Lenny's Podcast

From skeptic to true believer: How OpenClaw changed my life | Claire Vo

with Claire Vo
29 Mar 2026 22 min read 1h 18m
TL;DR

Claire Vo transformed from a skeptic who lost her family calendar on day one to running nine OpenClaw agents across three machines because she found genuine product-market fit once she stopped expecting a single agent to solve everything. The key unlock is treating OpenClaw agents like you'd onboard a real employee: give them their own account, progressive access, and specialized roles rather than dumping all tasks into one system.

Key Moments
Claire Vo
“My first install, I truly spent 8 hours getting OpenClaw up and running. In return for those 8 hours, I got my personal family calendar deleted.”
Claire describing her initial skepticism and the mistake that almost turned her away from OpenClaw entirely
▶ 0:03
Claire Vo
“It just hit me with enough joy and enough utility when it wasn't deleting my calendar that I knew something was there.”
Claire explaining how she recognized genuine product-market fit despite the calendar disaster
▶ 10:34
Claire Vo
“Where people stumble with OpenClaw is they think they can throw any task at a single agent and get great results. And then they get really frustrated. I won't sugarcoat it. It's a pain to set up. It is not hands-off, but the value is so high. I am willing to go through the pain.”
Claire identifying the most common failure mode and why persistence pays off
▶ 0:42
Claire Vo
“I don't feel like I'm using Claude. I feel like I'm using Polly. I feel like I'm using Finn. I feel like I'm using these things that I built and that sense of crafting your personal agent experience as opposed to giving your tasks to a general purpose agent, I think just changes the the user agent interaction in a very interesting way.”
Claire explaining why building and naming individual agents creates psychological ownership and better outcomes
▶ 16:37
Claire Vo
“You would set it up its own local admin account. And then you know, going back to what I was using it for, personal assistant, executive assistant. And I have had as an executive executive assistant, so I know how to onboard them. And you don't onboard your EA by giving the password to your email account.”
Claire detailing the security model for setting up OpenClaw based on how she'd hire a real employee
▶ 18:57
About Claire Vo
›
Claire Vo is the host of the sister podcast How I AI and a longtime engineer who has served as chief product officer three times and founded her own AI startup. She's become a power user of OpenClaw, running nine instances across three computers, despite starting as a skeptic. Her pragmatic, technical approach to evaluating AI products has made her insights on agentic systems particularly valuable.
Takeaways
1
Specialized agents beat generalist agents Claire runs eight different agents (Polly for scheduling, Sam for sales, Q for kids' homework) rather than one catch-all system. This specialization dramatically improves results because each agent can be deeply optimized for its specific domain and won't get confused trying to handle unrelated tasks. Most users fail by throwing everything at a single agent and expecting great results.
2
Treat OpenClaw like hiring a real employee Rather than giving OpenClaw admin access to your main machine and all passwords, set it up on a separate computer with its own email account, calendar access, and progressively expanded permissions. This mental model—starting with read-only access to calendar, then delegated email, then draft permissions—matches how you'd actually onboard an executive assistant and dramatically reduces security and operational risks.
3
Use good models and embrace the setup friction Claire uses Claude Opus, Sonnet, and GPT-4 rather than cheaper models because they're hardened against prompt injection and deliver better results. The setup takes hours and isn't hands-off, but recognizing the real value upfront makes the friction feel worthwhile rather than a barrier to entry.