Hey HN! Why I built this: Over the last 10 years I relied heavily on Twitter as my primary source of information. I could get a wide range of information in one place, and each information nugget came in a small format of about 280 characters which was good for summarizing. But recently due to algorithmic changes and the AI frenzy, Twitter has become very noisy and hard to keep up with. So I built GoldenScoop. It takes personalized information requests and runs a swarm of agents in the background. Our swarm reads multiple news sources every day and provides customized information based on user requests. With just a click of a button, users can convert these scoops into podcasts presented by our AI podcasters. They can keep generating podcast episodes, editing and presenting a wide range of information personalized for each user. What it does:

AI agents process 10,000+ sources daily Creates personalized Twitter-style feeds with original source references One-click conversion to HD AI podcasts Continuous content synthesis based on your interests

Try it: [https://goldenscoop.live/] You can listen to our intro podcast episode on the home page.

See a screenshot(https://pkwtjnzsjqxaeavnjyrk.supabase.co/storage/v1/object/p...)

Our top sources: TechCrunch, Reuters, Bloomberg, Hacker News, Wall Street Journal, BBC News, The Verge, MIT Technology Review, Nature, Financial Times, ArXiv, Scientific American, Washington Post, GitHub Trending, Y Combinator Blog, Fast Company, Wired, CB Insights, InfoQ, CNN, Associated Press, Forbes, The Guardian, NPR, Ars Technica

Built with: Google Gemini, Mistral Medium 3, ElevenLabs for HD voice synthesis, React and AGNO agentic framework

Open to feedback and questions!