By playing, you agree to uListen's Terms and Privacy Policy and YouTube's Terms of Service.

The Calum Johnson Show

The Million Dollar Vibe Coder: The Fastest Way To Get Rich In The New Era of Al (No Code!)

Timestamps About Riley Brown Riley Brown is a content creator and AI educator who went from marketing analyst to one of the most influential voices in the AI creator space.

Riley BrownguestCalum Johnsonhost

May 19, 20251h 6mWatch on YouTube

Summaries, chapters, transcripts, and FAQs on this page are made by uListen, not YouTube.

Great conversations run long. Their best ideas don't have to.

uListen is free on iPhone, iPad, Android, Chrome, and Claude.

CHAPTERS

Jump to publisher timestamps

Chapter summaries

Why Vibe Coders Skip the Playbook
Riley Brown argues that with all of human knowledge in an AI model, you don't need to follow rules or structure, pointing to LevelsIO's income. Calum Johnson then introduces Riley as a non-engineer who raised $9 million for his Vibe Code app and frames the episode's goal.
- •Riley says he follows no playbook and the most successful people he sees have no structure
- •LevelsIO is cited as making around $300K a month
- •Riley raised $9 million for his AI app, Vibe Code
- •The episode aims to show the fastest way to launch an AI business and get feedback in hours
What Vibe Coding Is and Why Obsession Matters
Calum asks what value a person with saved-up business ideas can take from the conversation, and Riley says anything doable on a computer is possible with about a month of learning AI. He then explains vibe coding as AI generating code, so fast iteration on ideas becomes possible.
- •Riley urges obsession with AI and warns that some people sell snake oil in the space
- •Traditionally an app required hand-writing and debugging many code files
- •AI generating code is, in Riley's view, its best skill
- •Critical thinking and spotting problems are the key skills Riley names for the AI age
Fail Fast: Launch Quickly and Take Many Shots
Calum shares that slow launches killed his early businesses, and Riley passionately argues for picking a path, failing fast and not clinging to one identity. He describes his own path from reselling to content creation to building a startup with top engineers.
- •Calum says wanting everything perfect delayed his launches
- •Riley says to just pick one path and find out quickly if an idea is stupid
- •He went from a reselling business to content creation to starting companies
- •Riley says his team's engineers prefer building to hand-coding and favor a bias toward action
Claude Artifacts and the Idea of Personalized Software
After a sponsor break, Riley shares his screen and shows Claude Artifacts, the first tool where he saw AI code render live. He builds a simple calculator and edits it by prompt to illustrate vibe coding's generate-then-edit loop.
- •Claude Artifacts shows generated code and a rendered app side by side
- •A calculator is created and then edited with a follow-up prompt
- •Riley sees this as the start of an era of personalized software
- •You don't need to know how the code works, just as you don't need to understand a camera
Finding a Problem and Prompting an App Into Existence
Riley uses the Vibe Code mobile app to prompt a voice-to-YouTube-outline app, then diagrams a problem and ideal state for a Paddy Galloway-style script coach. They discuss turning expertise into personalized, dynamic AI products, and Riley says prompting is a fundamental skill.
- •Riley records a voice prompt to build a YouTube outline app using Whisper
- •He diagrams problem and ideal state, using Whimsical
- •The idea: an AI that asks dynamic questions and writes scripts in Paddy Galloway's style
- •Consultants and experts can convert what's in their heads into software
- •Riley says prompting is programming and a core skill
Using ChatGPT Workflows to Research and Refine Your Idea
Riley demonstrates ChatGPT with search and the o3 model to research Paddy Galloway and draft user questions. He and Calum discuss whether prompts need heavy context, with Riley favoring quick iteration and conversation.
- •Workflows honed in ChatGPT for your own problems can become apps for others
- •Calum says ChatGPT should be treated as a conversation, not as Google
- •Riley counters that search gathers context for you and you shouldn't over-polish a prompt
- •The user picks favorite questions in a collaborative back-and-forth
The Mobile App Result and the Case for Vibe Coding Tools
Riley tests the app made from one prompt and tweaks it by asking for bigger text. He explains Vibe Code's wedge as mobile app building compared with web-focused tools, and why vibe coding beats slow, costly engineering planning.
- •The app worked from one prompt and compiled as a real mobile app
- •Changes like bigger text are made by simply asking
- •Bolt, Lovable, Replit, Cursor and Windsurf mostly make web apps, while Vibe Code targets mobile
- •Riley expects a multimillion-dollar company to come out of a vibe-coding hackathon
Building a Web App in v0 From a Screenshot
Riley moves to v0, creates a Paddy GPT interface, then screenshots Lovable's landing page and prompts v0 to restyle the app. He also mentions WhisperFlow for voice input and generates a matching background image in ChatGPT.
- •v0 builds Next.js web apps with a high-quality design look
- •A screenshot plus a prompt restyles the app while keeping its functionality
- •WhisperFlow lets Riley dictate prompts instead of typing
- •ChatGPT's 4o model generated a background image, which was added to v0
- •Riley's hack: say "change nothing else" to keep the AI on task
Recap: Prototypes, Tools and Tinkering Without a Playbook
Calum recaps the steps so far, and Riley explains that they built mock-ups of a mobile app and web interface. He stresses focusing on types of tools, experimenting, and notes that LevelsIO's unstructured approach led to a $50K flight simulator game.
- •Step one was identifying a pain point and a solution, then building a v1 prototype
- •Prototypes and mock-ups communicate ideas better than a Google Doc
- •Riley says to focus on the types of tools rather than specific ones
- •Block out four hours and tinker, since problems reveal themselves along the way
- •LevelsIO vibe coded a flight simulator game that made 50K
When to Go Live and Deploying in 15 Seconds
Calum asks when to start getting feedback, and Riley says to make it solve your own problem, spend a week refining, and ask AI for steps to go live. He then deploys the v0 app to the internet in about 15 seconds and recommends sharing it with honest friends.
- •Make the product work for your own use case and refine the prompting for about a week
- •Ask AI for the steps to go live; it's a permissionless world
- •Deploying from v0 took about 15 seconds
- •Send the link to a trusted friend who will be honest, not to your mom
Monetizing and Scaling: Payments, Security and Distribution
Calum asks how to scale toward revenue like LevelsIO's. Riley suggests hiring a developer for payments and a security check at roughly $200-250 for the payment piece, versus $5,000-$10,000 for a full build, and says creators with distribution have an edge.
- •Stripe payments could be added by a developer for around $200-250
- •Riley doesn't think vibe-coded apps can yet be fully trusted on bugs and security
- •A full build by an agency or developer would cost about $5,000-$10,000
- •With 150-200 users paying $30 a month, you can invest in making the app scalable
- •He sees a golden period of content creators combined with vibe coding
The Saturday Challenge: Build, Record, and Post on Twitter
Calum asks what one thing a viewer should do on a free Saturday. Riley says to drop the money motive, build an app in any tool, screen record the process and post it on Twitter, and he offers to retweet the first 50 people who tag both of them.
- •Let go of the desire to make money immediately and focus on building a skill
- •Screen record yourself building and share what you made at the end of the day
- •Twitter is Riley's recommended place to start, where communities emerge
- •Riley made Vibe Code content before he was an expert
- •He'll retweet the first 50 people who tag both hosts to his roughly 70,000 followers

Publisher timestamps

Get more out of YouTube videos.

Read along on your phone, search it beside the video, or ask your favorite AI app what was said.