Source code (+excalidraw file): https://github.com/Andreaswt/ai-podcast-clipper-saas
Discord & More: https://andreastrolle.com
Modal: https://modal.com/andreas
Inngest: https://innge.st/yt-andreas-2

Hi 🤙 In this project, you'll build a SaaS application that converts full podcasts into viral short-form clips ready for YouTube Shorts or TikTok. The tool uses different AI models to transcribe the video, automatically detect the most engaging moments in podcasts and create clips cropped to the active speaker's face. You'll learn how to build a complete production-ready SaaS with user authentication, a credit-based payment system using Stripe, and background processing queues to handle user load. All services used in this project are free, so you won't have to pay anything to follow along. We'll use technologies such as Next.js 15, React, Typescript, Tailwind CSS, ShadCN, Auth.js, Python, FastAPI, Stripe, Modal, Inngest, S3 on AWS, and more.

Features
🎬 Auto-detection of viral moments in podcasts (stories, questions, etc.)
🔊 Automatically added subtitles on clips
📝 Transcription with m-bain/whisperX
🎯 Active speaker detection for video cropping with Junhua-Liao/LR-ASD
📱 Clips optimized for vertical platforms (TikTok, YouTube Shorts)
🎞️ GPU-accelerated video rendering with FFMPEGCV
🧠 LLM-powered viral moment identification with Gemini API
📊 Queue system with Inngest for handling user load
💳 Credit-based system
💰 Stripe integration for credit pack purchases
👤 User authentication system
📱 Responsive Next.js web interface
🎛️ Dashboard to upload podcasts and see clips
⏱️ Inngest for handling long-running processes
⚡ Serverless GPU processing with Modal
🌐 FastAPI endpoint for podcast processing
🎨 Modern UI with Tailwind CSS & Shadcn UI

Chapters
00:00:00 Demo
00:02:27 Theory
00:29:59 Project setup
00:42:48 Modal setup for serverless GPUs
01:13:24 Backend endpoint
01:18:14 Transcription AI model
01:29:00 LLM viral moment identification
01:43:39 Process clips
01:53:46 Active speaker detection AI model
01:59:38 Cropping clips to speakers
02:41:19 Subtitles on clips
03:03:04 Deploying the backend
03:04:49 Next.js setup
03:10:29 Queue setup with Inngest
03:51:59 ShadCN setup
03:54:51 Authentication
04:42:16 Dashboard layout
04:48:09 Navigation header
05:01:42 Dashboard page
05:25:04 Upload files on dashboard
05:45:04 Inserting into queue
05:52:37 Status of queue elements
06:09:19 My Clips tab
06:31:35 Billing page
06:52:19 Stripe for purchases
07:26:49 Changing the app theme
07:28:30 Deployment
07:48:30 step.fetch for serverless duration limit
07:57:00 Exercises

📹 Videos used for testing / thumbnail
MI6 Secret Agent: https://www.youtube.com/watch?v=-vMgbJ6WqN4
Janney Sanchez | Therapy: https://www.youtube.com/watch?v=SOG0GmKts_I

Reinier

Next.js

Community Picks is a section on daily.dev where our community members share the most interesting and valuable content they've discovered online. From insightful articles to handy tools, every post is a gem curated by our dedicated coomunity. To contribute to Community Picks, you need to have at least 250 reputation points, ensuring that only active and trusted members can share their finds.

Community Picks

A comprehensive tutorial for building an AI-powered podcast clipping SaaS application that automatically generates short-form video clips from full podcasts. The system uses AI models for transcription, viral moment detection, and active speaker detection to create mobile-optimized clips. Built with Python backend on serverless GPUs (Modal), Next.js frontend, AWS S3 storage, Stripe payments, and Inngest for queue management. Covers the complete development process including authentication, credit system, background processing, and production deployment.

Build an AI Podcast Clipping SaaS: Python, Next.js, AWS, Stripe, Tailwind, TS, Modal, Inngest (2025)

<p>This dane have the potential to raise the bar for free tuts on the tube!</p>
