Commercial · in active development; the public site shows how a program works
NextConsensus
Estimates how likely a guideline, label, or coverage decision is to change by a set date.
AI product leader and hands-on builder · Chicago · Open to relocation
I turn new AI capabilities into products for hospitals and other organizations with strict deployment requirements
14 years leading and building products in clinical care, oncology, and AI. My background is in biomedical engineering and molecular biology research. See where I’d fit →
- calls on an average workday
- 300,000+
- hospitals and health systems using it
- 250+
- since I left, still running
- 9 years
Doximity Dialer, which I originated as founding product lead. Doximity’s own figures, February 2026.
Where I come in
I come in before there’s a roadmap, build and test an early idea, and then decide what to scale, what to cut, and who should own it.
I originated Doximity Dialer, led R&D at CancerCompass, and co-founded Andwise. Andwise showed in year one that physicians wanted it; I did not work out who would pay until year two, and by then every path that could fund growth made someone other than the physician the paying customer. I chose to wind down the company in 2024. Since then I have built NextConsensus, Ambit, and Refract, putting new AI capabilities into production alongside cross-functional teams.
Current work
The two products deal with one problem: records that go out of date while people still rely on them, like a guideline that no longer matches the evidence.
Experimental · open source, with a live demo and a roadmap that says what is not built
Ambit
Distinguishes the tools an agent can reach from the actions it is authorized to take.
- Refract Records when a source changed and what changed. Live · open source, with a daily observation bot in production and a 16,146-edit public test set
- Fast Harm, Slow Repair A draft test of how far an AI mistake spreads and whether the fix reaches everyone it affected. Draft · 1 of 12 development cases built; no model results yet
Selected work
The patient sees the office, not the cell phone
Adoption may come before official approval only while the risk stays with the person choosing. When the risk is clinical, approval comes first.
Dialer still carries more than 300,000 calls on an average workday across 250+ hospitals and health systems, 9 years after I left.
Read full case study →
Four specialty programs on one routing architecture
One system can carry many programs once every decision point has a named owner.
Routing delays stopped being the primary bottleneck, and the gap between the best and worst service lines narrowed.
Read full case study →
Reading the contract before the physician signed it
Who pays decides who the product works for. Adoption proves people want it; it does not settle that.
1,200+ physician users, then a deliberate wind-down in 2024 rather than a business model that made an employer or advisor the customer.
Read full case study →Revised