Applied Scientist for Digital Biology at NVIDIA · PhD candidate at the University of Oxford
Machine Learning & Data Science
I work on machine-learning research, product strategy and implementation with scientific, engineering and product teams, including external research collaborators.
Previously Global Life Sciences Research Alliances Lead at NVIDIA, with more than 15 years of experience across software, product, machine learning and data science.
Current research
Learnable geometric features for point clouds using diffusion geometry and differential forms.
A training-free time discretisation for low-budget flow and Schrödinger-bridge sampling.
Diversity-regularised direct preference optimisation for more varied peptide sequences while retaining structural similarity.
Ensemble-first structure prediction across ordered, flexible and intrinsically disordered proteins.
Selected work
Managed an eight-person research and engineering team working on ProteinMPNN fine-tuning, diversity-regularised preference optimisation and sequence–structure evaluation.
Read the caseWorked with research, product and engineering teams on BioNeMo examples, multi-node workflows and technical evaluations for life-sciences organisations.
Related experienceBuilt and deployed a reinforcement-learning and Bayesian system using Spark, PyTorch, production APIs, Azure and a web interface.
Read the caseHow I work
Method design, baselines, controlled experiments and scientific writing.
Managing research and engineering teams, reviewing technical work and assigning clear responsibilities.
GPU workflows, production ML, APIs and reproducible implementation.
Technical evaluation, scientific context and ecosystem work to support venture investment decisions.
