StarZero All shows Try the podcast clipper →

Dec 1, 2025 · 33m · startup-ideas

Be a 10x Vibe Coder (Claude Code + Cursor + MCP)

Chris Raroque · 26m spoken Greg Isenberg · 3m spoken
0:00 / 0:00
▶ Watch on YouTube →

gold bands on the timeline = statements, start to end. Hover to read, click to jump. CC turns on captions

Host Greg Isenberg interviews solo developer Chris to reveal advanced 'vibe coding' workflows that combine Cursor, Claude Code, Model Context Protocol (MCP) integrations, and voice dictation to build polished applications at 10x speed.

How this conversation actually went

Every chapter scored 0–10 on four independent dynamics. Hover any point for the reasoning behind the score. Greg holds 11% of the talking time here. How this is scored →

Greg as informed peer 1.5 Guest teaching 7.1 Guest disagreement 0.5 Greg pushing back 0.3
05100:0010:0020:0030:000:56–4:24 · Greg as informed peer 2/10 Welcoming Chris and Setting Episode Goals Greg introduces Chris and frames the episode around actionable vibe coding workflows. Chris explains his dual setup running Cursor and Claude Code simultaneously side-by-side.4:24–8:27 · Greg as informed peer 1/10 Navigating Model Strengths and Opus Rate Limits Chris breaks down his model routing strategy, managing Opus rate limits and sharing a contrarian tip to use GPT-5.1 high specifically for Cursor plan mode while reserving Sonnet for execution.8:27–11:14 · Greg as informed peer 4/10 Choosing Beginner-Friendly Platforms for Mobile Apps Greg tests Chris's tool recommendations by bringing up competitor mobile coding apps like Roark and VibeCode. Chris defends CreateAnything as superior for design fidelity and instruction following.11:15–15:39 · Greg as informed peer 0/10 Live Demonstration: Prompt Dictation in Xcode Chris delivers a live tutorial in Xcode showing how detailed voice dictation via Whisperflow allows rapid, high-context prompt engineering without manual typing.15:40–17:44 · Greg as informed peer 2/10 Debugging Animations and Comparing Plan Mode Behavior Chris demonstrates iterative UI debugging in the iOS simulator using slow-motion animation playback to refine Claude's output.17:45–20:54 · Greg as informed peer 3/10 Head-to-Head Test: Cursor Plan Mode vs. Claude Code Greg asks Chris to predict the winner between Claude Code and Cursor Plan Mode in a head-to-head test. When Cursor wins on first prompt, Greg validates the outcome.20:55–24:12 · Greg as informed peer 0/10 Pro Tips: Mandatory Plan Mode and the UltraThink Keyword Chris educates the audience on mandatory plan mode in Cursor and Claude Code, revealing the undocumented 'ultra think' keyword in Claude Code that triggers deeper reasoning.24:14–28:27 · Greg as informed peer 0/10 Pro Tip: Running Server Background Tasks in Claude Code Chris outlines advanced workflows including background server execution for automatic log parsing, plus Context7 and Supabase MCP integrations, anticipating developer skepticism over direct DB tool access.28:28–30:55 · Greg as informed peer 0/10 Automated Code Review and Security Audits with AI Bots Chris explains how solo vibe coders can replace human code reviews by integrating Bugbot and PR review agents into GitHub repositories to catch vulnerabilities.31:04–33:07 · Greg as informed peer 3/10 Technical Architecture Research Using Claude Desktop Chris shares how to utilize Claude Desktop Deep Research for 12-minute architectural reviews before prompting the code editor; Greg synthesizes the episode insights and concludes.0:56–4:24 · Guest teaching 5/10 Welcoming Chris and Setting Episode Goals Greg introduces Chris and frames the episode around actionable vibe coding workflows. Chris explains his dual setup running Cursor and Claude Code simultaneously side-by-side.4:24–8:27 · Guest teaching 7/10 Navigating Model Strengths and Opus Rate Limits Chris breaks down his model routing strategy, managing Opus rate limits and sharing a contrarian tip to use GPT-5.1 high specifically for Cursor plan mode while reserving Sonnet for execution.8:27–11:14 · Guest teaching 6/10 Choosing Beginner-Friendly Platforms for Mobile Apps Greg tests Chris's tool recommendations by bringing up competitor mobile coding apps like Roark and VibeCode. Chris defends CreateAnything as superior for design fidelity and instruction following.11:15–15:39 · Guest teaching 8/10 Live Demonstration: Prompt Dictation in Xcode Chris delivers a live tutorial in Xcode showing how detailed voice dictation via Whisperflow allows rapid, high-context prompt engineering without manual typing.15:40–17:44 · Guest teaching 7/10 Debugging Animations and Comparing Plan Mode Behavior Chris demonstrates iterative UI debugging in the iOS simulator using slow-motion animation playback to refine Claude's output.17:45–20:54 · Guest teaching 6/10 Head-to-Head Test: Cursor Plan Mode vs. Claude Code Greg asks Chris to predict the winner between Claude Code and Cursor Plan Mode in a head-to-head test. When Cursor wins on first prompt, Greg validates the outcome.20:55–24:12 · Guest teaching 8/10 Pro Tips: Mandatory Plan Mode and the UltraThink Keyword Chris educates the audience on mandatory plan mode in Cursor and Claude Code, revealing the undocumented 'ultra think' keyword in Claude Code that triggers deeper reasoning.24:14–28:27 · Guest teaching 9/10 Pro Tip: Running Server Background Tasks in Claude Code Chris outlines advanced workflows including background server execution for automatic log parsing, plus Context7 and Supabase MCP integrations, anticipating developer skepticism over direct DB tool access.28:28–30:55 · Guest teaching 8/10 Automated Code Review and Security Audits