AI now produces answers faster than anyone can check them. In health, safety, and science, an answer you can't verify is worse than none. I close that gap end to end: I design the evidence, build the product, and publish the methodology that proves it.
"I turn clinical evidence into health products people can use. Then I publish the methodology so you can verify it."
Models are fluent, fast, and confidently wrong. The scarce skill is no longer generating an answer. It is knowing which answers hold up, and being able to prove it to a regulator, a clinician, or a parent. I build that layer, and I am rare because I do the whole chain myself: the science that decides what is true, the system that ships it, and the published proof that stands up to scrutiny.
Founders, clinical and regulatory leaders, and research groups in health, safety, and regulated industry, who need their AI validated and defensible, not just demoed. I work as both strategist and hands-on builder.
PhD in chemistry, NYU. NIH Kirschstein NRSA Fellow. Eugene M. Farber Award, Society for Investigative Dermatology. Seventeen publications, and four ACS talks in 2026, including an invited COMSCI symposium and a Sci-Mix selection. Founder and principal scientist of Health AI. Member of the Coalition for Health AI, the American Chemical Society, and Women in AI.
ORCID · Scholar · LinkedIn · Research Lab · Health AI · GitHub
Strategy, AI validation and evaluation, evidence architecture, speaking, and writing. Available now for the right projects.
Start a conversation