Podcast Summaries & Show Notes

Show notes for any podcast episode

Paste a link from Apple Podcasts or Spotify. Get the summary, the chapters, the quotes and every book, person and product mentioned, each linked to the moment it was said.

Free plan, no card needed

Real show notes of a real episode: open the chapters, see everything mentioned, ask it something

Lenny's Podcast

Why PMs are suddenly valuable again | Tulsee Doshi (Head of Product, Gemini)

Nov 16, 2026 · 59 minTranscribed from the audio

In one sentence

Gemini's head of product explains why models stay unpredictable, how to judge them with small evals and good vibes, and why PMs with taste matter again.

Lenny Rachitsky talks with Tulsee Doshi, head of product for Google's frontier models. They cover how models are shaped but never fully predicted, Google's full-stack strategy, safety through monitoring, the model's personality as its UX, what she looks for in PMs, how to start writing evals, the 60/20/20 rule for building ahead of the models, and how her team learns new tools without burning out.

Who's talking

  • Lenny RachitskyHost · Host of Lenny's Podcast
  • Tulsee DoshiGuest · Head of product, Google's frontier models (Gemini)

Speakers are inferred from the conversation.

Key points

  1. Model development has goals and evals, but the result always holds surprises; a model PM studies what comes out and builds on where it shines.
  2. Google's strategy is full stack, from hardware to products with over a billion users, with models across the cost-quality curve.
  3. The model has a UX of its own: how transparent it is, how it communicates and what it asks back; a behavior spec becomes rubrics, data and reward models.
  4. When building is cheap, taste and the ability to turn chaos into order are what make a PM valuable.
  5. Good evals start tiny, five to twenty prompts with yourself as the rater, and the judge, the serving stack and filters all affect the result.
  6. Build ahead of the models: most of the product works today, some almost works, and some waits for the next model.

Main claims

  • Google has 14 products with more than a billion users (she said she should check the number).

    Tulsee Doshi

  • The six-tone Fitzpatrick scale was built for healthcare and skews light, so it cannot represent a global population.

    Tulsee Doshi

  • In his survey of tech workers, more hours did not mean more burnout.

    Lenny Rachitsky

  • Chain-of-thought monitors that flag misalignment give her more confidence in deploying models.

    Tulsee Doshi

Quotes

  • The model isn't predictable.
    — Tulsee Doshi
  • Taste is now really important.
    — Tulsee Doshi
  • Start with even just five prompts.
    — Tulsee Doshi
  • What you should be providing is the insight. What the model should be providing is the data.
    — Tulsee Doshi

Takeaways

  • Write a mini eval this week: five prompts you care about, rated by you, run on two models.
  • Before doing a task twice, ask whether a model can do it, and share what works as a reusable skill.
  • Retest your harness with each new model; old workarounds can hold a better model back.
  • In AI-written docs, keep the data from the model and the insight from yourself.
  • Gemini
  • Product management
  • Evals
  • AI safety
  • AI agents
  • Burnout

How it works

  1. 1

    Paste an episode

    An Apple Podcasts or Spotify link, an RSS feed, or just the show's name. Or ask ChatGPT or Claude about an episode.

  2. 2

    We read or transcribe it

    The show's own transcript when there is one, otherwise the audio is transcribed. About a minute per hour of audio.

  3. 3

    Read, jump, ask

    Skim the show notes, jump to a chapter, open everything mentioned, or ask what the guest said about something.

The episode, without the hour

Any show, any app

Apple Podcasts, Spotify, an RSS feed or the show's name. Interviews, news, comedy, in any language.

Real show notes

A one-line TL;DR, an overview, who is speaking, the key points and what to take away.

Everything mentioned

Every book, person, company, product, paper and website, with what was said about it. Copy the list in one click.

Chapters on a timeline

See how the episode is built and jump straight to the moment a topic starts.

Claims and quotes

What was claimed and by whom, and short quotes with the second they were said.

Ask the episode

"What did she say about evals?" Get an answer with the quotes it rests on.

Catch up on a show

Fell behind? The latest episodes of a show, each with its summary and key points, at once.

Follow shows

Each new episode of the shows you follow is summarized and lands in your inbox.

Inside your assistant

Ask about an episode in ChatGPT or Claude and read the show notes right in the conversation.

Turn your next episode into show notes

Summarize your first episode free

Just ask your assistant

Once the connector is added, talk to ChatGPT or Claude the way you normally would.

  • Summarize this episode: https://podcasts.apple.com/…
  • Which books were mentioned in the latest Lenny's Podcast?
  • Catch me up on the last 3 episodes of Hard Fork
  • What did the guest say about pricing in this Spotify episode?

Use it in ChatGPT and Claude

Add Podcast Summaries & Show Notes as a connector, then say: summarize this episode.

Connector address

https://podcasts.inyour.chat/api/mcp
How to connect

Frequently asked questions

Which podcasts work?

Any public podcast with an RSS feed, which is nearly all of them, including shows on Apple Podcasts and Spotify. Spotify exclusives without a public feed do not work. The show notes can be written in your language even when the episode is in another.

Is it free?

Yes. You can open 3 episodes a week, each up to 120 minutes long. Reading them again, their chapters, everything mentioned and questions are never limited.

How do you get the transcript?

When the show publishes its own transcript in its feed, we use that. Otherwise we transcribe the episode's public audio, which takes about a minute per hour of audio. An episode opened by anyone before is ready at once.

Can I use it in ChatGPT and Claude?

Yes. Add Podcast Summaries & Show Notes as a connector and sign in. Your episodes are the same in ChatGPT, in Claude and on this website.

Can I follow shows?

Yes, with Pro you can follow up to 20 shows. When a new episode comes out, its summary is made and emailed to you.

More episodes than hours? Read them.

Start free