Podcast DJ: The Cliff Notes for Your Podcast Queue
with Kevin Smith, Snipd
Podcast DJ: The Cliff Notes for Your Podcast Queue
Show Notes
Most founders aren't short on information. They're short on signal. Kevin Smith, founder and CEO of Snipd, has been quietly solving the problem that every high-volume podcast listener knows intimately: you have hundreds of episodes queued up, a finite amount of attention, and no good way to extract the gold without sitting through the banter.
In this return visit, Kevin pulls back the curtain on Snipd's newest feature: the AI DJ. Think of it as a personalized audio moderator that analyzes an episode, identifies the highest-value moments, and guides you through them one by one — complete with intros, segues, and a closing takeaway summary. You still hear the real voices. You still get the original tone and energy. You just skip to the parts that matter. Kevin's promise: get through a podcast episode in 25% of the time.
Frameworks from This Episode
The Signal-to-Noise Problem in Content Consumption
Most podcast content is context-setting, banter, or setup. The insight-dense moments are a fraction of total runtime.
- ›AI DJ targets high-value moments specifically — not through summary but through curation of actual audio clips.
- ›The goal is preserving the emotional texture of the original while compressing the time investment.
- ›Founder takeaway: apply the same filter to every content input — ask what percentage is signal before subscribing.
The Highlight Reel Model
Kevin draws the analogy to sports: you watch the big match in full, but for the other 10 games you want highlights.
- ›Podcast listening should work the same way — full episodes for your favorites, AI DJ for everything else.
- ›This positions AI DJ not as a replacement for deep listening but as an expansion of total coverage.
- ›Founders can now follow 3× more voices in the same weekly time budget.
Personalization as the North Star
The current AI DJ surfaces the same highlights for all users. The roadmap moves toward a fully personalized agent.
- ›The agent will learn what you save, what you replay, and what you skip.
- ›Users will eventually prompt the DJ in natural language: 'Always include book recommendations from Lenny' or 'Find me everything about hiring from the last 30 days.'
- ›Personalization creates compounding returns — the longer you use it, the more precisely it matches your learning style.
The Hive Brain Advantage
Snipd is the only podcast app with data on which specific moments inside an episode users found valuable.
- ›This community-level signal feeds back into the AI DJ algorithm to surface the most-validated moments.
- ›Personal listening history layers on top of that collective signal for a two-axis personalization model.
- ›The data moat compounds: more snips → better signal → better DJ → more users → more snips.
Content Creator vs. Listener Control
Tools like Opus Clip give creators control over which moments get amplified. Kevin argues the future flips this.
- ›The listener's AI agent, not the creator, will decide what's worth surfacing for each individual.
- ›This represents a fundamental shift in the distribution and discovery model for audio content.
- ›Founders building content should optimize for the AI listener — not just the human one.
Founder Experiment: Build a Personal Podcast Knowledge Base
Use Snipd's export features and a simple AI pipeline to turn two weeks of podcast listening into a structured strategic briefing.
- 1Use Snipd for two weeks. Snip every moment that triggers a new idea, a framework you want to test, or a quote you'd reference in a pitch.
- 2Export your snips to Notion or a text file using Snipd's built-in export features.
- 3Paste your collection into a Claude or ChatGPT project with this prompt: "You are my business strategy advisor. These are the insights I've personally curated from the last two weeks of podcast listening. Identify the three recurring themes, surface any contradictions, and recommend one concrete experiment I should run this month based on what I've been consuming."
- 4Run the experiment. Document results.
- 5Repeat monthly. Over time you're not just listening more efficiently — you're building a feedback loop between your inputs and your decisions.
The compounding effect: Your snip library becomes a living record of your intellectual inputs. After six months, you can ask it what you were obsessing over in Q1, which ideas recurred across different guests, and whether your actual decisions tracked your stated priorities.
Key Terms
These terms have been added to the AI for Founders Glossary. Search by Kevin Smith to filter them.
Q&A
What is Snipd's AI DJ feature?
AI DJ is a Snipd feature that analyzes a podcast episode, identifies the most valuable moments, and plays them back with AI-narrated introductions, segues, and a closing summary — allowing listeners to get through an episode in roughly 25% of the original runtime.
How does Snipd decide which podcast moments are most valuable?
Snipd uses a combination of AI analysis of the full episode transcript and metadata, community-level snipping data showing which moments users most commonly saved, and personal listening history to weight clips that align with what a specific user has found valuable before.
Is the AI DJ feature currently available?
Yes. As of this episode, AI DJ is live for all Snipd users for English-language podcasts.
How is AI DJ different from a podcast summary tool?
Unlike text summaries, AI DJ plays the actual original audio with the real voices and original intonation intact. It does not replace the content with a generated description. It selects and connects specific moments from the episode.
Can I customize what the AI DJ surfaces for me?
Not yet in full. The current version surfaces the same highlights for all users. Personalization and natural language prompting are on the active roadmap.
What integrations does Snipd support for exporting snips?
Snipd supports export to Notion and other platforms. It also generates transcripts, book lists, quotes, auto chapters, and custom summaries from episodes.
How many languages does Snipd support?
Snipd supports up to 26 languages, though the AI DJ feature currently works only with English-language podcasts.