with AI Bots Chris explains how solo vibe coders can replace human code reviews by integrating Bugbot and PR review agents into GitHub repositories to catch vulnerabilities.31:04–33:07 · Guest teaching 7/10 Technical Architecture Research Using Claude Desktop Chris shares how to utilize Claude Desktop Deep Research for 12-minute architectural reviews before prompting the code editor; Greg synthesizes the episode insights and concludes.0:56–4:24 · Guest disagreement 0/10 Welcoming Chris and Setting Episode Goals Greg introduces Chris and frames the episode around actionable vibe coding workflows. Chris explains his dual setup running Cursor and Claude Code simultaneously side-by-side.4:24–8:27 · Guest disagreement 1/10 Navigating Model Strengths and Opus Rate Limits Chris breaks down his model routing strategy, managing Opus rate limits and sharing a contrarian tip to use GPT-5.1 high specifically for Cursor plan mode while reserving Sonnet for execution.8:27–11:14 · Guest disagreement 1/10 Choosing Beginner-Friendly Platforms for Mobile Apps Greg tests Chris's tool recommendations by bringing up competitor mobile coding apps like Roark and VibeCode. Chris defends CreateAnything as superior for design fidelity and instruction following.11:15–15:39 · Guest disagreement 0/10 Live Demonstration: Prompt Dictation in Xcode Chris delivers a live tutorial in Xcode showing how detailed voice dictation via Whisperflow allows rapid, high-context prompt engineering without manual typing.15:40–17:44 · Guest disagreement 0/10 Debugging Animations and Comparing Plan Mode Behavior Chris demonstrates iterative UI debugging in the iOS simulator using slow-motion animation playback to refine Claude's output.17:45–20:54 · Guest disagreement 1/10 Head-to-Head Test: Cursor Plan Mode vs. Claude Code Greg asks Chris to predict the winner between Claude Code and Cursor Plan Mode in a head-to-head test. When Cursor wins on first prompt, Greg validates the outcome.20:55–24:12 · Guest disagreement 0/10 Pro Tips: Mandatory Plan Mode and the UltraThink Keyword Chris educates the audience on mandatory plan mode in Cursor and Claude Code, revealing the undocumented 'ultra think' keyword in Claude Code that triggers deeper reasoning.24:14–28:27 · Guest disagreement 2/10 Pro Tip: Running Server Background Tasks in Claude Code Chris outlines advanced workflows including background server execution for automatic log parsing, plus Context7 and Supabase MCP integrations, anticipating developer skepticism over direct DB tool access.28:28–30:55 · Guest disagreement 0/10 Automated Code Review and Security Audits with AI Bots Chris explains how solo vibe coders can replace human code reviews by integrating Bugbot and PR review agents into GitHub repositories to catch vulnerabilities.31:04–33:07 · Guest disagreement 0/10 Technical Architecture Research Using Claude Desktop Chris shares how to utilize Claude Desktop Deep Research for 12-minute architectural reviews before prompting the code editor; Greg synthesizes the episode insights and concludes.0:56–4:24 · Greg pushing back 0/10 Welcoming Chris and Setting Episode Goals Greg introduces Chris and frames the episode around actionable vibe coding workflows. Chris explains his dual setup running Cursor and Claude Code simultaneously side-by-side.4:24–8:27 · Greg pushing back 0/10 Navigating Model Strengths and Opus Rate Limits Chris breaks down his model routing strategy, managing Opus rate limits and sharing a contrarian tip to use GPT-5.1 high specifically for Cursor plan mode while reserving Sonnet for execution.8:27–11:14 · Greg pushing back 2/10 Choosing Beginner-Friendly Platforms for Mobile Apps Greg tests Chris's tool recommendations by bringing up competitor mobile coding apps like Roark and VibeCode. Chris defends CreateAnything as superior for design fidelity and instruction following.11:15–15:39 · Greg pushing back 0/10 Live Demonstration: Prompt Dictation in Xcode Chris delivers a live tutorial in Xcode showing how detailed voice dictation via Whisperflow allows rapid, high-context prompt engineering without manual typing.15:40–17:44 · Greg pushing back 0/10 Debugging Animations and Comparing Plan Mode Behavior Chris demonstrates iterative UI debugging in the iOS simulator using slow-motion animation playback to refine Claude's output.17:45–20:54 · Greg pushing back 1/10 Head-to-Head Test: Cursor Plan Mode vs. Claude Code Greg asks Chris to predict the winner between Claude Code and Cursor Plan Mode in a head-to-head test. When Cursor wins on first prompt, Greg validates the outcome.20:55–24:12 · Greg pushing back 0/10 Pro Tips: Mandatory Plan Mode and the UltraThink Keyword Chris educates the audience on mandatory plan mode in Cursor and Claude Code, revealing the undocumented 'ultra think' keyword in Claude Code that triggers deeper reasoning.24:14–28:27 · Greg pushing back 0/10 Pro Tip: Running Server Background Tasks in Claude Code Chris outlines advanced workflows including background server execution for automatic log parsing, plus Context7 and Supabase MCP integrations, anticipating developer skepticism over direct DB tool access.28:28–30:55 · Greg pushing back 0/10 Automated Code Review and Security Audits with AI Bots Chris explains how solo vibe coders can replace human code reviews by integrating Bugbot and PR review agents into GitHub repositories to catch vulnerabilities.31:04–33:07 · Greg pushing back 0/10 Technical Architecture Research Using Claude Desktop Chris shares how to utilize Claude Desktop Deep Research for 12-minute architectural reviews before prompting the code editor; Greg synthesizes the episode insights and concludes.

