Denys is a PhD student at the NYU Center for Data Science, advised by Prof. Julia Stoyanovich. His research interests lie in AI safety, time-series modelling, and healthcare applications. He aims to develop fair, stable, robust, and transparent AI systems that can be deployed responsibly in real-world, high-stakes environments such as clinical decision support.
He earned his B.S. in Computer Science summa cum laude at Ukrainian Catholic University (UCU). He then completed a postgraduate research program at Imperial College London, where he worked on privacy and robustness in decentralized learning. Before joining NYU, he spent a year and a half as a research engineer at Seattle Children’s Research Institute, building real-world AI applications on biomedical signals. Along the way, he spent three years as a research fellow in the RAI for Ukraine program, organized by NYU and UCU, working at the intersection of model stability, uncertainty quantification, and algorithmic fairness.
Selected publications
VirnyFlow: A Design Space for Responsible Model Development
Denys Herasymuk, Nazar Protsiv, and Julia Stoyanovich
@article{herasymuk2025virnyflow,title={{VirnyFlow}: A Design Space for Responsible Model Development},author={Herasymuk, Denys and Protsiv, Nazar and Stoyanovich, Julia},year={2025},eprint={2506.01584},archiveprefix={arXiv},journal={CoRR},primaryclass={cs.LG},keywords={working,RAIforUkraine,data}}
Still More Shades of Null: An Evaluation Suite for Responsible Missing Value Imputation
Falaah Arif Khan, Denys Herasymuk, Nazar Protsiv, and Julia Stoyanovich
@article{shades,title={Still More Shades of {Null}: An Evaluation Suite for Responsible Missing Value Imputation},author={{Arif Khan}, Falaah and Herasymuk, Denys and Protsiv, Nazar and Stoyanovich, Julia},year={2025},journal={Proc. {VLDB} Endow.},volume={18},number={9},doi={14778/3746405.3746416},keywords={journal,RAIforUkraine,data}}
Responsible Model Selection with Virny and VirnyView
Denys Herasymuk, Falaah Arif Khan, and Julia Stoyanovich
In Companion of the International Conference on Management of Data,
SIGMOD/PODS, Santiago, Chile 2024
@inproceedings{DBLP:conf/sigmod/HerasymukKS24,author={Herasymuk, Denys and Khan, Falaah Arif and Stoyanovich, Julia},editor={Barcel{\'{o}}, Pablo and Pi, Nayat S{\'{a}}nchez and Meliou, Alexandra and Sudarshan, S.},title={Responsible Model Selection with Virny and VirnyView},booktitle={Companion of the International Conference on Management of Data,
{SIGMOD/PODS}, Santiago, Chile},pages={488--491},publisher={{ACM}},year={2024},url={https://doi.org/10.1145/3626246.3654738},doi={10.1145/3626246.3654738},keywords={conference,demo,RAIforUkraine,data},author+an={1=self;2=self;3=self}}
On Fairness and Stability: Is Estimator Variance a Friend or a Foe?
Falaah Arif Khan, Denys Herasymuk, and Julia Stoyanovich
@article{DBLP:journals/corr/abs-2302-04525,author={Khan, Falaah Arif and Herasymuk, Denys and Stoyanovich, Julia},title={On Fairness and Stability: Is Estimator Variance a Friend or a Foe?},journal={CoRR},volume={abs/2302.04525},year={2023},doi={10.48550/arXiv.2302.04525},eprinttype={arXiv},eprint={2302.04525},keywords={working,fairness,stability,RAIforUkraine},}
