Knowledge representation
Structured scientific context that both a person and a model can reason over.
Research
Research that turns into working software
We work on what sits between an experiment and a result you can trust — and judge it by whether it reaches a researcher as something they can run.
Reproducible workflows
Standardizing, validating, and scaling pipelines that mix lab work and computation.
Human-AI collaboration
Interfaces and agent patterns that raise throughput without taking the researcher out of the loop.
Data and model provenance
Lineage across data, transformations, model versions, and outcomes.
Publications
Publications and systems work
The bibliography below spans self-driving experimentation, scientific software, and applied AI for materials and chemistry — the same themes that now shape our product direction.
Self-driving experimentsScientific softwareMaterials and chemistry AI
We use this body of work as both scientific evidence and a design guide for the workflows we build for research teams. We work in the open where we can, and with design partners where the science is theirs to publish.
Using data science and AI approaches toward science and technology convergence research
Yang, Quanpeng, et al. "Using Data Science and AI Approaches toward Science and Technology Convergence Research." arXiv preprint (2025).
A machine learning van der Waals potential and Monte Carlo simulation Python package
Yang, Quanpeng, Safak Callioglu, and Gaurav Arya. "A Machine Learning van der Waals Potential and Monte Carlo Simulation Python Package." (2024).
Cluster-move Monte Carlo simulation with analytical van der Waals potential
Yang, Quanpeng, Safak Callioglu, Joseph Laforet Jr., Yuanchuan Shao, and Gaurav Arya. "Cluster-Move Monte Carlo Simulation with Analytical van der Waals Potential." (2024).
A multiphysics approach for self-assembly of nanocrystal checkerboards via non-specific interactions
Wang, Y., Zhou, Y., Yang, Q., Basak, R., Xie, Y., Le, D., Shipley, W., Frano, A., Arya, G., and Tao, A. "A Multiphysics Approach for Self-Assembly of Nanocrystal Checkerboards via Non-Specific Interactions." Nature Communications 15, 3913 (2024).
Probabilistic approach to low strain rate atomistic simulations of ultimate tensile strength of polymer crystals
Cobena-Reyes, J., Yang, Q., Burns, A., Stober, S. T., and Martini, A. "Probabilistic Approach to Low Strain Rate Atomistic Simulations of Ultimate Tensile Strength of Polymer Crystals." Journal of Chemical Theory and Computation 19, 18 (2023).
Temperature-dependent density and viscosity prediction for hydrocarbons: machine learning and molecular dynamics simulations
Panwar, P., Yang, Q., and Martini, A. "Temperature-Dependent Density and Viscosity Prediction of Hydrocarbons: Machine Learning and Molecular Dynamics Simulations." Journal of Chemical Information and Modeling (2023).