Open to opportunities

Anikeya Aditya

@anikeyaaditya

Multidisciplinary computational scientist combining applied mathematics, ML, and HPC software to enable high-scale simulations and discovery.

Belgium

Message

What I'm looking for

I’m seeking fully remote roles in computational science, AI/ML for science, data science, or sports analytics—where I can build scalable HPC workflows and production-quality research software in distributed, async teams.

I’m a multidisciplinary Computational Scientist and Applied Mathematician with 6+ years of research experience across molecular dynamics, density functional theory, machine-learned interatomic potentials, and applied deep learning. My work bridges Bayesian inference, stochastic processes, and differential equations with production-grade HPC software engineering in Python.

As a Computational Research Scientist (PhD Researcher) at University of Southern California, I designed and validated NequIP/Allegro equivariant neural network force fields on ab initio MD datasets, reaching DFT-level accuracy and enabling nanosecond-scale MD simulations for systems up to 363,744 atoms. I also architected MXM, a production-quality Python package for generating periodic multi-layer moiré supercells, and engineered scalable SLURM pipelines for multi-TB workflows with 10× throughput improvements.

I’m equally hands-on across modeling and implementation: I trained CNNs for crystalline phase and defect identification in millions of atoms, and I built Bayesian & statistical methods to quantify structural uncertainty across large configuration spaces. Earlier, I developed DFT-based photoluminescence lineshape modules and built Bayesian inference pipelines using Monte Carlo and inverse transform sampling—always prioritizing reproducibility and delivery in distributed, async collaborations.

Experience

Work history, roles, and key accomplishments

Computational Research Scientist

University of Southern California

Aug 2019 - Dec 2025 (6 years 4 months)

Designed and trained NequIP/Allegro equivariant ML force fields on ab initio MD datasets, achieving DFT-level accuracy (test RMSE 3.88 meV/atom; 0.083 eV/Å forces) to enable nanosecond-scale MD for systems up to 363,744 atoms. Architected HPC-ready Python software and SLURM workflows, improving MD post-processing throughput by 10× and applying Bayesian/statistical methods to quantify structural un

NequIP Allegro ML Force Fields PyTorch DFT (VASP)Slurm

Research Software Engineer

University of California, Santa Cruz

Aug 2018 - May 2019 (9 months)

Developed Python modules to compute photoluminescence lineshapes for NV-center point defects using DFT (Quantum ESPRESSO) combined with Fourier-transform generating functions. Replaced numerically expensive integration routines with analytical expressions in Mathematica to significantly reduce CPU overhead while preserving simulation fidelity.

Python Quantum ESPRESSO DFT Photoluminescence Mathematica NV Center Modeling Git Github

Undergraduate Researcher

University of California, Santa Cruz

Sep 2016 - Jun 2018 (1 year 9 months)

Built a Bayesian inference framework to infer primordial black hole mass functions from LIGO/Virgo gravitational-wave data, extending monochromatic merger-rate calculations to lognormal and power-law extended models. Implemented inverse transform sampling and Monte Carlo methods and validated results against published work in limiting cases.

Monte Carlo Methods Inverse Transform Sampling Gravitational Wave Data Analysis Statistical Modeling LIGO Virgo Uncertainty Quantification Scientific Computing

Education

Degrees, certifications, and relevant coursework

University of Southern California

Master of Science (MS), Computer Science

2023 - 2025

M.S. in Computer Science from USC (2023–2025).

University of Southern California

Doctor of Philosophy (PhD), Materials Science

2019 - 2025

PhD in Materials Science from USC (2019–2025). Dissertation focused on emergent phenomena in two-dimensional materials via MD and ML for strain, grain boundaries, and twisted superlattices.

University of Southern California

Master of Science (MS), Materials Engineering

2019 - 2025

M.S. in Materials Engineering from USC (2019–2025).

University of California, Santa Cruz

Bachelor of Science (BS), Physics

2014 - 2018

Grade: Highest Honors

B.S. in Physics (Highest Honors) from UC Santa Cruz (2014–2018). Thesis: inferring the mass function of primordial black holes from gravitational wave observations (advisor: Prof. Stefano Profumo).