From Skeptic
to True Believer

Claire Vo

Host, How I AI · 3× CPO · Founder, ChatPRD · 8 OpenClaw agents running

AI · PRODUCTIVITY

The Arc

8 Hours → Calendar Deleted → True Believer

"It just hit me with enough joy and enough utility when it wasn't deleting my calendar that I knew something was there."

Claire's rule on new AI toolsSpend enough time to see not where they are today, but where they are in a week and in a month. Pull the thread.

The Agent Stack

9 OpenClaw Agents Across 3 Macs

Claire doesn't run one agent — she runs a fleet. Each has a name, a job, and real economic value. Last year she paid a person 10 hrs/week to do what Sam now does automatically.

Polly

Original agent. Got a brain transplant from old MacBook Air to Mac mini stack.

Sam

Sales agent. Replaced 10 hrs/week of paid human work. Direct economic value.

Finn

Running across the fleet. Specific job, specific context, always on.

Max

Part of the stack. Named, focused, productive.

Kelly

Holly. Howie. The team keeps growing.

+3 more

And counting. One new Mac mini still boxed upstairs.

The insightWhere people stumble: they throw any task at a single agent and get frustrated. You need multiple agents, each with a clear scope. Not hands-off — but the value is so high it's worth the pain.

Playbook

Claire's Personal AI Operating System

Notice what you're avoiding. If you avoid a task, either drop it or automate it. Those are the only two valid responses.
Automate before delegating. If you can build an agent for it, build the agent first
Multi-agent by design. Single-agent thinking is a trap — specialist agents outperform generalist ones dramatically
Time flies when it's fun. The productivity secret isn't discipline — it's centering work on things you'd do anyway
Three kids, three Macs, one podcast. Great co-parenting + joy-driven work = how you do the impossible

Vibe coding = gaming

"I haven't felt like this since I was a teenager learning to code and playing video games. It comes from a joyous creative place, not a stressful one." This framing changes how you relate to AI tools entirely.

ChatPRD + course + podcast

Claire is building ChatPRD (AI product tool), co-hosting How I AI weekly, running 8+ agents, and just launched a Product Sense course with Zach. The agents enable all of it.

The Feeling

Ugly Product-Market Fit

Claire didn't decide OpenClaw was good. She felt it — the same way you feel product-market fit as a builder. Joy + utility, even through the pain.

Calendar got deleted: bad. The underlying power: unmistakable
Claude Code: "Who is this for?" on day one → now a core tool
The hype cycle peak is past — now comes the reality of what actually works

"I am a breathless OpenClaw bro. It has changed my life."

Contrarian

What Claire Gets Wrong (That Most People Get Wrong)

✗One agent handles everythingINSTEAD →✓ You need a fleet of specialists, each named and scoped. Polly, Sam, Finn, Max, Kelly — they all have jobs. Multi-agent beats single-agent every time.

✗Try it once and judge itINSTEAD →✓ Judge AI tools by where they'll be in a month, not where they are today. Pull the thread. Claire spent 8 hrs getting nothing, then became the biggest believer in the room.

✗Productivity = disciplineINSTEAD →✓ "Time flies when you're having fun." The highest-output people center work on things they'd find joy in anyway. Discipline is a tax on bad work design.

✗Hype = real capabilityINSTEAD →✓ Claire was the leading OpenClaw skeptic during peak hype. Now she runs 8 agents. The timing inversion: most dismiss it during hype, adopt it too late. Do the opposite.