PodcastingCAN I VIBECODE
Podsqueeze?
BUILD FIRST
- One defined user and a repeatable end-to-end job
- A small, clear data model with sign-in only if it is genuinely needed
- Loading, empty, error, and success states that make the workflow feel complete
LEAVE FOR LATER
- Large integration marketplaces
- Multi-region scale and enterprise administration
- Every power-user setting from the established product
RESEARCH NOTES
What this tool actually does.
Podsqueeze is an AI-powered podcast production and promotion platform. Users upload audio/video or connect an RSS feed; the AI transcribes episodes with speaker identification and timestamps, generates SRT subtitle files, structured show notes (titles, summaries, chapters, highlights, bullet points, mentioned resources), blog posts, newsletters, and social posts. It repurposes episodes into captioned clips and audiograms for TikTok/Instagram/YouTube Shorts via chapter selection, a subtitle-based text editor for trimming, and one-click AI-generated ~60-second shorts. Audio enhancement tools remove ums/silences, reduce noise, and normalize volume. It also builds branded, SEO-optimized podcast websites with interactive episode players, auto-generated episode landing pages, shareable timestamps, and quote images. A podcast topics finder assists episode planning, and the AI voice can be tuned
RESEARCHED BUILD PROMPT
Give this to your coding agent.
Build a focused replacement for Podsqueeze's core podcast repurposing workflow: a web app where a user uploads an audio/video podcast episode (or provides an RSS feed URL) and receives AI-generated transcripts with speaker labels and timestamps, downloadable SRT files, structured show notes (title, summary, chapters with timestamps, bullet-point highlights, mentioned resources), and short-form social posts. Secondary workflow: chapter-based clip export with burned-in captions and a subtitle text editor for trimming ends.
User-facing requirements: single-page upload with progress; results dashboard listing downloadable artifacts; clip preview with caption style presets; minimal project management (list episodes, re-download outputs).
Technical requirements: backend job queue (e.g., BullMQ + Redis) calling a transcription provider (e.g., AssemblyAI or Whisper API) and an LLM (e.g., GPT-4o via OpenRouter) for show notes and social copy; storage via S3-compatible object storage; frontend in Next.js/React with a video preview player (e.g., Mux or native video element) for clip trimming.
Implementation approach: two-server architecture—API (Node/Fastify) and worker—with webhooks for transcription completion; store job state in Postgres; generate clips via ffmpeg on the worker; use SRT from transcription as the source of truth for editing.
Do not copy Podsqueeze's brand name, logo, color scheme, or UI layout. Deliberately out of scope: full audio cleanup (noise removal, um/ah removal), podcast website builder, newsletter creation, topics finder, multi-show agency dashboard, and payment/billing. Include a README covering setup steps (env vars, Docker Compose for Postgres/Redis, ffmpeg install), provider API keys, job flow explanation, and known limitations (transcription latency, model hallucination on speaker names, clip encoding costs, no auth beyond a demo single-user mode).