.030) of the layers of non-spherical nuclei (nuclear pasta) in the crust.
Southern Methodist University
Advisor: Dr. Fred Olness
Nuclear astrophysics: inferring neutron-star structure and cooling from dense-matter physics, and building machine-learning methods for nuclear cross-sections with the SURGE collaboration.
Texas A&M University–Commerce (now East Texas A&M)
Advisor: Dr. William G. Newton · Thesis: Inference of whole neutron star EOS from combined astrophysical and nuclear observables
Nuclear theory and astrophysics: constraining the equation of state and nuclear symmetry energy by combining neutron-star observations with laboratory nuclear data under Bayesian inference.
Texas A&M University–Commerce · Magna Cum Laude
Advisor: Dr. Kent Montgomery
Early research on the nuclear symmetry energy and neutron-star crust — where the dense-matter work I still do began.
Collin College
A foundation in visual art and design — the creative roots of how I teach and communicate science.
Quantum chromodynamics, parton distributions, and electroweak collider phenomenology.
High-energy frontier theory — effective field theory and beyond-the-Standard-Model physics.
A selective U.S. school in nuclear theory and experiment — nuclear structure, QCD, and nuclear astrophysics.
France's international school on nuclear physics and astrophysics, named for Frédéric & Irène Joliot-Curie.
Open to collaboration in nuclear and dense-matter astrophysics, and to teaching and research conversations. I'm also actively seeking finance internships and full-time roles — quant research and data science are my strengths, but I'm genuinely open across finance. Email's the best way to reach me.