The elephant grew. The stake stayed put.

# 📬 POD CAPO — Issue #021

*Your ears rang. We took notes. You're welcome.*

Friday · 6 min read · Lightly edited by a human who definitely didn't listen to 2.5x speed (lie)

---

Good morning, capo. This week was about the Sphinx and the stake. On 20VC, ClickHouse CEO Aaron Katz dated a billion dollars of ARR (over/under December 2027; he takes the under) while the ladder already runs 0 to 1 to 50 to 200, north of 500 this year. One host on My First Million brought the elephant that grows strong enough to rip the stake and still doesn't.

David Senra, eight weeks running, told the story of Henry Singleton, the Sphinx who bought back over 90% of Teledyne and ignored the herd on purpose. Also, about 250 humans at the AI frontier, a Nvidia quarter that prints like a factory, and CFOs shopping for dumped models. Let’s go count someone else’s money, especially the money still tied to a stake somebody outgrew.

20VC — “ClickHouse CEO: AI Margins Need to Improve” ([listen](https://www.youtube.com/watch?v=Wna2Ds4dt50))

1h+ → 90 seconds

Aaron Katz, co-founder and CEO of ClickHouse, the open-source OLAP database now used by nearly every AI-native shop he can name. The cold open is already a ladder: ”We’re just getting started. This seems to be accelerating at an unprecedented pace. We haven’t seen revenue growth like this in our lifetime. We went 0, 1, 50, 200 and we’ll finish this year north of 500.”

Then the clock. ”We need to get to a billion dollars of ARR as quickly as possible. I’d put the overunder at December 2027, and I would take the under.”

Who pays for the picks and shovels, in his telling: ”It’s the world’s most popular open source database… and it’s used by nearly every AI native company, Anthropic, OpenAI, Weights and Biases.” Asked how this cycle compares to internet, mobile, social as a builder rather than a pontificating VC: ”Those cycles in my experience were much more gradual. This seems to be accelerating at an unprecedented pace.” Concentration? The AI basket on ClickHouse is under 12% of revenue. He still treats anything over 10% like an operator problem, not a vibe.

Takeaway for operators: put a dated ARR number on the whiteboard and take a side. If your growth story only works as a mood, you are not in Katz’s bet. Pull concentration below the threshold you would actually lose sleep over, then ask whether the ladder still climbs when half the AI logos churn. Own that math. Or stop borrowing theirs.

**My First Million — “7 things Bezos, MrBeast & Thiel do that

you don’t”** ([listen](https://www.youtube.com/watch?v=jZT04e4yBb0))

1h+ → three things worth stealing

One host opened with the diagnosis: ”There’s a good chance you’re not being hardcore enough.” The product is a pile of CEO moves ”that sound wrong, sound almost illegal, but they actually work.” He calls the collection frame-breaking. You see the world inside a border, then a story lands outside it, and the frame has to expand.

1. Study the elephant on the stake. ”The elephant who as a young age like they tether him to the stake in the ground and then he grows and now he’s so powerful he could easily rip out the stake but he doesn’t.” The host’s point is the illusion of limits. Most of the “I couldn’t possibly” list is a stake you outgrew and never tested.

2. Clone the lieutenant. MrBeast, as the host retells it, on how he trains people who never went to business school: ”Oh, I just do cloning.” The table’s reaction: ”That sounds illegal. What does that mean?” The method: ”If I hire you in one of like kind of the key roles, you just follow me everywhere.” Shadow until the clone can run the room. One early hire was allegedly the alarm clock.

3. Ship the one-bit escalation. Bezos as email router. Customer mail in, one character out to the team: a question mark. Encoded meaning, per the host: is this true, why is it happening, what are we going to do about it. Annoying on purpose. Fast on purpose.

Takeaway for operators: pick one stake you have been polite about and rip it this week. Then pick one key hire and make them follow you for 72 hours with no curriculum, just proximity. If your escalation still needs a paragraph, you do not have a question mark yet. Build the one-bit version.

Founders — “How Henry Singleton Worked” ([listen](https://www.youtube.com/watch?v=fC6ICHKJBXU))

1h+ → three quotes and a vibe

Senra’s eighth week running, and the object is a Sphinx. Henry Singleton: world-class mathematician, chess blindfolded, MIT’s early computers, radar-dodging tech in the war, then Teledyne.

