Introduction

AI systems can now write usable code from plain-spoken language. Not demos, but software that runs, can be modified, and improves over time. What required venture-backed engineering teams until now is rapidly becoming available to anyone looking in the right places.

I’m not an engineer by training. But I have led and product managed software projects alongside various startup teams for 10+ years. All the while, the way I managed roadmaps, spoke with users, and influenced what we built was in the context of a constant … code is expensive. I remember interviewing at a San Francisco startup in the early 2010s. While sitting down in a general seating area I saw a sign that said, “shh engineers are working.” Whatever your thoughts on the implied hierarchy, there was alignment.

When I first started coding with natural language I panicked. Imagine a big part of your job was to skillfully manage a scarce resource. And all of a sudden, that resource is wildly abundant. But the more I thought about it, the more excited I got. Today I’m convinced that we’re on the verge of a golden age for solopreneurs.

Cheap code changes what's worth building. Software has historically needed massive returns to justify its cost. Countless worthy problems identified by entrepreneurs have died in the graveyard of "not venture-scale." In other words, where ideas not likely enough to be worth a billion dollars went. All of a sudden, a long tail of projects have new ROI math to consider.

It also changes what gets built well. Without billion-dollar requirements, builders can target people like themselves, and that deep problem understanding is the essential ingredient for building software people love.

Put it all together, and what used to be disadvantages for solo builders are now exponential force multipliers. One person pursuing a familiar problem and building a solution in days will almost always outpace a team with the same mandate. Why? No consensus process, no roadmap theater, no organizational inertia. Coordination takes skill, trust, and time.

Today, the speed at which someone can move while building something they deeply understand is mindblowing. It will result in some of the best software ever built. A wave has started to form. People investing in the tools behind this dramatic shift will be the first ones to ride it.

Natural language coding with AI is not without new challenges.

If you've built with tools like Lovable, Claude Code, or Bolt, you've felt it. Things that seem simple eventually become fragile. Small changes break working systems. Progress feels fast, until it suddenly isn't. You can build more than ever, but it's also easier to lose control.

When I started building with these tools I ran into robustness issues almost immediately. Not because the tools were weak, but because I didn't have a way of working that complimented natural-language coding’s strengths and weaknesses. What became clear was simple:

"

Natural-language coding doesn't remove the need for rigor. It changes where rigor belongs.

The challenge is no longer writing code. It's directing it.

The Jetty Method is a practical framework for building real, maintainable software without an engineering background. The goal isn't to turn you into an engineer. It's to help you think clearly, build intentionally, and stay in control.

The Jetty Method is for people who want to build things for a big audience. Not people who want to talk about building, but people with a problem worth solving.

You might be a founder without a technical co-founder. A product manager tired of waiting for engineering. A solo entrepreneur racing the runway. What you have in common is that you have something at stake, and you want to move fast while building a strong foundation.

This guide isn't so much for professional engineers. You have workflows that serve you. It isn't for people who enjoy the chaos of prompting an AI. It isn't for those seeking no-code simplicity, because natural language coding is still coding. And it isn't for hobbyists with nothing on the line, because the structure won't feel worth it when the stakes are low.

The Jetty Method isn't complicated. Most of it becomes second nature quickly. But it does require following the plan rather than winging it. That's the trade: a bit of structure upfront for software that doesn't fall apart when you try to change it. The method exists because building without a plan feels fast until it isn't.

Let’s get to work.