Vision
Over the next fifteen years I hope to train as a physician-scientist, contribute research that changes how disease is detected and treated, and help build the tools and teams that bring that research to patients everywhere.
AI can read patterns across millions of data points that no single clinician could review. The hard part is making it accurate, fair and understandable enough to earn a place in real decisions.
How do we prove a medical model is safe to trust?
The brain is the least understood organ we have. New recording, imaging and brain-computer interface tools are opening questions that were out of reach a decade ago.
What could data teach us about how the brain heals?
Sequencing a genome is now fast and affordable. Turning that information into a decision for an individual patient is still the open problem.
Which variants actually change care?
Treatments are moving from the average patient to the specific one, matched to a tumor's biology or a person's genes. My cancer research sits squarely in this shift.
How early can we choose the right therapy?
Quantum methods may one day simulate molecules too complex for classical computers, which could change how drugs are discovered. It is early, and I want to understand it well.
Where will quantum methods help biology first?
None of these tools matter unless patients and clinicians can use them. The best technology fits into care instead of adding to its burden.
Who is left out when new tools are designed?
Roadmap
A direction, not a schedule. The path will change; the purpose shouldn't.
Today
Cancer research at Stanford Medicine, independent ML projects, science leadership and a first venture.
2027+
Study biology, computer science and medicine side by side. Join a research lab early and keep publishing.
2031+
Medical school and graduate research, ideally in a combined physician-scientist program.
2035+
Translate research into tools, whether inside a hospital, a lab or a company.
2040+
Lead teams that bring better diagnosis and treatment to patients at scale.