What you get

What you get from a Dwarkesh Podcast episode.

Episode summary

The substance of the episode, distilled — what was discussed and why it matters, minus the runtime.

Key moments timestamped

The 5-10 moments worth revisiting, each linked to the exact second in the video.

Quotable lines

The most shareable lines, attributed and timestamped, ready to save or post.

Export anywhere

Markdown export into Notion, Obsidian, or any notes app. Build your own episode archive.

How it works

Three steps. No watching required.

01

Paste the episode URL

Any Dwarkesh Podcast episode on YouTube — back catalog and new drops both work.

02

Pick your depth

TL;DR, full summary, or detailed notes.

03

Read, share, or save

Skim in minutes, share a quote, or save the notes.

Examples

What goes in, what comes out.

▸ Long interview episode
Summary + key moments + quotes
▸ Guest deep-dive
Topic summary + takeaways
▸ Solo / monologue episode
Main points + standout lines
▸ News / commentary episode
Substance + timestamps
Built for

Built for Dwarkesh Podcast listeners.

Regular listeners

Stay in the loop on episodes you can't finish — the substance in a five-minute read.

Researchers & writers

Find and cite specific moments without scrubbing the whole episode.

Busy fans

Decide which episodes are worth the full listen, and jump straight to what matters.

Questions

Honest answers.

Does this work on every Dwarkesh Podcast episode?

Yes — every public Dwarkesh Podcast episode on YouTube, including the back catalog and new releases within hours of posting.

Is it really free?

Yes. 3 free summaries per month with no credit card. Pro is $5.99/mo for unlimited.

How accurate is the summary?

The model transcribes the actual audio, not just auto-captions, so names, claims, and quotes are preserved accurately.

Can I export to Notion or Obsidian?

Yes. Every summary exports to Markdown, which drops cleanly into Notion, Obsidian, or any notes app.

About the Dwarkesh Podcast

About the Dwarkesh Podcast

Dwarkesh Patel built a reputation for unusually well-researched interviews — he reads everything a guest has written before they sit down, which means episodes go deeper than the usual press-tour conversation. Guests have included AI researchers, historians, economists, and biologists, and the questions tend to probe rather than promote.

Because the episodes are dense and often run two to three hours, a summary that pulls out the actual claims and the points of disagreement saves real time. RecapGPT extracts the substantive arguments — not just topic headlines — so you can follow Patel's line of questioning without the full runtime.

About the Dwarkesh Podcast

About the Dwarkesh Podcast

Dwarkesh Patel built a reputation for unusually well-researched interviews — he reads everything a guest has written before they sit down, which means episodes go deeper than the usual press-tour conversation. Guests have included AI researchers, historians, economists, and biologists, and the questions tend to probe rather than promote.

Because the episodes are dense and often run two to three hours, a summary that pulls out the actual claims and the points of disagreement saves real time. RecapGPT extracts the substantive arguments — not just topic headlines — so you can follow Patel's line of questioning without the full runtime.

About the Dwarkesh Podcast

About the Dwarkesh Podcast

Dwarkesh Patel built a reputation for unusually well-researched interviews — he reads everything a guest has written before they sit down, which means episodes go deeper than the usual press-tour conversation. Guests have included AI researchers, historians, economists, and biologists, and the questions tend to probe rather than promote.

Because the episodes are dense and often run two to three hours, a summary that pulls out the actual claims and the points of disagreement saves real time. RecapGPT extracts the substantive arguments — not just topic headlines — so you can follow Patel's line of questioning without the full runtime.

About the Dwarkesh Podcast

About the Dwarkesh Podcast

Dwarkesh Patel built a reputation for unusually well-researched interviews — he reads everything a guest has written before they sit down, which means episodes go deeper than the usual press-tour conversation. Guests have included AI researchers, historians, economists, and biologists, and the questions tend to probe rather than promote.

Because the episodes are dense and often run two to three hours, a summary that pulls out the actual claims and the points of disagreement saves real time. RecapGPT extracts the substantive arguments — not just topic headlines — so you can follow Patel's line of questioning without the full runtime.

Related

More ways to use RecapGPT.

Stop watching. Start reading.

3 notes free every month. Pro is $5.99/mo. No credit card required to start.

Get started — free →