Nuclear structure, end to end
Connecting nuclear theory to data with Bayesian emulators, and measuring short-lived isotopes by laser spectroscopy at FRIB.
Physicist · MIT · García-Ruiz Lab
Jose M.
Munoz
I am a physicist at MIT working on the nuclear many-body problem from both ends: the statistics and computation that connect it to data, and the precision experiments that test it. I like hard problems wherever they live.
About
A featherless biped, curious by nature and trained as a scientist. I like asking questions and taking apart hard problems.
I am a Physics PhD student at MIT, in the Laboratory of Exotic Molecules and Atoms. I work on the nuclear many-body problem from both ends. On one side I build the computational and statistical methods that connect nuclear theory to data. On the other I take precision laser-spectroscopy measurements of short-lived isotopes at FRIB.
My path has crossed a few fields. Before nuclear physics I worked on particle physics with the CMS experiment at CERN, neutrino phenomenology, econometrics, and more recently generative models for finance. Statistics is the thread that ties them together.
01
What I’m working on
Inference for hard problems
Bayesian inference, emulation, and interpretable models that make large many-body calculations tractable and worth trusting.
Structure in messy systems
Complex systems and emergent phenomena across fields, from particle physics to generative models in finance.
02
Selected work
Featured
Global Framework for Emulation of Nuclear Calculations
A hierarchical Bayesian neural network (BANNANE) that emulates ab initio nuclear calculations across the chart, predicting energies and charge radii together with calibrated uncertainties.
Physical Review Letters 136, 082501 · 2026↗
Nuclear Charge Radii of Aluminium Isotopes at the Proton Drip Line
arXiv:2605.09139 · 2026
Linking Electromagnetic Moments to Nuclear Interactions with a Global Physics-Driven Machine-Learning Emulator
arXiv:2603.26905 · 2026
Discovering Nuclear Models from Symbolic Machine Learning
Communications Physics 8, 101 · 2025
A General Framework for Equivariant Neural Networks on Reductive Lie Groups
NeurIPS 36 (2023) · 2023