About me
I work at the intersection of artificial intelligence and biology — building machine-learning methods that learn the language of cells, sequences and molecules.
Most of my research is generative: I design flow-matching, diffusion and probabilistic models that simulate cellular dynamics and predict how biological systems respond to genetic and chemical perturbations. More broadly, I work across sequence- and structure-aware modelling, uncertainty estimation and representation learning — with a consistent goal of turning predictive models into mechanistic, interpretable and experimentally useful insight.
- PhD Researcher Cambridge · Sanger
- AI × Bio Cells · Sequences · Molecules
- Generative Modelling Perturbations · Structure · Dynamics
- Discovery Targets · Molecules · Experiments
What I work on
-
Generative AI for Biology
Flow-matching, diffusion and probabilistic models that simulate cellular states and generate realistic biological data.
-
Single-cell & Perturbations
Predicting how cells respond to genetic and chemical perturbations from single-cell and multimodal data.
-
Sequence & Structure Modelling
Representation learning across sequences and molecular structure for multitask learning and drug discovery.
-
Probabilistic & Open Science
Uncertainty-aware, interpretable methods — released as open, reproducible research software.
News & Milestones
-
Started industrial placement at AstraZeneca
Multitask learning of antibodies' developability — Cambridge.
Jul 2026 -
Paper at ICML 2026
Active hit discovery accepted at ICML; attended in Seoul!
2026 -
Papers at ICLR 2026
Retrieval-augmented perturbation prediction and GGE (top 10%) at the Gen2 Workshop, ICLR; attended in Rio!
2026 -
Visiting PhD student in Bayesian deep learning
Fortuin Group, Helmholtz AI, Munich — uncertainty estimation for single-cell models.
2024 -
Started PhD in AI for Biology
University of Cambridge & Wellcome Sanger Institute.
2023
