Search for a command to run...

Timestamps are as accurate as they can be but may be slightly off. We encourage you to listen to the full context.
In this engaging episode, Pat Walls, founder of Starter Story, walks through his experience of building a YouTube alternative using Claude Code after receiving strikes from YouTube that threatened his business. (07:00) What began as a weekend panic project turned into a powerful demonstration of how AI coding tools can help entrepreneurs rapidly prototype solutions to business threats. Pat demonstrates step-by-step how he used Claude Code to recreate core YouTube functionality, deploy it live, and ultimately shift his business model from YouTube-dependent to a hub-and-spoke approach where his own platform becomes the central hub.
Pat Walls is the founder of Starter Story, a media and education company that helps entrepreneurs learn how real businesses get built. Starter Story generates over $2 million annually through subscriptions, courses, and content, with its YouTube channel reaching millions of viewers each month. A former software engineer turned creator, Pat is known for experimenting publicly with new tools and recently used AI to rebuild his own video platform when YouTube put his business at risk.
Andrew Warner is the founder of Mixergy, where he interviews entrepreneurs about how they built their businesses. He has conducted thousands of interviews with successful entrepreneurs and is known for his direct questioning style and ability to extract actionable insights from business founders.
Pat demonstrates that you don't need to be a coding expert to build sophisticated applications when you treat AI as a collaborative partner. (09:00) He emphasizes approaching AI coding with humility, saying "please" and "thank you," and working iteratively rather than expecting perfection immediately. The key insight is that successful AI development requires thinking like a product manager - breaking down complex features into simple, single-purpose requests. This approach allows non-technical entrepreneurs to rapidly prototype solutions that would traditionally require months of development and significant technical expertise.
Rather than reinventing complex infrastructure, Pat strategically chose to build on established frameworks like Next.js and services like Bunny.net for video streaming. (22:00) His philosophy is clear: "Don't reinvent the wheel." This approach allowed him to focus on the unique value proposition of his platform rather than getting bogged down in technical details like video compression and encoding. Smart entrepreneurs leverage existing solutions and APIs to accelerate development while maintaining professional quality and reliability.
Pat's YouTube strikes became a catalyst for a fundamental business model shift from platform dependency to a hub-and-spoke approach. (48:00) He realized that treating external platforms as the hub made his business vulnerable to arbitrary policy changes. By building his own platform as the central hub, with YouTube, LinkedIn, and other channels as spokes, he gained control over his audience relationship and content distribution. This mindset shift transforms potential threats into opportunities for greater business resilience and customer ownership.
Pat consistently uses screenshots and visual references to communicate his vision to Claude Code, rather than relying solely on text descriptions. (16:00) He takes screenshots of existing interfaces, searches for design inspiration online, and pastes visual examples directly into his AI prompts. This technique bridges the gap between human creative vision and AI execution, ensuring that the end result matches the intended user experience. It's particularly powerful for non-technical users who can visualize what they want but struggle to articulate technical requirements.
Instead of asking AI to build everything at once, Pat focuses on implementing one feature at a time, testing it, and then moving to the next enhancement. (15:00) He starts with the core functionality (video playback) before adding secondary features like chapters and timelines. This approach prevents overwhelming the AI with complex requirements and allows for easier debugging when issues arise. Each iteration builds confidence and understanding, making subsequent features easier to implement and customize.