# kanav.net — Kanav Jain Last modified: 2026-10-06T00:02:49-05:00 · revision 0e3f2a2 ## Who this is AI product leader, founder, and hands-on builder taking consequential AI from prototype to adoption in healthcare and science. The essays and the Ethotechnics standards serve the product work rather than a separate career as a writer or policy researcher. ## Availability Open to new roles — I’m looking for a senior product role building products and leading product teams. Based in Chicago, IL. Open to relocation, including to Australia. U.S. and Australian citizen — no sponsorship needed in either country. Best fit: I work best before there’s a roadmap: defining the product, building a prototype, and testing whether it solves the problem. I design the architecture, write software, and lead cross-functional teams as the product grows. That draws on a decade of organizational leadership, including a 14-person R&D team. Contact: kanav@kanav.net (mailto:kanav@kanav.net?subject=Full-time%20role) Structured form of this block: https://kanav.net/api/brief.json (~2KB, everything in this section plus checkable proof) Fuller: https://kanav.net/api/profile.json · https://kanav.net/resume.json · https://kanav.net/api/case-studies.json Over MCP: https://kanav.net/mcp (Streamable HTTP, stateless; tools get_brief, get_profile, get_resume, list_case_studies, get_case_study — the same documents as above) ## What I do 14 years leading and building products in clinical care, oncology, and AI. My background is in biomedical engineering and molecular biology research. Doximity Dialer, which I originated as founding product lead, carries more than 300,000 calls on an average workday across 250+ hospitals and health systems, 9 years after I left. This work grew out of years of clinical software: escalation paths at Epic, verified identity at Doximity, reliability reviews at CancerCompass, accountable routing at Transcarent, and physician sign-off at Andwise. I now build the same mechanisms into systems where a model makes the recommendation. ## Operating posture - Human beings should not be treated as the invisible error-correcting layer of institutions. - Apparent performance = designed performance + unaccounted human compensation. - A right that cannot be exercised at tolerable cost is an incomplete form of agency. - Authority is legitimate only while there remains a working return path by which the experience of governed people can alter what governs them. ## Key pages - /about/ — Career narrative and sourced claims - /about/#beliefs — What I believe: I treat workarounds as a fault in the system. Whoever controls a system should answer for it. Being able to do something isn’t permission to do it. When cases don’t fit, I check the category first. I judge a system by what happens when it’s wrong. - /work/ — Case studies: Epic — After the software is installed (decided by ownership); Doximity — The patient sees the office, not the cell phone (decided by approval); CancerCompass / CTCA Marketplace — Scaling oncology navigation for 30 million annual visitors (decided by attention); Transcarent — Four specialty programs on one routing architecture (decided by accountability); Andwise — Reading the contract before the physician signed it (decided by who pays) - /current-work/ — Products and systems being built now: two primary products, supporting work, and additional range. Every one of these projects deals with the same failure: a record that has gone out of date and is still being relied on. A guideline that no longer matches the evidence. A permission that no longer matches what the system can reach. A correction that never reached the people who acted on the error. Each project handles one step. NextConsensus estimates when an institution will change its position. Ambit records an agent’s capabilities and the actions a person has authorized. Whether provides weekly guidance on hiring, spending, and fundraising as conditions change. Refract detects when a source has changed. Ethotechnics specifies who may act, on what evidence, and who may object. The pieces refer to one another through one shared record — who handed a decision to software, on what evidence, and how to take it back — published as an open standard, STD-07, the Revisable Delegation Record. https://ethotechnics.org/standards/std-07-revisable-delegation-record/ - /current-work/#nextconsensus — NextConsensus: Registers a dated probability that the FDA, a guideline committee, or a payer acts by a team’s deadline, built from that body’s own review record and scored in public once the deadline passes. - /current-work/#ambit — Ambit: Capability and authorization infrastructure for AI agents. It records which tools an agent can use, which it has been shown to use correctly, and which a person has allowed it to use, so autonomy can be granted one action at a time rather than all at once. - /current-work/#refract — Refract: Tells you when a source changed and which claims or citations built on it may now be out of date. The open-source engine (AGPL-3.0) replays a source’s revision history into a record anyone can re-run and get the same result. The caller still decides whether a change matters. - /current-work/#fast-harm-slow-repair — Fast Harm, Slow Repair: A draft test that measures how far an AI error spreads, how long repair takes, and what stays wrong afterward. - https://whether.work — Whether: Operating posture. Turns macro and capital signals into weekly decisions and review triggers for startup founders. - /framework/ — Ethotechnics: an open framework for making consequential AI deployable and correctable, covering the method, the principles, and the operating model. - /framework/#method — The method: what makes a claim verifiable - /framework/#operating-model — The stages of approval, the review that runs when evidence changes, who handles escalation, and independent checks - /work/ — Five case studies and contemporaneous decision records: what was known before the outcome, what was chosen, and what happened - /framework/#principles — Eight principles for building AI products in healthcare - /research/ — The research program: when people keep fixing a system’s mistakes, does their work hide the need to redesign it? Hypothesis, what would decide it, and field observations, including the pilot observation of clinician deference after seeing a recommendation - /current-work/#how-i-build — How I build: the agent-driven toolchain, capability accounting with Ambit, local model routing, agent evaluation, one change traced commit by commit, and where I stopped trusting the agent - /writing/ — Essays on AI governance, institutional power, and cognitive scarcity - /resume/ — Full resume - /contact/ — Contact and role target ## Architecture Schematics ### Agent-system accountability map ``` [ System Capability ] │ ▼ [ Action Authorization ] ─ (What it may do) │ ▼ [ Provenance Record ] ──── (What changed and what it did) │ ▼ [ Recovery Evaluation ] ── (What remains after an error) │ ▼ [ Human Override ] ─────── (Where escalation sits) ``` ## Machine-readable data - /api/profile.json — Full profile (career, case studies, beliefs, writing, links) - /api/case-studies.json — Structured case study data - /resume.json — JSON Resume format - /rss.xml — Writing feed ## Essays - The Stop Button Was Removed — https://kanav.net/writing/the-operational-andon-cord/ - People Are Not the Error-Correcting Layer — https://kanav.net/writing/the-invisible-error-correcting-layer/ - Nobody Points “Next Best Action” at Themselves — https://kanav.net/writing/nobody-points-next-best-action-at-themselves/ - The Official Record Is Late — https://kanav.net/writing/the-official-record-is-late/ - Don’t Let Reassurance Do Engineering’s Job — https://kanav.net/writing/dont-let-reassurance-do-engineering-s-job/ - If Every User Is a Potential Threat — https://kanav.net/writing/if-every-user-is-a-threat/ - You Don’t Have the Right — https://kanav.net/writing/you-dont-have-the-right/ - How to Design for Cognitive Scarcity — https://kanav.net/writing/how-to-design-for-cognitive-scarcity/ - Pending: The Political Economy of Waiting — https://kanav.net/writing/pending/ - Toothless Ethics: Why Principles Don’t Stop Machines — https://kanav.net/writing/toothless-ethics/ ## How to contact Email kanav@kanav.net. Include: what you’re building, what’s at stake, what you’ve tried, and whether you want a conversation, audit, or second pair of eyes.