How to Use AI Effectively | A Conversation with Claude
About this episode
Vance turns the interviewer's chair over to Claude for a full-episode role reversal, walking through his personal AI operating system: a "thinking bot" (Claude, paid subscription, unlimited exploratory use) paired with a metered "doing bot" run through OpenRouter for anything that costs money or touches the live website. He narrates concrete use cases — QuickBooks reconciliation including a Bitcoin deposit that broke the software's normal categories, digitizing old lab printouts to spot health trends, and using a GitHub repo as version-controlled "instructions" storage to avoid contradictory prompt buildup. He draws a hard line against using AI to write anything personal (an email, a testimonial to someone he knows) and recounts an embarrassing failure where he let an agent send an uncorrected calendar invite to a whole network. The episode closes on unease: Vance compares onboarding people to AI to handing them the apple in Eden — once you see what it can do, there's no going back — before reframing that as a net positive if the goal is buying back time with his kids rather than just producing more work. A guest segment following the main interview is a separate Legacy Interviews client testimonial (the Ontario farm family), not related to the AI conversation.
“I don't think people leave money on the table, they treat the first output like gospel when really it's just a starting point for dialogue.”
“The more context it has about you in the instructions, the easier it is for it to give you the information you need... I have the AI ask me questions. I'll say, Hey, I want you to ask me, you know, five questions that will help you understand what instructions would give you the sort of context that would help us work better together.”
“AI is going to be so ubiquitous that we will never go back. And I think the nostalgia that we will have for a period of time before everything was known digitally, that we will have a level that will be more intense than all of us that still feel like, oh remember back in the day when you didn't have cell phones.”
Key moments
- Vance frames the episode's origin — a friend, Zach Smith, asked him to explain AI use to hesitant late adopters.
- The QuickBooks story — AI handles routine transaction categorization at "bookkeeper level," saving what would be two days of work.
- The Bitcoin deposit edge case — Vance has AI backwards-engineer 2025 transactions to check whether he booked a crypto payment correctly.
- Vance explains his "ask AI to ask you questions" technique for building instruction sets from scratch.
- The thinking bot / doing bot split — Claude for unlimited exploration, an OpenRouter-based agent metered per query for anything that costs money.
- The "apple in the Garden of Eden" moment — Vance describes feeling conflicted mid-conversation with Sean Newman about how addictive/irreversible AI adoption is.
- Failure case — an AI-drafted calendar invite sent to a whole network with the wrong time, requiring an embarrassing round of corrections.
- Yosha Bach's rule, endorsed by Vance: never let AI write personal communication to someone who has a relationship with you.
- The GitHub repo as "library" metaphor — version-controlled instruction files to prevent contradictory rules stacking on top of each other.
Notable quotes
“It's a little like handing somebody the apple in the Garden of Eden... once you get them on board, they're never gonna want to get off.”
“I said, hey AI, send them a calendar invite and before I could even finish, it was like, there I did it and it was for the wrong... time.”
“Ask the AI to ask you questions so that it has the context that it needs.”
“What would've taken me probably two full days of work was about two and a half hours.”
Bring this conversation to your organization. Vance Crowe speaks to conferences, boards, and leadership teams on communication and negotiation and leading change.
Book Vance to speak