“Best” is usually a matter of taste. We use a count instead. For each show, we count the quotes we hold that are checked word for word against the episode. More checked quotes means more lines from founders, chief executives and researchers.
A count measures how much leaders say on a show. It doesn’t say whether you’ll enjoy it.
The twelve, as of Nov 14, 2026
| Rank | Show | Checked quotes | Latest episode |
|---|---|---|---|
| 1 | All-In | 261 | Nov 10, 2026 |
| 2 | MTS | 181 | Nov 13, 2026 |
| 3 | TBPN | 161 | Nov 13, 2026 |
| 4 | Lightcone | 158 | Nov 11, 2026 |
| 5 | Training Data | 155 | Nov 5, 2026 |
| 6 | Sourcery | 154 | Nov 13, 2026 |
| 7 | Latent Space | 153 | Nov 10, 2026 |
| 8 | a16z Podcast | 135 | Nov 13, 2026 |
| 9 | Moonshots | 116 | Nov 13, 2026 |
| 10 | Lenny’s Podcast | 107 | Nov 8, 2026 |
| 11 | Lex Fridman Podcast | 104 | Nov 10, 2026 |
| 12 | No Priors | 99 | Nov 12, 2026 |
The full list of the shows we follow, with their counts, is on our coverage page.
A recent line from four of them
On TBPN, Dylan Patel of SemiAnalysis named a gap in export rules.
Selling data is something that has not had any scrutiny yet
On Lightcone, Amjad Masad said how Replit expects agent use to grow.
our prediction internally is that you go from everyone using one agent to 10 agents, to 100 agents, to 10,000 agents at some point.
On Training Data, Ronak Malde talked about keeping people in the loop.
you need to allow these models and build the algorithm so that it keeps humans in the loop.
On Lenny’s Podcast, Tulsee Doshi of Google DeepMind gave product teams a place to start.
start with even just five prompts, right? Like start, you don’t have to start large.
How the count works
We follow each show by name. Every quote is checked word for word against its episode, and only checked quotes count. The counts grow every week, so this list will change.