A surgical scalpel on a dark, marked work surface

Surgeon-led AI for medical practices

The scalpel has not changed. Your practice can.

One painful practice problem. One working AI solution. Every week. Taught by a surgeon who actually builds and uses these systems.

See the next live buildRead the playbook

For physicians and the teams who keep their practices moving. No software-engineering background is required for the live-build track.

Featured live build

The Op-Note That Refuses to Invent

Watch a fictional dictation become a structured operative-note draft that flags what was never said instead of filling in the blanks.

Duration
60 minutes
Format
Live online
Cost
Free

Registration is open. The live date is announced by email.

Get the next live date

What you will leave with

  1. The rule that forces missing facts to be flagged, not fabricated

  2. How a template becomes the structure the AI must follow

  3. How to stress-test the workflow before it touches real work

  4. How to preserve a human review point

The live-build method

Every class ends with something that works.

  1. See it run

    Start with the finished workflow.

  2. Learn the rules

    Identify instructions, guardrails, and human decisions.

  3. Build your version

    Adapt the system to the practice's tools and constraints.

  4. Test the edges

    Try missing facts, conflicts, and failure cases.

Process proof

Real systems. Fictional demonstrations.

Working patterns from systems Jesse has built, shown publicly with fictional information.

Case file 01Documentation workflow

Clinic Notes

Documentation workflows that keep clinic and operative-note processing separate and flag unsupported facts.

Case file 02Correspondence workflow

JD Mail

Correspondence workflows that turn messy source material into concise, review-ready drafts.

Case file 03Knowledge workflow

Practice Brain

Approved knowledge made searchable, with an explicit not-found response instead of a guess.

The first eight builds

A curriculum built around the work that actually hurts.

  1. The Safe AI Practice

    Map a recurring practice problem into a bounded workflow with clear inputs, outputs, and ownership.

    GuardrailChoose tools and a deployment model only after review by your organization, compliance team, or qualified counsel.

    Open the build notes

  2. The Operative Note That Refuses to Invent

    Turn fictional dictation into a structured operative-note draft that identifies what was never said.

    GuardrailFlag unsupported facts instead of filling gaps, and preserve a qualified human review point.

    Open the build notes

  3. The Letter Queue That Eats Your Evening

    Turn messy source material into a concise, review-ready correspondence draft.

    GuardrailUse only supplied source material, identify conflicts, and keep a person responsible for the final message.

    Open the build notes

  4. The Prior-Authorization Packet

    Organize a fictional request into a clear packet with requirements, supporting material, and unresolved gaps.

    GuardrailDo not infer criteria, history, or supporting facts that are absent from the supplied material.

    Open the build notes

  5. The Denial Appeal and Peer-to-Peer Brief

    Turn a fictional denial and its source record into a focused appeal outline and conversation brief.

    GuardrailSeparate quoted policy language, supplied facts, and open questions so a reviewer can verify each one.

    Open the build notes

  6. Know the Chart Before You Enter the Room

    Condense a fictional chart into a focused pre-visit brief with unanswered questions made visible.

    GuardrailTreat the output as a draft for review, not a substitute for the source chart or clinical judgment.

    Open the build notes

  7. Give Your Staff a Practice Brain

    Make approved practice knowledge searchable so staff can find an answer or receive an explicit not-found response.

    GuardrailAnswer only from approved source material and say when the answer is not present.

    Open the build notes

  8. Build Your Practice's 90-Day AI Plan

    Sequence a small set of practice workflows into a practical 90-day build and review plan.

    GuardrailStart with bounded, reversible work and assign a human owner and review point to every workflow.

    Open the build notes

Your instructor

Jesse Dashe, MD

A doctor teaching doctors, because the workflow is the hard part.

Jesse is a board-certified orthopedic hand surgeon in active practice. He teaches systems he has built.

Safety boundary

HIPAA-conscious

Useful first. Safe by design. Honest about the limits.

  • Public demonstrations and community examples use fictional data.
  • AI produces drafts; a qualified person reviews and decides what is used.
  • Visitors choose their tools and deployment model.
  • Real implementations require review by the visitor's organization, compliance team, or qualified counsel.

Educational content only. Not medical, legal, or compliance advice. AI Surgeon does not guarantee that any workflow or setup is HIPAA compliant.

Ways in

Start free. Add support when the work calls for it.

Every path begins with the free community.

Questions, answered

Before you bring the workflow.

Who is AI Surgeon for?
AI Surgeon is for physicians, practice owners, and the teams who keep medical practices moving.
Do I need to know how to code?
No. The live-build track is designed for people without a software-engineering background.
Will you show real patient data?
No. Public demonstrations and community examples use fictional data.
Is this compliance training?
No. AI Surgeon is educational content, not medical, legal, or compliance advice. Real implementations require review by your organization or qualified advisers.
How do I receive the next live date?
Join the list with your name and email. The next live date is announced by email.

Bring one real problem

Bring the workflow that keeps following you home.

We will solve it one system at a time.

See the next live buildJoin the free community