Vaibhav Venkat

Computational Research · Berkeley, CA

I'm currently researching how ABPs behave in a crystalline cylindrical system under Prof. Omar and Dan Evans. I've generated over 40 HOOMD-blue particle simulations with geometry modifications to log coarse-grained, hexatic ordering, topological defects, chirality, transport, and clustering behavior.

Most of my analysis consists of plotting and analyzing autocorrelations, spectral transforms (i.e Fourier + Laplacian transforms), entropy analysis, Stochastic and Partial Differential Equation fitting with validation. Recently, we are studying motility-induced phase separation (MIPS) in these long cylindrical systems through local-density distributions, dense–dilute coexistence estimates, and stochastic analysis of certain dilute "bands" on the surface.

The project also uses reproducible safetensor data and environment tooling to support scalable, multi-case CPU and GPU analyses. I used freud to log the structural properites, and mostly Scipy and Jax for the bulk of the mechanical and statistical analysis. The PDE fits used Shenfun with the SINDY algorithm, and the SDE fits used Numpyro and Optimistix. Plots primarily used Matplotlib and Seabornl.

View research code on GitHub →

Computational Researcher · Remote

I developed 11 Python, NumPy, and SciPy modules for generating structures used in CREASE scattering analysis, including patchy particles and network models. I also authored the documentation and built a dashboard connecting structures with their scattering profiles.

View research code on GitHub →

Tutorial: Machine-learning-based CREASE-2D analysis of 2D SAXS profiles to characterize anisotropic nanostructures in soft materials

ACS Measurement Science Au · February 2026

A tutorial for extending the CREASE-2D framework to interpret experimental two-dimensional small-angle scattering data.

Analysis of SANS profiles using CREASE to characterize the nanoscopic domains in lipid bilayer vesicles

APS Global Physics Summit · March 2026

Project Lead Intern · Hybrid

Led a team of five interns to design and pitch a React Native app for urban parking, including authentication, real-time parking spot logging, and a competitive ranking system.

Learning to Lead Intern · Austin, TX

Analyzed client business models and technical challenges, co-developed a strategy for an operational problem, and presented the solution to a judging panel.