Lucio Santana

@helloiamleonie he loved it. printed result turned out really cool!

@luciocs nice. did your son enjoy the book?

just curious: what’s the most useful thing your OpenClaw, Hermes Agent, etc. is doing for you?

@chamath It's not easy to go forward with meaningful AI applications cause their not where hype is. But I'll keep trying anyway. Check out paracasainclusivo.com.br

AI doomerism exists, in part, because solutions to our most important problems still remain unsolved while examples of superficial process automation abound. We’ve lost the script. If AI wants to be celebrated, it should put solutions to some of society’s greatest problems at the front of the line. Case in point…why hasn’t anyone focused on an AI for fixing the growing illiteracy of our children?

@AdolfoUsier All the agent required was a logged session to the print-on-demand platform.

@luciocs bro the agent created the actual product thats wild — what did the prompt look like

My OpenClaw AI agent (Friday) couldn't find the magic potions book my son wanted. So it created one and published it on a print-on-demand platform. The copy arrived today. He's 12. He loves Harry Potter. He's speechless.

@AdolfoUsier No prompt. Just my kid texting what we wanted it to look like. A few mockups and feedbacks. And agent did all the rest.

Passando rapidinho aqui para compartilhar uma pequena alegria. Ao fazer minha declaração de IR, tinha no meio receita de royalty do meu livro A Era do Impacto. Agora me sinto oficialmente um escritor 😃 📚

For anyone starting out: don't try to replicate everything at once. Start with identity (CLAUDE.md). Build memory gradually. Connect your tools. Automate what's recurring. The best version of your setup is the one 30 days from now — not today's.

The evolution came in layers: Weeks 1–2: identity and memory. Weeks 3–4: first skills and connections. Month 2: automations, expanded memory. Month 3: full autonomy, team skills, browser prototyping. Each built on the last: memory enabled depth, depth trust, trust delegation.

The most important thing I learned building my AI agent setup isn't technical. It's a mindset shift.

I don't know which tool "wins." Maybe neither of the current ones. But the direction is clear: agents operating with increasing autonomy, in continuous and uninterrupted cycles. The question isn't if this happens. It's how you'll be ready when it does. #AIAgents #OpenClaw

One thing is becoming clearer: this is an infrastructure question. Continuous autonomous execution consumes resources. Running it on your work machine creates bottlenecks. Dedicated hardware or cloud for agents isn't a luxury — it's a necessity when the operation scales.

People keep asking: why use two agent frameworks? Wouldn't one be enough? Honest answer: I'm using both to test, experiment, and learn. But there's a real difference that matters.