#1 From Black Box to Playground

Three weeks ago AI felt like a black box, big and intimidating and I had no idea where to start.

What I knew was that I learn by doing, not by watching and not by reading, but by getting my hands on the thing and finding out what happens. The cost of starting with Claude Code was high before the return was visible and it needed trust before it gave anything back. I needed to put the time in before I even knew what I was building toward.

With some great help from people around me it started to move from black box to playground, a playground with endless possibilities and people way smarter than me already figuring out how to use it.

But honestly, I got overwhelmed. I watched everything I came across and started replicating what others shared openly to learn more.

My social media algorithm got crazy and became both very helpful(!) and very loud by showing me everything everyone was building and all the things I could copy with one command.

The playground got bigger and noisier every day and somewhere underneath the noise was a question I couldn't quite name yet.

What were they seeing that I couldn't see yet? Or was everyone just trying like me but screaming very loud?

Outside in or?

I really tried the things.

Teresa Torres' task manager, I built it from scratch, but I couldn't use it. It didn't fit my brain. Zevi's workflow as well, very smart and well-designed but still not mine.

Every system I tried had the same problem, it was built for someone else's way of thinking and working and I kept trying to fit myself into it.

And slowly I started to realise I wasn't missing out on anything. I was just at the beginning of learning how to build around my way of thinking instead of trying to fit into someone else's.

That is also when I learned about Jevon's Paradox, that when things get easier you do more not less and more skills don't mean more value, they usually mean more noise.

The trap was that I was building the wrong things by copying instead of creating, trying to fit in instead of building from the inside out.

The Shift

So I stopped and I started to ask different questions.

Not what are the smartest people building and how do I replicate it but what does my brain actually need and what would a system look like if it started there?

Building from the inside out

The answer was to build my own operating system and agent from scratch. Something that would start with how I actually think instead of asking me to adapt to how it worked. I called her Pippi because I had never done it before so I should be able to do it.

The first thing I did was not to build commands or automations or workflows but asking Claude to understand me, my goals, my strengths, my struggles, how I work, what drains me and what excites me. I built the brain before I built anything else.

And then something shifted. The system started to feel like something I was building around what I already was, not something new I had to adapt to, not a template I was trying to fit, but something that reflected how I actually think and built structure around it.

Pippi does the research now, not me finding what smart people are doing, but Pippi finding what's relevant for my goals, my thinking, my way of working and bringing it to me. We add what I need and we cut what I don't and it evolves with me.

This is my bumpy road from overwhelmed by noise to building exactly what I actually need. And it didn't start with the right system.

It started with the right questions.