I don’t know about you, but for me, AI has been very helpful these last few years.
It’s helped me do things faster - especially when it comes to analysing, creating docs, speeding up research. It genuinely made a difference.
But somehow, it still felt far away from what it could be. From what people keep telling us it will be. Something was missing. I was faster, sure. But I was still sitting behind my computer. Still doing the things behind my desk.
But that’s not the life I want to live.
At no point did I ever do a vision board at the beginning of the year and write down “I want to spend X amount of hours in front of my PC.” My vision of life is that we are in flow. That we are creative. That we are spending time with friends and family, that we are taking care of our health. Not sitting behind a chair.
So for a long time, I kept waiting. Waiting for the one tool that could literally do anything. And if you’re following the AI world right now, a lot of people are talking about Claude Cowork - and yes, it can do a lot. I’m using it. But I’m still sitting at my computer when I do.
And then, I discovered OpenClaw. And I feel like I’m getting my time back.
Let me explain how.
I’m not using a tool that lets me “build AI agents” in some complicated interface. I just talk to it on Telegram. I record a voice note. Like right now - I’m literally talking out loud and dictating this newsletter. That’s it.
I have one core agent. I call him VAI - short for Virtual Assistant AI. When VAI receives a task, he figures out what needs to happen and dispatches it to the other agents I’ve built around him.
Let me give you some concrete examples.
Sales calls. When I do a sales call, I used to spend three to four minutes afterwards going into my Excel spreadsheet, logging the lead info, writing down the key takeaways from the call, figuring out how to follow up. Then I’d go to my inbox, pull up a template, and draft the follow-up email. It was fine. It worked. But it was me doing it.
Now? I send a one-minute voice note about how the call went. My agent automatically goes into Google Sheets, updates the lead information, and then sends me - via email - a perfectly drafted follow-up for that specific lead, ready for me to review and send.
I could go a step further and authorize it to send the email on its own. I’m not there yet. But I might.
Client deliverables. I create a lot of deliverables for clients - pitch deck reviews, outreach messages, email campaigns. They all live in Google Docs, they all have formatting rules, and they take time to get right.
Before AI: would take me three hours.
With ChatGPT - where I had specific prompts, copied and pasted things one by one, then went into Google Doc and formatted everything myself: took me thirty minutes.
Now: I give one instruction to my agent. Three to four minutes later, I get an email with links to two perfectly formatted Google Docs, ready to send to the client. Three minutes. That’s it.
And it doesn’t stop at work.
Some friends of mine have a shared agent that lives in a Telegram group. It takes care of all of their admin, to-dos, groceries - they collaborate through it together.
I personally have another agent that specialises in writing the newsletter I send out for work. I just dictate out loud what I want to write. It knows my tone of voice. And it creates newsletters that are, genuinely, like I wrote them myself. Word for word, my voice.
The list goes on. And it keeps growing.
I’ve shown this setup to a lot of friends recently who have been mind-blown by the setup.
So this Saturday, I’m going to do a 30-minute live session where I’ll walk you through everything - the agents I’ve built, what each one does, the tools I use, how they’re all connected, and how you can start building something like this for yourself too.
This is just the beginning. The world is changing fast. And AI agents are here, finally.
Come see what that actually looks like.
Register here → https://luma.com/innuewet
See you Saturday,
Gui