speaking balance: gold is Greg, purple is the guest (3 minute bins)

0:00 · Greg 44.4% · guest 55.6%0:00 · Greg 44.4% · guest 55.6%3:00 · Greg 25.2% · guest 74.8%3:00 · Greg 25.2% · guest 74.8%6:00 · Greg 11.5% · guest 88.5%6:00 · Greg 11.5% · guest 88.5%9:00 · Greg 16.8% · guest 83.2%9:00 · Greg 16.8% · guest 83.2%12:00 · Greg 0% · guest 100%12:00 · Greg 0% · guest 100%15:00 · Greg 0.6% · guest 99.4%15:00 · Greg 0.6% · guest 99.4%18:00 · Greg 4.8% · guest 95.2%18:00 · Greg 4.8% · guest 95.2%21:00 · Greg 0% · guest 100%21:00 · Greg 0% · guest 100%24:00 · Greg 0% · guest 100%24:00 · Greg 0% · guest 100%27:00 · Greg 0% · guest 100%27:00 · Greg 0% · guest 100%30:00 · Greg 0% · guest 100%30:00 · Greg 0% · guest 100%33:00 · Greg 80% · guest 20%33:00 · Greg 80% · guest 20%
Sharpest disagreement ▶ 26:15 Preempting traditional developer security objections to Supabase MCP

