Autonomous agents will change what we call software

Agentic products begin with outcomes and help determine the sequence of work.

I am still learning what autonomous agents will change what we call software means in practice. That is the useful part of building in public: an idea does not have to arrive as a final opinion. It can begin as a question, become a prototype, meet real people, and change shape. I pay attention to the small details because they reveal whether technology is actually helping. Does the interface make the next step clearer? Does the AI explain its limits? Can a person correct it without fighting the product? These questions matter as much as model capability or technical novelty. They keep the work grounded in the people who will use it. In future of work, I am trying to choose clarity over performance, usefulness over theatre, and curiosity over certainty. I expect my thinking to evolve as I build more, read more, and hear from people with different experiences. For now, this article is a record of the reasoning behind the experiment and an invitation for readers to question it with me. I also think about the practical side of the work: performance, accessibility, privacy, and the experience of someone arriving with no context. Search engines may help a reader discover an article, but usefulness is what makes them stay. That means writing specific explanations instead of stuffing a page with repeated keywords. It means naming the tradeoffs, sharing what did not work, and connecting a broad idea to an actual product decision. When I mention AI, I want to explain the human problem around it. When I mention a technology, I want to say what it enabled and what it complicated. This is how I try to make technical writing both discoverable and worth discovering. The web is full of confident claims, so I would rather offer a careful perspective from the work I am actually doing. A reader should be able to understand the idea, see why it matters, and decide whether it applies to their own project. That standard keeps me honest. It also makes the act of writing part of the engineering process: I have to define the problem, describe the constraints, and explain the result clearly enough for someone else to follow. The more I build, the more I value this kind of clarity. It is not less ambitious than chasing novelty. It is what allows a new idea to become useful beyond the person who first imagined it.

I do not think good technology comes from pretending to have every answer. It comes from making a thoughtful first version, observing what happens, and improving the next version with humility. That is the habit I want this portfolio to document.