About Me
I'm deeply interested in entrepreneurship, product management, system architecture, operations, UI/UX, and design.
I enjoy working end-to-end. From understanding user problems to designing workflows and shaping systems that scale.
I prefer shipping prototypes over writing long documents, and I learn fastest by building real things and iterating based on feedback.
Operating Principle
“I believe good products come from understanding real workflows, not just building features.”
100% with AI. Since 2022.
When ChatGPT launched in November 2022, I made a decision that shaped everything after it. I would build 100% with AI. Not as a side experiment. As my primary method.
Every project on this site was built that way. From the first one in 2023 to the company I am running right now. I do not write code by hand. I describe what I want, and I work with AI until it exists. I have been doing this for over three years.
"It works. Unreasonably well. A single person operating this way can ship at the speed of a full engineering team."
I have built hardware prototypes, delivery MVPs, browser concepts, an AI operating system, gym platforms, cafe systems, a real estate app, a gamified learning product, and an autonomous betting agent. All solo. All with AI.
But the more I built, the more I noticed a pattern. The AI was never the bottleneck. I was. Specifically, the way I talked to it.
Every project started the same way. I would describe what I wanted. The AI would build something close but wrong. I would spend hours correcting assumptions I did not know I was making. The model did not lack capability. It lacked context. It lacked boundaries. It lacked the things a domain expert would have said out loud before writing a single line of code.
That is the real problem with AI right now. Not that it cannot build. It can. The problem is that the person asking does not know what they do not know. And the AI is not going to tell them. The gap between "I have an idea" and "the AI built exactly what I wanted" is not a knowledge gap. It is a specification gap.
So I built Kosmo. A prompt compiler.
You describe what you want in plain English. Kosmo asks the questions a domain expert would ask. It checks current official documentation. It injects guardrails, stopping conditions, and hidden reasoning loops. Then it outputs a structured, tool-aware prompt specifically adapted for the AI you are using, whether that is Claude, Gemini, Cursor, or ChatGPT.
It is not another AI wrapper. It is the layer that should have existed from the start. The layer between the person with the idea and the tool that builds it.
I have been on both sides of this problem longer than almost anyone. I know what it feels like when the prompt is wrong because I have felt it hundreds of times across dozens of projects over three years. Kosmo is not a theory. It is the solution to a problem I lived.
Discover Kosmo