Conglomerates were the internet stocks of the 1960s. He ran his differently. Senra, via Thorndike: ”He aggressively repurchased his stock, eventually buying over 90% of Teledyne shares. That’s nuts.” The scoreboard: ”The annual compound return to his investors was an extraordinary 20.4%. If you had invested a dollar with Singleton in 1963, by 1990… it would have been worth 180.”

Munger, as Senra stamps it: returns ”a mile higher than anyone else’s… utterly ridiculous.” And the raw IQ line: ”Henry Singleton was the smartest single human being I’ve ever known in my entire life.” The nickname was the Sphinx because he would not perform for analysts or journalists. When Fortune 500 peers finally piled into buybacks late, Singleton’s reply was the whole capital-allocation religion in one sentence: ”If everyone’s doing them, there must be something wrong with them.”

The vibe is not inspiration. It is a man who treated the herd as a contrary indicator and bought the company back from the people who wanted a story. The stake, for once, got pulled.

Invest Like the Best — “She Knows the 250 People Building AI. Here’s What They Actually Believe.” ([listen](https://www.youtube.com/watch?v=hY6S__xeCjg)): Patrick restated what Conviction partner Mike maps: ”There’s sort of 250-ish people that he thinks about… that are what some combination of entrepreneurs and researchers… building AI.” Conviction’s Sarah, on the tape, signs the great-person bet: ”I believe in like the great man and great woman theories of history… if you have very high agency people… you change outcomes.” The FOMO she relayed from an investor friend: ”I keep saying I want to press the brakes as hard as I can, but I’m not doing it… going 90 miles an hour.” The frontier is not a market. It is a contact list.

All-In — “Nvidia’s Historic Quarter / SaaS Comeback” ([listen](https://www.youtube.com/watch?v=1u5dMAKl_ks)): The table opened on the print, not the culture war. One host on the quarter: ”96.2 billion in revenue… up 106% year-over-year… 60 billion in net profit. That is the most profitable core business quarter of any public company ever.” Chamath on why Salesforce ripped with it: models were phase one, harnesses phase two, and the next step needs systems of record with context.

All-In — “Nvidia’s Historic Quarter / SaaS Comeback” ([listen](https://www.youtube.com/watch?v=1u5dMAKl_ks)): The table opened on the print, not the culture war. One host on the quarter: ”96.2 billion in revenue… up 106% year-over-year… 60 billion in net profit. That is the most profitable core business quarter of any public company ever.” Chamath on why Salesforce ripped with it: models were phase one, harnesses phase two, and the next step needs systems of record with context. Freeberg’s operator translation after trying to rebuild a CRM over a weekend: spend the hours on software unique to your vertical, not on recreating Slack, Gmail, or Salesforce. His label: a vertical SaaS apocalypse, not a horizontal one. The factory is still printing. The roadkill narrative is the thing that died.

Prof G — “China Is Dumping AI to Crash Silicon Valley” ([listen](https://www.youtube.com/watch?v=JSk1-GBKeQU)): Galloway put the business thesis on the table and asked Josh Tyrangiel to swing. ”It’s engaging in massive AI dumping with cheap subsidized LLMs that CFOs across America that are increasingly asking difficult questions around ROI are going to opt for the lesser model that costs 1/35th of the frontier model.” The compression line: ”Beijing is going to do to Silicon Va/codelley in 30 weeks what Tokyo did to Detroit in 30 years.” Josh: ”Totally possible… very plausible,” then added distillation as ”essentially bootlegging.” His claim, not ours. Mine the CFO question. Skip the rest of the shouting match.

The Sphinx ignored the herd. The elephant is still practicing politeness. We remain long the dated bet and the lieutenant who wakes the boss up.

If you only listen to one episode this week, make it the 20VC Katz tape. A dated billion, a lived ARR ladder, and a builder’s answer to the bubble sermon.

If you only have 20 minutes, the My First Million hardcore reel. You will walk away knowing what an elephant on a stake sounds like when a CEO is still tied to it.

Senra has now surfaced eight weeks in a row. Thiel, the merchant bankers, Nolan, Plank, Shannon, Dylan, Hopkins, and now Singleton. He is no longer a guest on the calendar. He is a cryptid, and we would not have it any other way.

Forward this to a founder still polite about a stake they outgrew. Or don’t. We’re a newsletter, not a

— The Capo 🎙️🤌

Made with coffee, too many podcast apps, and the firm belief that nobody has time to listen to 15 hours of content per week. That's our job now.