Heidelberg AI News

Video · Riley Brown ·

I Vibe Coded a $100K/Month App in One Afternoon (Claude Projects)

In short

The video demonstrates building Glaze, a Screen Studio-style Mac and iPhone app, plus its landing page, launch video, and UGC ad from a phone using Claude Projects and parallel sessions. It shows how AI can compress product development and marketing into a few hours, while highlighting that desktop quality and visual polish still require iterative testing and refinement.

Key points

  • The project used about 20 to 25 prompts across parallel Claude Code sessions.
  • Glaze included screen recording, cursor effects, backgrounds, device frames, presets, and exports.
  • The Mac app initially had low-resolution output, excessive zoom blur, and editing flashes that required fixes.
  • The iPhone app supported recording, importing from Photos, frame and background customization, and camera-roll exports.
  • The workflow created a landing page, a 37-second Blender launch video, and a subtitled UGC ad.
  • Testing on real Mac and iPhone devices drove iterative changes to lighting, icons, motion blur, and app design.

Written by an AI model from the video’s captions and description.

I spent a few hours building an entire product launch from my phone using Claude Code in Claude Projects, with a mix of Opus 5.5 and Fable 5.2. Just by talking to my phone, I built Glaze, a Screen Studio clone for Mac, plus an iPhone app, a landing page, a 3D launch video in Blender, and a UGC-style ad.

Watch on YouTube The opening of the article at youtube.com.

In this story

A tool or model opens in the Heidelberg AI Curriculum: what it does, what it costs, how to start. A repo opens on GitHub.

Related stories

All stories

VideoWes Roth

Claude JUST got MYTHOS...

The video argues that Claude is evolving into an AI operating-system layer that coordinates office tools, meetings, data, software, and third-party apps rather than replacing each app directly. It also explains how eligible independent researchers, open-source developers, and vulnerability hunters may apply for Mythos 5.2 through Anthropic’s Cyber Verification Program, while challenging the interpretation of the MIT study claiming 95% of enterprise AI projects produce no ROI.