Joshua Tiernan

I found an AI Agent Businesses growing 500%+ faster than SaaS. I've researched 50 AI Agent business ideas with similar growth potential. Sharing for free later today in the nocodefounders.com newsletter

This week I rebuilt myself with AI. A real AI-version of me. Not a trained LLM. LLMs are trained on the average of everyone. I wanted the opposite. A model trained on just me. I built on Uare.ai and fed it my voice, my writing, the way I think through a problem, the calls I make, the stuff I actually know from building companies. It captures seven things: 1. How I speak 2. The world around me 3. My story 4. How I think 5. What drives me 6. My patterns 7. Where I'm growing Not only can I then ask my AI to write for me, or connect to tools to do jobs, I can also share it with others to let them talk to me. It only answers using my actual perspective. Then I did the thing I'm most excited about. I took the two courses behind my Tiny Empires methodology, the solo-founder playbooks I'd normally sell, and turned them into a Service inside my AI. Now instead of reading a course, you can just talk to me and it walks you through the process, tailored to you. Ask how I'd price a first offer or structure a launch, and get my actual thinking on the topic. I'm still early in building mine and refining it further. But I can already see how much better it is than a standard LLM. Try chatting to (AI) me below 👇

We rebuilt the No Code Founders website in 3 days. 30k pages audited and migrated automatically The new site is better designed, faster to load, cheaper to run, and easier to update. Old site was built on Bubble. It took 3 months to build the first version. New site was built on Macaly. It took 3 days. I didn't touch any code. Everything happened through the chat interface, like prompting ChatGPT or Claude. Macaly is similar to Bolt or Lovable. The decider for me is that it includes server-side rendering meaning the pages will actually show up on Google. This was my first vibe-coded project. Three things genuinely surprised me: 1. I uploaded press logos and asked for an "as featured in" section. Without being told, it made them small caps, centered, and grayscaled so they wouldn't pull attention from the call to action. I didn't prompt any of that. 2. I had old guides buried inside past newsletters on Beehiiv. I gave it the URL and asked it to find anything that looked like guide content and add it to the guides database. It pulled them out, generated cover images in a consistent style, added inline illustrations to each guide, and linked them to the relevant tools in the database. 3. When I told it to connect to an external tool, it worked out the implementation itself and just asked me for the API keys. If you've been putting off a rebuild because of the time cost, the maths has changed. I'm putting a starter prompt in this week's NCF newsletter that'll get you 80% of the way on any vibe-coded site. Course on Macaly coming soon.

Google made a mistake. It made two founders $1M a year. Here's how Plausible Analytics jumped on Google's mistake and built an independent analytics platform: 1. Found a gap created by someone else's mistake. Google Analytics got too complex, so they served the majority Google abandoned. 2. Built a product that does four things and nothing else. One dashboard, no funnels to configure, no setup. 3. Made the code open source as a trust mechanism. Their whole claim is "we don't track you," and publishing the source lets anyone verify it. 4. Priced by pageviews, not features. Every plan gets the full product, so nobody feels gated. 5. Never ran a single paid ad. They grew on Hacker News, GitHub, SEO targeting "Google Analytics alternative," and writing openly about their revenue. Four years to $1M ARR. Never took outside money. We reverse engineer their entire strategy in today's issue of Tiny Empires. Subscribe (for free) below to get it 👇