(4nd) Accelerating from Inorganic Materials to Drug Discovery with Enhanced Sampling Methods and Machine Learning

I am a researcher and scientific software developer with experience in applying computational methods and machine learning to various topics in physics and chemistry. My experience comprises developing advanced sampling methods for atomistic modeling molecular molecular dynamic simulations, continuous modeling of liquid crystals, and leveraging scientific machine learning (AI for science) tools within those fields. I have worked in developing algorithms, including machine learning based strategies, and software for advanced sampling in molecular dynamics simulations. These efforts have produced methods to compute free energies at least 10 times faster than previous approaches. I have also lead the development and design of a JAX-based Python software platform for
enhanced sampling which supports different simulation tools including ASE, HOOMD-blue, OpenMM, LAMMPS, or JAX-MD.

I am particularly interested in the application of enhanced sampling methods to validate the applicability and accuracy of machine learning force fields and have shown how these strategies can increase the stability molecular dynamic simulations employing machine learned interatomic potentials.

The work I have carried out also involves applications to the study of protein—ligand complexes, and catalysis over metallic surfaces.

Teaching Interests

Mentoring is of upmost importance to me. I have mentored graduate students, and under my guidance, a project started to show results within 6 months, which is 3 times faster than what was done with previous efforts.