01Start with the new employee.
Here's the mental model I want you to carry through this whole course.
Imagine you hired a brilliant new assistant. Genuinely brilliant — top of their class, reads fast, writes beautifully, never gets tired. But it's their first day. They don't know your clients, your family, your tone, your pet peeves, or that you never schedule anything before 10am.
What would you do? You'd onboard them. Explain how you like things done, hand over the important documents, introduce them to your calendar, and check their work closely for the first couple of weeks.
That is exactly what this course is. AI tools like Claude or ChatGPT are the brilliant new hire. A digital twin is what you get after you onboard one properly — an assistant that has absorbed how you think, how you write, and what you care about.
02Blank chatbot vs. digital twin.
You may have used ChatGPT or Claude before and thought the results were… fine. Generic. That's not a flaw in the tool — it's the first-day-on-the-job problem. Same question, same AI:
Polite, correct, sounds like nobody in particular. You'd rewrite most of it.
Note: per your rules, I haven't sent this — it's a draft for your approval. I also didn't offer a discount, since deadline moves at the client's request don't qualify.
Everything that changed between those two screenshots is stuff you will build in Lessons 3–6. No magic, no code — just good onboarding.
03Personal twin, business twin — same recipe.
- › Weekly meal plan that knows the kids hate mushrooms
- › Travel itineraries in your style (museums yes, 6am hikes no)
- › Drafts the awkward messages — landlord, school, insurance
- › Gift tracker that remembers what you gave everyone last year
- › First drafts of client replies that sound like you on a good day
- › Turns messy meeting notes into your standard follow-up format
- › Answers your team's "how do we usually…?" from your playbook
- › Preps a one-page brief before every sales call
04What a twin can and can't do.
| Great at | Bad at — don't ask |
|---|---|
| First drafts in your voice | Sending things unsupervised |
| Remembering every preference you taught it | Knowing things you never told it |
| Turning messy input into your standard formats | Being right 100% of the time |
| Answering "what would I usually do here?" | High-stakes legal/medical/financial calls |
| Doing the boring 80% so you focus on the 20% | Replacing you — it's a twin, not a successor |
05The recipe we'll follow.
Every digital twin is made of the same four ingredients. This is your map:
- 01 — A brain (instructions): who the twin is, how it talks, its rules. Lessons 3–4.
- 02 — Knowledge: documents it should know by heart. Lesson 5.
- 03 — Memory: how it keeps learning over time. Lesson 5.
- 04 — Hands (connections): optional access to calendar, email, files. Lesson 6.
Notice what's not on the list: code. There isn't any.
Open a blank note and answer these three questions:
- What task do you do repeatedly that mostly involves writing or organizing?
- If you had a perfect assistant, what's the first thing you'd hand them?
- Is your twin's first job personal, business, or both?
Rule of thumb: pick something frequent and low-stakes. "Draft my routine replies" is a perfect first job. "Do my taxes" is not.
- ›A digital twin = a brilliant AI assistant after proper onboarding.
- ›Blank chatbots sound generic because they know nothing about you. That's fixable.
- ›Same recipe for personal and business: brain, knowledge, memory, hands.
- ›The golden rule: the twin drafts, you decide.