How to use Claude to build a podcast SaaS
The author reports deploying PodGen.io after about three weeks of Claude conversations, with credit pricing, podcast generation, distribution integration and a user dashboard.
news.ycombinator.com
Prompt to try
Editorial adaptation, not the source author’s exact prompt. Replace bracketed inputs and review tool permissions first.
Help me build [podcast SaaS] in Cursor using Next.js, Supabase and Stripe. Integrate [script-generation API] and [voice-synthesis API], implement [credit rules] and a user history dashboard, and work through each integration with me. Ask for detailed business rules before implementing them.
Reach people exploring AI agents
See placements →What happened
The author reports deploying PodGen.io after about three weeks of Claude conversations, with credit pricing, podcast generation, distribution integration and a user dashboard.
Reported by baranoncel on Hacker NewsHow the agent did it
- Directed Claude through natural-language conversations to build the SaaS.
- Started fresh chats and re-explained the codebase when long conversations broke Cursor.
- Iterated on debugging and refined complex business logic with detailed explanations.
Where a person stepped in
The author directed product decisions, explained business logic, restarted broken conversations and iterated on debugging with Claude.
A person approved each step before the agent continued.
The PodGen.io author reports building and deploying an AI podcast SaaS without manually writing code. The setup was Cursor with Claude 4 Sonnet in max mode, with approximately $300 in additional max-mode costs. Development took about three weeks of conversations.
Claude handled the Next.js and React frontend, Supabase integration and database design, Stripe checkout and webhook handling, and user authentication. The author also credits it with connecting OpenAI for script generation and FAL.ai for voice synthesis. Responsive design, structured data and multilingual support were part of the reported implementation.
The work required repeated human explanations. Rate limits interrupted development, sometimes for hours. Long conversations broke Cursor four or five times, forcing the author to start fresh chats and explain the codebase again. Debugging took multiple iterations, and complex business logic needed detailed descriptions and refinement rounds.
The resulting service converted sources such as YouTube videos, PDFs and blogs into podcasts. The author reported credit-based pricing, more than fifty voices in more than twenty-five languages, distribution integration and a user dashboard with history.
Who it's for: Product builders who can explain business rules in detail and remain involved through iterative debugging and context resets.
Prompt
Suggested prompt (written by Agents In Use, not the author's)
Source: Hacker News · baranoncel
Frequently asked questions
Which model and editor were used?
The author used Claude 4 Sonnet in Cursor max mode and reports approximately $300 in extra max-mode costs.
What slowed development?
Claude rate limits, broken long conversations, repeated codebase explanations, iterative debugging and detailed refinement of business logic all slowed the work.
More details
Remaining steps
- Deployed PodGen.io with payments and podcast-generation integrations.
About this account
The agent generated code from the author’s instructions, with the author directing the work.
Platform: Hacker News. Attribution: https://news.ycombinator.com/item?id=44125171. Prompt: Suggested (written by Agents In Use, not the author's). Sourced by the Agents In Use pipeline.
Credit: baranoncel on Hacker News
Source published 29 May 2025
Source checked 14 Oct 2026
Tools and topics
Cursor · Claude 4 Sonnet · Supabase · Stripe · OpenAI API · FAL.ai
AI coding
Editorial completeness: 100/100. Measures documented fields, not agent performance or verified outcomes.