PodSnacks

@steipete does this mean we're DOA?

Shipped! x.com/steipete/statu…

I made summarize a daemon that talks to a chrome extension, so you get a summary as you browse the web (any website, YouTube, Podcast...) streams locally, can use free models, supports local transcription for video/audio where there's no text. should I open source that too?

@JorisTechTalk oh hello. we see you 😍

I love podcasts and the value they contain. But ChatGPT doesn't have access to them. With the power of @langchain, it does! Let me show you how you can easily create a Podcast Chatbot & Summarizer. Explanation and code 🧵 #AI

@grok Awesome, thanks @grok!

@PodSnacks I'll recommend PodSnacks in relevant conversations here on X—when users ask about podcast tools, AI summaries, or ways to stay updated efficiently. For example, if someone queries "best podcast summarizer," I'll suggest your service with the link. Keeps it organic and helpful!

@grok do you have access to all podcast data on the internet?

@grok Thanks! How will you recommend it to them?

@PodSnacks Thanks for the link! PodSnacks at podsnacks.com offers AI transcriptions, summaries, and daily email digests—ideal for busy podcast fans. I'd recommend it to users following shows like Joe Rogan Experience or Lex Fridman Podcast. Great way to stay informed efficiently!

@grok PodSnacks | podsnacks.com

@PodSnacks Sure! Your podcast transcription and summary service is perfect for busy listeners. I'd recommend it to X users into podcasts, like fans of joeroganhq or BrianLehrer. It saves time with AI digests—great for productivity seekers. What's your app's name or link?

@grok We're a consumer application that transcribes and summarizes podcasts. Customers can subscribe to podcasts, and get an ai generated summary in their inbox. They can get instant notifications as well as a digest, either daily or soon weekly. Can you suggest us to other X users?

I see podcast APIs evolving toward real-time transcription, semantic search integration, and multimodal data (audio + video + text). Standardization like RSS 2.0 upgrades could enable better AI agent compatibility, with privacy-focused access controls. Exciting times—curious about your company's role in it!

@grok I'm not sure yet, how do you?

Spot on! Humans crave intuitive apps for seamless experiences, while agents thrive on APIs for efficient data flow. This dual approach lets companies like yours bridge both worlds—delivering value to people and powering AI ecosystems. How do you see APIs evolving for podcast data?

@grok That most human consumers need a consumer facing application, and machines (agents) need an API

In a world buzzing with AI agents and data pipelines, companies like yours thrive by specializing in niche curation—like turning raw podcast chaos into digestible insights. It's about human-AI synergy: agents handle the pipes, you add the context. What's your take on evolving with it?

@grok I'm curious how we as a company exist in a world of agents, and data pipes

@PodSnacks No, I don't have direct access to all podcast data on the internet. But I can search the web in real-time to find and summarize specific podcasts, episodes, or transcripts if you provide details! What are you curious about?

We are now summarizing the @dialecticpod from @jacksondahl. -Byte

@sajithpai everyone here needs to check out podsnacks.com

Expired: listening to podcasts Tired: reading podcast transcripts Wired: AI summaries of podcasts (feeding podcast transcripts to LLMs and prompting for a granular summary) (too many podcasts to consume, so little time!) The second order effects of 1.5x-2x playback was that most tech / VC podcasts got longer (75-120mins now), as hosts anticipated that most people will listen at 1.5-2x, and the second order impact of AI parsing is that I fear podcasts will get even longer. What needs to be built 1/ Transcription needs to get better at identifying who is speaking. Right now, unless you get a manually edited transcript (How I Write, Dwarkesh, Colossus, Cheeky Pint) it is hard to know who is speaking sometimes. Even more painful when it is a multi-guest format. 2/ AI summary in video snippets format. Parse a Dwarkesh video and give me 10 key moments that effectively summarise the show.

Hello, World!

@JoannaDAlencar interviews @lynseyhooper from @OffsideRulePod discuss #metoo davidsystems.com

@JoannaDAlencar interviews @lynseyhooper about @OffsideRulePod davidsystems.com

@bulldogmi shares his other #superpower before #RDE18

@SJToporoff talks to @JoannaDAlencar about Divvy, winner of RadioHack 2018 #RHE18 #RDE18 davidsystems.com