The Non-Coder's
Manifesto
20+ products. Zero lines of code. What happens when anyone can build.
lines of code I can write. Not approximately zero. Actually zero. Every product you see in my portfolio was built without writing a single line.
The gatekeepers are gone
For decades, building meant code. Code meant engineers. Engineers meant budgets, timelines, and permission. The barrier to entry was so high that ideas died in slide decks. Strategy documents gathered dust because nobody could turn them into products. That era is over. AI didn't lower the barrier. It removed it entirely.
I'm not technical. That's the point.
Every product in my portfolio was built by describing what I wanted in plain English to Claude Code. No frameworks, no dependencies, no Stack Overflow. The skill isn't coding. It's knowing what to build, why it matters, and how it should feel. Fifteen years of strategy gave me that. AI gave me the tools to act on it.
live products. CultureTerminal, The Pattern, Modern Retro, Trove, Curio, Oishii London, Little London, Pub Guide, First Out, Taste OS, The Visible Shelf, and more. All shipped. All live. All built by a non-coder.
Strategy used to end at the deck
I spent fifteen years writing strategy documents that other people built. Sometimes brilliantly, sometimes badly, always slowly. The gap between the idea and the execution was where good strategies went to die. Now I build the prototype before I write the strategy. The product IS the strategy. The deck is just the footnote.
The 9pm to midnight window
Everything gets built between 9pm and midnight. After my son's bedtime, after the house goes quiet. Three hours. That constraint forces ruthless editing. No unnecessary features. No scope creep. Ship tonight or it doesn't exist. The constraint is the creative advantage.
Taste is the skill that matters now
When everyone can build, what separates the good from the forgettable is taste. Knowing what to keep and what to kill. Knowing when something feels right. AI can generate infinite options. The non-coder's job is to be the editor, not the author. Curation is the new creation.
The best product people won't be the ones who can code. They'll be the ones who know what's worth building.
What a non-coder builds differently
I think in experiences, not functions. In stories, not systems. In feelings, not features. That's not a limitation. It's a perspective engineers don't have. I approach every product as a user first. What does it feel like to land on this page? What's the first thing someone notices? Does it make you want to stay? Code serves the experience. The experience serves the person.
AI is my co-founder
I describe what I want. AI writes the code. I review the result. We iterate. It's a conversation, not a command. The best outcomes happen when I bring the vision and the taste, and AI brings the execution and the patience. Neither of us could do this alone. Together we ship faster than any team I've ever managed.
per month. That's the total cost of running 24 live websites, 7 automated pipelines, and a full product studio. The economics of building have collapsed. The only barrier left is ambition.
The portfolio is the proof
I don't have a CS degree. I don't have a GitHub contribution graph. What I have is 20+ live URLs that you can visit right now. Products that people actually use. Sites that run themselves. Automated pipelines that publish daily. The proof isn't in the code. It's in the output.
Every product taught me something a course never could
CultureTerminal taught me data pipelines. The Pattern taught me editorial automation. Modern Retro taught me image generation at scale. Trove taught me database design. First Out taught me data visualisation. None of these were planned learning exercises. They were things I wanted to exist. The education was the side effect.
The barrier was never code. It was permission. AI removed the gatekeepers, and the gatekeepers didn't even notice.
Why strategists should build
A strategist who can build their own prototype will outperform one who can only describe it. Every time. Because building forces you to make decisions that decks let you defer. Which features actually matter? What can you cut? What does the user really need? Strategy through building is faster, more honest, and more useful than strategy through slides.
The agency model is breaking
The old model needed fourteen people to ship a campaign microsite. Strategist, project manager, UX designer, UI designer, front-end dev, back-end dev, QA, producer, account director, and five people in meetings about it. I ship the same output in a night. The economics don't lie. One person with AI and taste is a viable creative studio.
Non-coders see what engineers miss
An engineer looks at a problem and thinks about architecture, scalability, and technical debt. A non-coder looks at the same problem and thinks about the person using it. Both perspectives matter. But in a world where AI handles the architecture, the human perspective becomes the differentiator. Build for feelings first. The code will follow.
days a year The Pattern publishes. Fully automated. No manual intervention. Built by someone who can't write a for loop. The system runs itself because I designed the experience, not the engineering.
Ship, don't spec
The fastest way to validate an idea is to build it. Not to write a brief about it. Not to run a workshop about it. Not to schedule a meeting about it. Build it. Tonight. If it works, iterate. If it doesn't, you've lost three hours instead of three months. The non-coder's advantage is the willingness to be wrong quickly.
The tools are ready. Are you?
Claude Code, Cursor, Replit, v0. The tools for non-coders to build real products exist right now. They're not toys. They're not demos. They're production-grade. The only question is whether you have something worth building. That's a taste question, not a technology question.
What I'd tell every strategist
Stop waiting for engineers to build your ideas. Stop writing decks about products you'll never ship. Stop treating 'I'm not technical' as an excuse. It's 2026. The tools exist. The cost is negligible. The only thing standing between your strategy and a live product is the decision to start. Open a terminal. Describe what you want. Ship it tonight.
The best things I've built started as sentences, not specifications.
The compound effect
Twenty-five products in four months. Each one builds on the last. The design system from one becomes the template for the next. The data pipeline from CultureTerminal powers The Pattern. The editorial voice from the blog shapes the decks. Nothing is wasted. Everything compounds. A non-coder with range and taste builds a flywheel that accelerates with every project.
slides. Zero code. This deck was built the same way as everything else. Plain English to Claude Code. Shipped in under an hour. The medium is the message.
There's a quality ceiling. AI-assisted building gets you to 80% fast, but the last 20% still requires understanding what's happening under the surface. And 'vibe coding' at scale creates maintenance debt that compounds invisibly until something breaks.
The barrier to starting has disappeared. The barrier to building well hasn't.
The Non-Coder's Manifesto says the barrier to building is gone. Permission Not Required says the barrier to starting was never technical. It was cultural. One removes the tooling excuse. The other removes the permission excuse.
Code is becoming a commodity. Taste is not. The future belongs to non-technical builders with something to say.
I can't write a single line of code. I've shipped more products this year than most engineering teams.
