While consumer AI faces volatility, rapid FDA policy maturation is converting pharma AI uncertainty into investable risk. Read the essay →
Product Builder Computational Scientist Biologist
I build products and systems for scientific discovery, turning advances in biological computing into tools scientists can use.
I currently work at Biohub (previously the Chan Zuckerberg Initiative) at the intersection of product, AI, and biology. Before that I worked on computational target discovery at Kallyope, and before that on lymphoma therapeutics at Columbia.
My path from the lab to computational science to product still shapes how I approach the work. It lets me engage with research early, understand what is technically possible, and help shape what becomes a product. That often means working alongside researchers before there is a finished model, a clear use case, or even a settled product direction.
VariantFormer predicts how an individual's genetic variants affect gene expression across tissues. For its release, I built the technical product and GTM layer around the model: an interactive playground, analysis notebook, explainer, and launch materials. The case study visualizes predictions for 17,859 genes across 63 tissues.
Read the case study →



While consumer AI faces volatility, rapid FDA policy maturation is converting pharma AI uncertainty into investable risk. Read the essay →
The Full-Stack Playbook Every Pharma Will Copy Read the essay →
Clinical validation arrived. Now comes the pay-to-play. Read the essay →
I'm interested in AI for biology, scientific software, new products, and collaborations around them.