Molecular modeling algorithms
Inverse kinematics, ensemble sampling, stochastic reaction-diffusion, geometric constraints, and quantitative model validation.
Computational biophysics × scientific software
I build reusable tools for molecular structure and dynamic research.
C++ and Python systems for protein conformational sampling, biomolecular self-assembly, structural bioinformatics, and high-performance scientific computing.
structure → dynamic
method → software
What I work on
Methods that survive contact with real scientific workflows.
My work spans algorithm design, scientific validation, scalable implementation, and the interfaces researchers use to for research.
Scalable scientific systems
C++ and Python infrastructure spanning MPI parallelization, reproducible workflows, structural data processing, and automated validation.
Scientist-facing software
Command-line tools, desktop plugins, web interfaces, molecular visualization, and code-generation systems built for maintainability.
9 applications
Structural bioinformatics tools developed
3 deployment targets
VMD, PyMOL, and web from a single GUI specification
~1,000 structures
Sampled across protein-backbone and IDR ensembles
~3,000 molecules
Minute-scale virus self-assembly trajectories
Selected work
Research methods developed as reusable infrastructure.
Each case study separates the scientific problem, the algorithm or architecture, my contribution, and the evidence used to validate the result.
Open sourceCross-platform GUI generation
A Python framework that converts a designed interface and CLI metadata into a validated, platform-independent JSON specification, then generates consistent scientific applications for desktop molecular viewers and web environments.
Read case study
Ongoing workProtein backbone sampling
An extension of tripeptide loop-closure sampling that releases terminal constraints, introduces SE(3) motions and shifted sampling frames, and couples broad backbone geometric exploration to all-atom structural validation.
Read case study
Open sourceNERDSS + ioNERDSS infrastructure
A connected software stack that transforms PDB or mmCIF structures into simulation-ready coarse-grained models and runs explicit particle-based reaction-diffusion and self-assembly at larger spatial and computational scales.
Read case study
PublishedBiomolecular assembly models
Structure-resolved stochastic and kinetic models of clathrin lattices, HIV-1 Gag assembly, and dynamin recruitment that integrate experimental constraints and produce testable system-level predictions.
Read case studySelected publications
Mechanistic modeling, molecular structure and dynamic, and scientific computing.
A focused selection of work in protein assembly, structural workflows, high-performance simulation, and research software.
2026
Preprint · arXiv · arXiv:2602.16047
A Unified, Cross-Platform Framework for Automatic GUI and Plugin Generation in Structural Bioinformatics and Beyond
2026
Preprint · bioRxiv · 2026.01.27.702082
Transforming Macromolecular Structures into Simulations of Self-Assembly
2025
Journal article · Journal of Computational Chemistry
Parallelization of Particle-Based Reaction–Diffusion Simulations Using MPI
2024
Journal article · Nature Communications · 15, 4060
Dynamin1 Long- and Short-Tail Isoforms Exploit Distinct Recruitment and Spatial Patterns to Form Endocytic Nanoclusters
2023
Journal article · eLife · 12:e84881
Structure of the HIV Immature Lattice Allows for Essential Lattice Remodeling within Budded Virions
Background
Physics training, biological questions, engineering execution.
I work where molecular modeling and software engineering meet: turning research algorithms into maintainable systems that scientists can validate, scale, and reuse.
Current
Research Engineer · Inria
Developing inverse-kinematics protein-backbone samplers and a cross-platform framework for automatically generating VMD, PyMOL, and web applications.
Previously
Johns Hopkins University · Biophysics
Built NERDSS and ioNERDSS infrastructure and led computational studies of clathrin, retroviral Gag, and membrane-associated assembly.
Training
Ph.D. in Condensed Matter Physics
Developed quantitative models connecting molecular conformational changes and chemical transitions to the emergent mechanics of kinesin motors.
Interests
Reusable research infrastructure
Scientific software, protein conformational search, hybrid physics/ML workflows, simulation platforms, and agent-usable tools for molecular discovery.
Contact
Building a molecular-modeling method or scientific platform?
I am interested in research and engineering roles that connect computational methods to reliable, scalable software used by scientists.