Book 2 · Chapter 10: Your Second AI-Powered Week
This page lays out a second AI-powered week, day by day, with a measurement log and the chapter's prompts.
This page assumes you've read the chapter. It hands you the tools, not the lessons.
Second AI-Powered Week Plan
Monday: Claude Desktop + your personal context
45 minutesSet up Claude Desktop (from Chapter 1) with your personal context library. Not your company's. Yours.
Tuesday: AI research from scratch
60 minutesPick a real research question from work. Not hypothetical. Something you need to know.
Wednesday: Files, uploads, and your first real workflow
75 minutesCreate a workflow from Chapter 4-5 using real files you work with.
Thursday: Messy data + workflow refinement (the hard part)
90 minutes (this is the hard day)Take your Wednesday workflow and break it on purpose.
Friday: A full AI-powered workday
Most of your day, but integrated, not separateStop treating AI as a special tool. Use it for everything today.
Weekend: Reflection and adjustment
No task today except this: answer these questions with no sugarcoating.
From the Week Overview section.
Measurement Log
Pick three things you used AI for this week. For each one, ask:
- How long would this have taken before?
- How long did it take with AI?
- How much of the time savings came from speed, and how much from not having to think as hard?
Also notice
What made you NOT use AI when you could have?
What did you use AI for that you're not sure was worth it?
Where did you have to throw away AI's output and start over?
From the Week Two: Measuring What Changed section.
- Monday: does your AI know your context?
What's the main thing that wastes my time in my work right now?
- Tuesday: research-question starters
What's the current state of [technology/market/regulation] in the next 6 months? How do other companies solve [your specific problem]? What data would I need to decide between [two options]? What am I missing about [competitor/trend]?
- Tuesday: dig deeper on a conclusion
What's the strongest argument against this conclusion?
- Wednesday: a specific instruction that works
Extract the three main decision points from these notes and format them as bullets with the business impact of each decision
- Thursday: refine a workflow that's breaking
What assumptions are you making about this data? If I want you to handle inconsistencies like this automatically, what would you need to know? Can you show me your thinking on this decision you made? Where are you most uncertain?
- Thursday: when your AI gets confused
What's confusing about this task?