Carlos Xavier Hernández

About

I am a Senior Research Scientist at Meta Reality Labs, where I develop machine learning systems for neuromotor interfaces—recently leading development of gesture recognition models for the Meta Neural Band. Previously, I was an early member of the team at CTRL-labs, a neurotechnology start-up acquired by Meta in 2019.

Prior to joining industry, I earned a PhD from Stanford University as an NSF Graduate Research Fellow, working with Vijay Pande on statistical modeling of protein folding. My research interests span signal processing and representation learning—with applications from molecular dynamics to brain-computer interfaces.

Publications

Software

Photography

In my spare time, I take photos— mostly of wildlife and street scenes. Purchase prints or book a session →

📄 Download PDF

Carlos Xavier Hernández

Senior Research Scientist
🐙 github | 💼 linkedin | 🌐 website | ✉️ email

I am a Senior Research Scientist at Meta Reality Labs, working on machine learning to enable neuromotor interfaces. Prior to that, I worked with Vijay Pande at Stanford on statistical modeling of biomolecular dynamics.

Experience

Meta Platforms, Inc.

Senior Research Scientist · New York, NY, USA · 2019 – Present

CTRL-labs

Research Scientist · New York, NY, USA · 2018 – 2019

Stanford University

NSF Graduate Research Fellow · Stanford, CA, USA · 2013 – 2018

Education

Stanford University

Ph.D. in Biophysics · Stanford, CA, USA · 2013 – 2018
Advisor: Vijay Pande

Columbia University in the City of New York

B.S. in Applied Mathematics · New York, NY, USA · 2009 – 2013

Skills

Languages & Frameworks: Python, PyTorch, NumPy, SciPy, Pandas
Domains: Time-series modeling, signal processing (DSP), causal inference
Methods: Deep learning (RNNs, Transformers), Markov state models, information theory

Selected Publications

A Generic Non-Invasive Neuromotor Interface for Human-Computer Interaction

P Kaifosh, TR Reardon, and CTRL-Labs Nature · DOI · 2025

Variational Encoding of Complex Dynamics

CX Hernández*, HK Wayment-Steele*, MM Sultan*, BE Husic, and VS Pande Phys. Rev. E · DOI · 2018

Using Deep Learning for Segmentation and Counting within Microscopy Data

CX Hernández, MM Sultan, and VS Pande arXiv · arXiv · 2018

Selected Software

MDTraj: A Modern, Open Library for the Analysis of Molecular Dynamics Trajectories

RT McGibbon, KA Beauchamp, MP Harrigan, C Klein, JM Swails, CX Hernández, CR Schwantes, LP Wang, TJ Lane, and VS Pande Biophysical Journal · DOI · Python · ⭐ 683 🍴 290

VDE: Variational Dynamical Encoder for Complex Dynamics

CX Hernández, HK Wayment-Steele, MM Sultan, BE Husic, and VS Pande GitHub · msmbuilder/vde · Python · ⭐ 189 🍴 42

MSMBuilder: Statistical Models for Biomolecular Dynamics

MP Harrigan, MM Sultan, CX Hernández, BE Husic, P Eastman, CR Schwantes, KA Beauchamp, RT McGibbon, and VS Pande Biophysical Journal · DOI · Python · ⭐ 161 🍴 94

MolEncoder: Molecular Autoencoder in PyTorch

CX Hernández GitHub · cxhernandez/molencoder · Python · ⭐ 92 🍴 18

Osprey: Hyperparameter Optimization for Machine Learning

RT McGibbon, CX Hernández, MP Harrigan, S Kearnes, MM Sultan, S Jastrzebski, BE Husic, and VS Pande JOSS · DOI · Python · ⭐ 73 🍴 26

Posters & Presentations

Neural Control of Movement Society Meeting

Poster · Panama City, PAN · 2025 “Stable Control through sEMG Input: Hand Gesture Recognition on a Population with Hand Tremor”

Convolutional Neural Networks for Visual Recognition (CS231N)

Invited Presentation · Stanford, CA, USA · 2017 “Using Deep Learning for Segmentation and Counting within Microscopy Data”

Biophysical Society Meeting

Poster · Los Angeles, CA, USA · 2016 “Intrinsic Disorder in the P53 C-Terminal Regulatory Domain Yields Multiple Pathways for Folding-Upon-Binding”

Workshop on Molecular and Chemical Kinetics

Poster · Berlin, DEU · 2015 “Inferring Causality Along Transition State Pathways”

Honors & Awards

Press