From Operator to Builder
I have no coding background. But I do have years of experience building businesses, designing workflows, managing finances, and figuring out what users actually need.
AI has given me a new way to turn that experience into working software. I’m building a custom operating system for COZi Living and exploring how far the same approach can take Karri, my newest venture.
This is my ongoing experiment in becoming an AI-enabled builder — what works, what breaks, and what I’m learning along the way.
I’ve done this before — twice, actually. As a two-time tech startup founder, I worked side by side with developers to launch a prototype. I know what it takes to translate an idea into something real, and how much of that depends on the person sitting next to you who can actually build it.
This time, my developer doesn’t talk back in meetings. It’s AI.
I’m still the founder. But now I’m building alongside a non-verbal developer — one that never gets tired and never needs the idea re-explained, but also can’t read a room, push back over coffee, or tell me when I’m solving the wrong problem. This is where I document what that partnership actually looks like: what frustrates me, what amazes me, what AI still can’t do, and whether a builder-operator with strong domain expertise is best positioned to thrive in this stretch. My working belief is that in this AI era, the ones who roll up their sleeves and bring real domain insight are the ones who can launch the best — and the fastest. But I’m still testing that.
How far can a non-technical founder go when AI makes domain knowledge executable?
This isn’t a learn-to-code tutorial. It’s a real-world experiment involving live businesses — employees, contractors, owners, guests, customers, payments, and everything that goes wrong once real people are on the other end.
What I’m building, in public

Building the software my existing business needs
A custom property-management and operating system, developed from inside a 50+ room hospitality business.
- Reservations and calendars
- Cleaner scheduling and payments
- Owner reporting
- Automation and exception handling

Building an AI-native business from the beginning
An experiment in using AI not only to build software, but to validate, operate, and grow a new luggage-delivery venture.
- Product development
- Operational design
- Service quality and trust
- Building the company while delivering the service