Building myqi

I am someone with a lot of ideas floating around in my head. For most of my life that is where they stayed. I have never written code, and every idea that needed software needed someone else to build it, so it went on a list.

Like most people, I had been using AI as a chat app. I asked it questions and it answered. Then one day I described one of those ideas to it and it built it. It was a small thing, but it was like a light bulb went off in my head. An idea I had carried around for years existed now, because I said it out loud. That moment changed what I thought I was capable of, and I dove into the rabbit hole to see what else it could do.

I started to think of ways that I could use AI in my life that I never could before.

Making fun side projects

Once I could build things, I built a lot of them. A finance tracker. A collection of our daughter’s photos, each one a trading card the family can collect week by week. A birthday gift for my wife. Each one lived in its own conversation with AI, and every conversation started from nothing. I would explain what the project was, where I had left off, what I had decided and why. The next day, again. An AI that forgets you every morning is a very smart stranger.

Building a brain

So I built a memory. It is a folder of plain text notes. One page per project with its current state, the last thing that happened, and the exact next step. A page for who I am. My family. My goals. Lessons kept apart from facts. When the AI starts, it reads a short summary and a pointer to where I left off, then goes and finds what it needs. I started calling it my second brain, and it is the one thing every project shares.

Over time the brain grew past the projects. My goals, what I wanted to work on in a given week, what my preferences and habits are, all of it ended up in there too. Instead of constantly thinking about it in my head, I dumped it into my second brain.

Then I wondered what all of it added up to

At this point a lot of my life was passing through one system. Every project’s state. Every decision and why. My notes. Months of conversations. And I noticed that none of it was being used for anything except answering the question I happened to ask that day.

So I asked a different kind of question. Not “what should I work on today,” but “what am I actually working toward?” Not this week. In a year.

The AI did not answer that for me. It remembered what I said on Tuesday when I contradicted it on Thursday. That turned out to be the thing I needed, and it was only possible because the brain was there.

myqi is what came out of that

That planning page became a private app on my phone, and it is where myqi lives now. It shows the life I am working toward, the outcome I chose, and every project with its honest status. Behind it sit three things.

• Think with me. A chat that can read my brain and nothing else. Ask what I decided about a project a few weeks ago and it finds the note, quotes it, and says if it is out of date. It cannot change anything.
• Research. Every morning it looks for one real problem worth building for, with strict evidence rules: dated first-hand accounts, an existing workaround people already pay for, the smallest test that would prove anything. If it cannot find that, it says so. It stops when three briefs are waiting for me.
• Cards. When it thinks it has learned something about me, or thinks I should commit to a next step, it puts it on a card. I press Accept, Edit, or Dismiss. Nothing is saved without a button.