Work with me

I’ve taken healthcare products from demo through hospital review into daily use

Now I build AI products the way I built clinical software: find the decision buried so deep in the workflow that nobody treats it as one, then build the version that removes the workaround. At Doximity, it was which phone number a patient sees. For an AI agent, it is what the agent is allowed to do.

See case studies See current work →
Portrait of Kanav Jain

Systems designer and independent researcher · Chicago

Value Measure Source
300,000+ calls on an average workday Doximity Dialer, which I originated as founding product lead. Doximity’s report for October–December 2025.
9 years later, in daily use

Selected case studies

Doximity

2013–17

The patient sees the office number, not the doctor’s cell

Read full case study →
Doximity Dialer “Call from” selector set to OFFICE, with a tooltip explaining the recipient sees this number as caller ID
The identity decision in the product’s own words: the clinician picks which number the patient sees, with the office line preselected.
CancerCompass / CTCA

2019–21

Plain-language next steps for 30 million annual visitors

Read full case study →
Transcarent

2021–22

Four care programs, one system, and an owner for every case

Read full case study →
See all case studies → Work with me →

What connects my work

Doximity Dialer lets a clinician call from a personal phone while the patient sees the office number, and it is still in daily use. It expanded what a clinician could do. A working solution like that easily becomes the default, even when other approaches are possible. Can a tool expand what someone can do while also giving them more say in how it works?

Research →

What I’m building

For sale

NextConsensus

Estimates how likely a guideline, label, or coverage decision is to change by a set date.

NextConsensus share card for FF-011: “Will FDA approve ivonescimab for EGFR-mutated non-small cell lung cancer after EGFR TKI therapy by Nov 14?” with a PDUFA goal date of 14 November 2026 and a registered probability of 15%
Enlarge image
Registered 2026-09-26 at 15%, against a 70% authority base rate. Unresolved at the 2026-11-04 registry snapshot.

Experimental

Ambit

Maps what an agent setup can already do, what one more step would unlock, and which of those tools actually work.

A terminal running “ambit approve” on an agent’s proposal: approved by Jordan, applicable true, with the note that the approval artifact is signed and expires
Enlarge image
The agent drafted this change over MCP, which has no tool to approve or apply it. A person approves it in a terminal under a declared name, and the approval is a tamper-evident record that expires after 24 hours.

  • Live

    Refract Replays source revisions to catch a change before the citations that rely on it go stale.

Current work: code and checks →

Writing and standards

  1. Research and essays

    From a first-author ACS Symposium Series chapter on RNA thermodynamics to 230+ essays at The Crumple Zone on how institutions distribute burden, absorb failure, and decide who can exist inside them. Selected writing →

  2. Proposed AI standards

    I wrote Ethotechnics’ framework for recording approval, challenges, and corrections. ethotechnics.org

What I want to test next

Whether a record can carry a real dispute end to end: an agent acts, a person challenges the action, and the correction reaches every decision that depended on it.

Staff Fix the Errors Your Dashboard Doesn’t Show An automation can report 94% straight-through processing while staff fix mismatched codes in shadow spreadsheets.

I take advisory and fractional product leadership roles for healthcare and AI ventures, partner with commercial co-founders on new builds, and research how institutions correct their failures. If you have a product that works and a review standing between it and its users, I’d like to hear from you. kanav@kanav.net

Revised