01 · Learn
Understand the move
Short written modules explain the method, the safety rules and the tools — in plain language, for physicians, not engineers.
Build a practical personal AI workflow for your medical work.
Learn practical AI workflows with clear safety boundaries and no patient data required. The course is taught in Claude, with adaptation guidance for ChatGPT and Gemini. You’ll create a setup shaped around how you work, repeatable routines for verifying outputs and citations, and clear practices for keeping patient identifiers out of what you share with AI tools. Includes 10 hands-on modules and 6 copy-paste toolkits.
The shift
Most people use AI like a vending machine.
Ask a question, get an answer, walk away — and nothing compounds. This course teaches the opposite: AI as an operating system for your professional life, where your notes, your research, and your outputs build on each other.
It starts with the first move — giving your AI a permanent memory of who you are and what you're working on. Everything after is built on top of that.
How the course works.
Practical and sequential. You build the system on your own machine while you learn it — and you keep every piece.
02 · Build
Build it as you go
Copy-paste toolkits and hands-on labs turn each lesson into a working part of your own system, running on your own computer.
03 · Keep
Leave with real assets
The outcome isn't notes. It's a memory layer, a research workflow, writing safeguards and repeatable systems that are yours to keep.
Safety is built into the workflow.
The guardrails aren't a disclaimer at the bottom of the page — they're the working rules you set up from the first lesson.
The AI drafts; you act
You set it up to draft only: sending, posting, prescribing, booking and signing stay with you. Every outward action is yours.
Cite it or flag it
Uncertainty is made visible before the work ever leaves your desk.
Your data stays yours
Patient data never leaves your control, and nothing you build depends on sending it away.
“A green checkmark is not proof of life.” You learn to verify that the note is right and the citation is real — not just that a tool reported success.
By the end of the course, you'll have a system that looks like this.
Not a folder of prompts — six working pieces you build alongside the lessons, each one solving a specific part of your week.
An AI that remembers your work
Your specialty, research interests, current projects and working preferences load automatically at the start of every session.
A research brief every morning
A short digest of new papers and guideline changes across the topics that matter to your work — waiting before you sit down.
Save a paper in one click
From an open browser tab to an organised, AI-readable research library — without breaking your reading.
Build a lecture deck faster
Go from verified research to a structured outline to a finished slide deck — without the blank-page start.
Catch fabricated citations before they ship
Any clinical claim is held with a [VERIFY] flag until it's checked against a real source — you approve it, or it doesn't ship.
A credible medical brand — without the hype
An evidence-based content system with approval gates, evidence tiering and a defined voice. It builds real credibility instead of chasing engagement — and nothing publishes until you say so.
The same week, before and after.
The point of the system isn't more tools. It's that your work stops starting from zero.
Before
Every task starts from scratch
- Open a chat and re-explain your context.
- Get an answer, then lose it by next week.
- Search PubMed by hand, tab by tab.
- Keep 19 papers open “to read later.”
- Start the slide deck from a blank page.
- Hope the citation you half-remember is right.
After
Your work compounds
- Claude opens already knowing your work.
- Your research brief is waiting each morning.
- Papers land in one organised library.
- Writing runs through a verification workflow.
- Decks begin from a verified outline.
- Every claim is checked before it ships.
Ten core modules.
You can start with Claude, ChatGPT or Gemini. Demonstrations use Claude, with guidance for adapting each workflow to your preferred app. The first four modules need no terminal or additional software. The Advanced Track then moves the same system into your terminal, which is where files, on-machine memory and automations become possible. From there it is a set of paths rather than a fixed sequence: productivity, research, teaching or content. Every terminal command is written out for you to copy and paste.
Foundations in the Claude app — no terminal, no install
- F020 min
The Mindset Shift & Ground Rules
From “using AI” to building a system — and the safety rules that make it usable in medicine.
- F0.525 min
How to Actually Talk to It
The six techniques that carry almost every clinical prompt, and a loop for fixing one that misbehaves — nothing to install.
- F130–45 min
Your AI's Memory in the Claude App start here
Projects and custom instructions — the fast, no-setup way to make Claude open every session already knowing your work.
- F245 min
Knowing It's Still Right
A golden set — twenty questions you know the answers to — so you can tell whether a change helped, and catch invention before it reaches a letter.
Advanced Track the same system in your terminal — every command written out to copy and paste
- A12–3 hrs
Claude in Your Terminal & the Context Window start of track
The step the rest of the track is built on: permanent on-machine context, your first CLAUDE.md, and a session-start briefing.
- A22–3 hrs
Productivity & Workflow Automation
Domain personas, morning briefs and guarded assistant workflows for a real clinical week.
