Welcome to Day 1.
You've probably lived this.
Every new chat starts from zero.
Same introduction. Same context. Same explanations.
Every single time.
"But chats have memory now."
They do. And somehow, that made it worse.
Because that memory is confused — and you don't control it.
Maybe you once mentioned a side idea, in some conversation, months ago.
Nothing important. A passing thought.
The AI decided it was important.
Now it keeps bringing it up. Making assumptions you never asked for. Pulling conversations where you didn't want them to go.
So you get frustrated and close the tab: "AI just doesn't work."
Or worse — you settle for a result that's... fine. Just fine. Nothing you'd put your name on.
Now think of a real colleague.
You introduce yourself once.
They learn who you are, what you're working on together, how you like things done.
And from that day on, they remember.
That's problem one on the map, and today we fix it:
Memory. Organized. Yours. On purpose.
So today we flip the whole thing around.
Today, AI interviews YOU.
Your fifteen minutes
Here's how it works.
You paste one text.
The AI starts asking you questions. One at a time.
About your role, your work, and the task that eats your week.
You just answer. Nobody on earth knows your job better than you.
At the end, it hands you two things.
One: your briefing. A short document about you, your work, and your rules. Save it anywhere you'll find it tomorrow — your notes app, a Word file, an email to yourself.
Two: your task. The ONE piece of your job (a report, a document, a process you repeat and want off your plate) that Bruno takes over this week.
Here's the text. Copy it, paste it, answer honestly:
"I'm a manager building an AI colleague for my work. Context for you, so you never need to ask me about it: this interview will produce a briefing. Tomorrow that briefing becomes the permanent memory file (CLAUDE.md) of an AI assistant living in a folder on my computer, and later this week it becomes the base for teaching that assistant to run one of my recurring tasks (a skill). You don't need to explain any of this to me.
Your job today: interview me, one question at a time, in plain language. No jargon, no technical questions, and never ask me why I'm doing this or what tool I'll use: you already know. Find out: my role, what my work looks like, how I like things done, and the ONE task I should delegate first. The task must pass five checks: I repeat it every week or month · it's built from files I already have on my computer · I can show a finished example of a good version · the result is a file of any kind (a report, a spreadsheet, a presentation, a written document) · it needs no live data from the internet. If my first idea fails a check, tell me why in one plain sentence and help me pick a better one.
When you have everything, write "BRUNO'S BRIEFING v1" as plain text I can copy and save anywhere: who I am, how I work, my rules, and the chosen task. No file formats, no technical vocabulary inside it. Then stop."
When the briefing is done, add one line to it yourself.
By hand.
The most valuable line I own:
"If you don't know something, write UNKNOWN. Never guess."
Why this line?
Because AI wants to please you.
And a colleague who wants to please you will invent an answer before admitting he doesn't know.
You've met humans like this.
Now yours has the rule in writing. Forever.
Congratulations.
Most people never get past asking AI random questions.
You just gave it a proper briefing, like a manager onboarding a new hire.
Problem 1 is already cracking.
Everything smooth, briefing saved.
Got there, but something felt off.
Stuck.
Whatever your color, DM me on LinkedIn: DAY 1 + your color.
Why do I ask? Because the score tells YOU if it worked — and it tells ME where you got stuck, so I can fix it.
This course is free, and it exists for one reason: to hand you in a few emails what took me twelve months to figure out.
Green, I celebrate with you. Yellow or red, I get you unstuck. I read everything.
Tomorrow, Bruno gets a body.
You'll build his home in about ten minutes, put your briefing inside...
...and watch him read YOUR files back to you, by name.
See you tomorrow.
Filippo