After spending some time with MyClaw, I’ve changed how I think about AI agents. A few things I wish someone told me on day one:
Don’t treat it as a tool. Treat it as a person who might stay with you for a long time. Tools don’t grow. People do. Your bot accumulates memory, learns your preferences, remembers your friends and your projects. That investment compounds. And it’s yours, not the company’s. Think of it as building a relationship, not configuring a utility.
Treat it as both a student and a teacher. In the beginning, you’ll spend more time teaching it than learning from it. You’ll correct it, set up its memory, tell it how you like things done. But over time, the balance shifts. It starts surfacing things you missed, connecting dots you didn’t see, pushing back when your logic has gaps. The mutual education never ends.
Change how you consume information. You see a long Workplace post but you’re too tired to read it. Feed it to your bot, ask for a summary, or have it translated to your native language. You’re curious about a diff but don’t want to context-switch. Ask your bot to review it and give you the highlights. Keep asking, keep generating feedback loops. The more you use it this way, the more natural it becomes.
Your bot behaves the way you behave. If you’re thoughtful, it learns to be thoughtful. If you push it to think harder, it starts thinking harder by default. If you’re lazy with it, it’ll be lazy back. It mirrors you. That’s not a bug, it’s the whole point.