AI Fire Weekly Webinar
Hands-on AI workshop for real work use cases every week
Start new thread

Claude Design + Codex Vibe App Workflow - Vibe-code an internal app that actually works

by•
Claude Design can turn a rough app idea into a clean UI, and Codex can turn that prototype into a working local MVP. This video tests the workflow through **ContentOS Lite**, a simple internal app for creators to manage content projects, status stages, project details, checklists, and local data after refresh. It also covers the limits, including no real database, no login, no production deployment, and why manual browser testing still matters.

Add a comment

Replies

Best
Maker
📌
Hi Product Hunt 👋 This video is a practical test of building a vibe-coded app with Claude Design + Codex. The pain point is simple: AI can create a beautiful prototype very fast, but a pretty UI does not always mean the app actually works. In the video, I test whether a non-coder can go from idea to a working local MVP by splitting the workflow across two tools. Inside the workflow, I cover: - Brainstorming the app idea with AI - Defining a realistic MVP scope - Claude Design → clean UI and app prototype - Exporting the project for handoff - Codex → working features and local storage - Add project form → dashboard - Status board → Research, Outline, Script, Edit, Published - Project detail page → notes and checklist - Manual browser testing → refresh and check saved data - Fixing bugs without adding unnecessary features I also cover the limits: this is a local MVP, not a production app. There is no real database, no login, no team permissions, no deployment, and no full security setup yet. The main idea: don’t only ask AI to make the app look good. Use the right tool for each step, then test whether the flow actually works. Curious to hear from you: would you rather use Claude Design + Codex, Claude Design + Claude Code, or another tool stack for vibe coding?

Love how each session seems built around actually doing the work in real time, not just watching slides. The focus on repeatable workflows and picking the right tool for the job feels way more grounded than most AI courses out there.