Chris anticipates and dismisses traditional developer criticism regarding giving AI agents direct MCP write access to backend databases, arguing the AI actually sets up security rules better than humans.

 Hardest push from Greg ▶ 9:16 Probing guest on alternative mobile vibe coding tools

Greg interrupts the guest's CreateAnything recommendation to specifically challenge whether competitor platforms like Roark and VibeCode app offer better alternatives.

 Biggest teaching moment ▶ 22:40 Demonstrating the undocumented UltraThink keyword

Chris reveals that typing 'ultra think' into Claude Code changes UI colors and forces the model into deeper reasoning mode, a hidden feature Greg had never heard of.

 Greg holds their own ▶ 9:16 Demonstrating domain knowledge of current mobile AI builders

Greg demonstrates active market knowledge of cutting-edge mobile developer tooling by citing Roark and VibeCode to benchmark Chris's assertions.
the scores for every segment, with the reasoning behind each
ChapterTopicGreg as informed peerGuest teachingGuest disagreementGreg pushing backWhy
Welcoming Chris and Setting Episode Goals 2500 Greg introduces Chris and frames the episode around actionable vibe coding workflows. Chris explains his dual setup running Cursor and Claude Code simultaneously side-by-side.
Navigating Model Strengths and Opus Rate Limits 1710 Chris breaks down his model routing strategy, managing Opus rate limits and sharing a contrarian tip to use GPT-5.1 high specifically for Cursor plan mode while reserving Sonnet for execution.
Choosing Beginner-Friendly Platforms for Mobile Apps 4612 Greg tests Chris's tool recommendations by bringing up competitor mobile coding apps like Roark and VibeCode. Chris defends CreateAnything as superior for design fidelity and instruction following.
Live Demonstration: Prompt Dictation in Xcode 0800 Chris delivers a live tutorial in Xcode showing how detailed voice dictation via Whisperflow allows rapid, high-context prompt engineering without manual typing.
Debugging Animations and Comparing Plan Mode Behavior 2700 Chris demonstrates iterative UI debugging in the iOS simulator using slow-motion animation playback to refine Claude's output.
Head-to-Head Test: Cursor Plan Mode vs. Claude Code 3611 Greg asks Chris to predict the winner between Claude Code and Cursor Plan Mode in a head-to-head test. When Cursor wins on first prompt, Greg validates the outcome.
Pro Tips: Mandatory Plan Mode and the UltraThink Keyword 0800 Chris educates the audience on mandatory plan mode in Cursor and Claude Code, revealing the undocumented 'ultra think' keyword in Claude Code that triggers deeper reasoning.
Pro Tip: Running Server Background Tasks in Claude Code 0920 Chris outlines advanced workflows including background server execution for automatic log parsing, plus Context7 and Supabase MCP integrations, anticipating developer skepticism over direct DB tool access.
Automated Code Review and Security Audits with AI Bots 0800 Chris explains how solo vibe coders can replace human code reviews by integrating Bugbot and PR review agents into GitHub repositories to catch vulnerabilities.
Technical Architecture Research Using Claude Desktop 3700 Chris shares how to utilize Claude Desktop Deep Research for 12-minute architectural reviews before prompting the code editor; Greg synthesizes the episode insights and concludes.

Statements from this episode (0)

Nothing in this episode matches those filters. clear them

Made with StarZero

Turn any episode into a week of clips.

This entire site, over 300 episodes transcribed, diarized, checked and made playable, runs on the StarZero media pipeline. Drop in your own episode and the podcast clipper finds the moments worth sharing, cuts them, captions them, and reframes them for every feed.

Try the podcast clipper Learn more