Aayush Gupta
Computational chemist and machine learning scientist for drug discovery.
I build machine learning and computational chemistry systems for drug discovery — spanning peptide modalities, protein binder design, molecular simulation, ADME prediction, and high-throughput screening. My work sits between chemistry, physics, and AI.
- 10+ years across academia & industry
- AI/ML, molecular simulation & cheminformatics
- PhD in Computational Chemistry, UIC
Snapshot
Scientific depth, model-building rigor, and product-facing execution
I work in the gap between research ambition and operational usefulness: building systems that help chemistry teams rank molecules, design experiments, and move with more confidence.
Current Focus
Building practical AI for molecular decisions
I focus on systems that help scientists rank, design, and test molecules with more confidence — ML-based molecular representations, protein binder design workflows, simulation pipelines, uncertainty-aware selection strategies, and production research tooling.
Career Arc
From physics-grounded chemistry to product-minded AI
My path runs from quantum chemistry and molecular modeling to production-facing machine learning for biotech. I have worked across research-heavy environments and fast-moving startup teams.
Selected Highlights
Work I am proud of
A few representative chapters where machine learning, simulation, and scientific tooling came together in ways that materially changed how teams operated.
Vida Vinci
Machine learning for peptide and protein design
- Built ML-cheminformatics encoder representations and predictive ADME models grounded in internal data.
- Worked on protein binder design, Free-Wilson analysis, and uncertainty-aware ranking workflows.
- Implemented BIBD and CP-SAT based library design for synthesis-ready candidate generation.
DeepCure
High-throughput molecular simulation at scale
- Built an end-to-end OpenMM MD + ABFE/RBFE platform with in-house Amber preparation.
- Engineered a high-throughput MD screening method that improved throughput by roughly 300×.
- Strengthened internal AI and computational chemistry platforms for molecule and pose quality.
Exscientia
Bridging medicinal chemistry and AI product work
- Worked across ML and medicinal chemistry teams on ADMET, active learning, and linker generation workflows.
- Led work inside the PhysicsML team and contributed to product direction.
- Helped deliver EXS-NNP and translate advanced modeling into usable capability.
Education
Training
Doctoral work under Prof. Huan-Xiang Zhou focused on machine learning methods for advancing computational chemistry.
Built the early foundation in chemistry, materials, simulation, and research.
Research Theme
Where my work creates leverage
- Turning advanced scientific methods into workflows that research teams can actually use.
- Connecting statistical learning with physics-based simulation.
- Designing decision systems that help rank compounds and understand uncertainty.
- Working across code, experiments, and multidisciplinary teams.