- A32–3 hrs
Knowledge Capture & Research Automation
One-click paper capture, an AI-readable research database, and an automated PubMed feed.
- A41.5–2 hrs
AI-Powered Presentations with Gamma
Move from verified research to a lecture structure and a finished deck.
- A52–3 hrs
AI for Literature & Manuscripts
Cite-or-[VERIFY] and safe co-writing workflows that keep you honest at speed.
- A62–3 hrs
Build a Credible Medical Brand with AI
Approval gates, evidence tiering and a voice contract — credibility, not influencer theatre.
Built from a working clinical system — not AI tips.
I'm Dr. Fatemah AlYaqout, a consultant allergist and immunologist, dual-trained in adult and paediatric Allergy & Immunology, with a clinical focus on immunodeficiency. Everything in this course comes from systems I built and run in my own week — the memory layer, the morning research brief, the writing safeguards — not from generic productivity advice.
I built these over more than six months, mistakes included, because the tools kept forgetting who I was and quietly getting things wrong. What's here is the distilled, transferable core: the parts that carry over to any physician's work. Evidence and elegance, with no previous coding experience required — though some modules include guided copy-and-paste technical setup.
Choose how much help you want.
Same system in every tier. The difference is how much of it we build together.
Self-Paced
US$399
Founding cohort: US$299 with code FOUNDING, valid 23 November–6 December 2026.
- All 10 modules and 6 toolkits, available today.
- All future updates to this course included.
- Help shape future updates through short feedback forms after each module.
- Full refund within 14 days of purchase, without giving a reason.
Free and separate from the course purchase, open to anyone who registers:
- Companion videos: 2 available today, more as they are released.
- One live group Q&A on Thursday 17 December 2026, 19:00 Kuwait time, on Zoom, recorded.
Guided
US$899
The course, plus real help getting it running.
- Everything in Self-Paced
- Two 60-minute implementation sessions
- Your setup reviewed with you, on your own machine
Concierge
Apply
Everything in Self-Paced, plus six implementation sessions in which we build your system together. Priced by formula rather than a flat fee: the course, plus six implementation sessions, plus a 10–15% project premium covering preparation, customisation, between-session review, documentation, support and final handover. Limited slots.
- Session 1 — workflow audit and implementation plan
- Session 2 — AI memory and context setup
- Session 3 — research and knowledge system
- Session 4 — writing, presentations and verification
- Session 5 — automation, permissions and safety
- Session 6 — testing, refinement and handover
Medical student access
Free
For currently enrolled medical students, subject to enrolment verification.
- Full access to all 10 modules and 6 toolkits.
- Companion videos as they are released.
- All future updates to this course included.
- The live group Q&A is free and open to anyone who registers.
Already enrolled, or found us elsewhere? Single implementation sessions are also available — an hour with me to unstick your specific setup. Get in touch about scope and availability.
Book an implementation sessionQuestions physicians actually have.
Do I need to know how to code?
No. You start in your chosen chat app—Claude, ChatGPT or Gemini—with no coding required for the first four modules. Demonstrations use Claude, with guidance for adapting each workflow to your preferred app.
Do I have to use the terminal?
Not to start, and not to get a real result — but you do need it for most of the course, so here is the honest split. The four Foundations modules run entirely in the Claude app you already know, and they leave you with an AI that remembers your work and your standards. The other six core modules are the Advanced Track, and those do use the terminal, because that is where files, on-machine memory, the automations, the research feed and reaching your assistant from your phone become possible. Every command there is written out for you to copy and paste, and you take those paths at whatever pace suits you.
What do I need to get started?
A Mac, Windows or Linux computer and a subscription to Claude, ChatGPT or Gemini. You begin in your chosen chat app, and the first four modules require no terminal or additional software.
Do I need Claude?
No. Demonstrations use Claude and Claude Code because those are the tools I use every day. The course provides guidance for adapting the chat-based workflows to ChatGPT and Gemini, and the terminal workflows to Codex CLI. The underlying approach is consistent across tools: a standing profile, a dedicated space for each project, your own documents, and standing instructions stored on your own machine.
Is this just “ChatGPT tips” repackaged?
No. It's a real system — a memory-equipped AI, an organised research engine, and safe writing and presentation workflows. You leave with working systems and assets, not a list of prompts.
What if the AI tools change after I buy?
Every tier includes lifetime updates. When Claude Code, Gamma or a workflow changes, the affected module is updated. You're buying the method, kept current — not a snapshot.
Start this weekend
Build the assistant that finally knows you.
The foundation fits in a single weekend. By Monday, your AI opens every session already knowing your work.