About
I want to build things. That's the clearest sentence I can write about myself, and even that one I'm not sure about. Maybe it's more accurate to say I want to have built things. The desire is for the finished object, not always the building. But the only way to have built something is to build it, so here I am.
What have I actually built? Not much yet. Apollo AI, a crop stress detection tool that won a hackathon. My own website, which I keep rebuilding because the previous version never feels right. I'm on my twentieth version, which is either persistence or inability to ship. Probably both.
Right now I'm working through what I call 101 Projects: building TCP, DNS, HTTP, encryption from scratch. The idea is simple. I don't understand something until I've built it myself. Feynman had a line about this. "What I cannot create, I do not understand." I used to think I understood how the internet worked. Then I tried to implement a DNS resolver and realized I understood nothing.
I also run R&D at Agora, a writing lab. The premise there is that writing is thinking made visible. You don't know what you believe until you try to write it down. I'm testing that premise on myself. These paragraphs are the test.
I read a lot. My bookshelf has a pattern I didn't notice until recently: Taleb, Popper, Kahneman, Feynman. Different fields, but they're all asking the same question. How do you act when you don't know? Which is most of the time.
Taleb taught me about skin in the game. Don't tell me what to do if you don't pay the price when you're wrong. Don't give advice you wouldn't take yourself. I try to follow this, though I probably fail more than I notice.
Popper taught me that knowledge grows by falsification, not confirmation. You don't prove theories true; you fail to prove them false. This changed how I think about my own beliefs. I hold them lighter now.
From all of them I got the same lesson: plans collapse on contact with reality. The world is too uncertain to predict. So instead of trying to know the future, I try to position myself so I don't need to know it. Small experiments. Fail cheap. Double down when something works.
I don't like labels. When someone identifies strongly with a title (entrepreneur, writer, philosopher) I get suspicious. Usually it means they're performing the role instead of doing the work. I'd rather be wrong about this than adopt labels myself.
I have no respect for authority. Not as rebellion, just as epistemics. Someone's title or reputation tells me nothing about whether their ideas are correct. Forget who said it. Look at the argument. Does it start somewhere reasonable? Does it end somewhere reasonable? Is the path between them sound? That's all that matters.
My favorite days are the ones where I forget to eat lunch because I'm building something. I call it work, but really it's play. The line between them dissolves when you're doing something you chose.
I keep a graveyard page on my site for dead projects. Ideas that didn't work. Things I started and abandoned. This might seem like advertising failure, but I think it's honest. Most things don't work. If you're not killing projects, you're not starting enough of them.
I'm twenty years old, which means everything I've written here will probably embarrass me in five years. That's fine.
If you're figuring things out too, maybe we should talk.