Roudy B 🇭🇹 retweetledi
Roudy B 🇭🇹
7.4K posts
Roudy B 🇭🇹 retweetledi
god i’ve watched this entire thing twice now.
krishna is a force at anthropic & he’s masterfully given them a winning hand.
the amount of alpha in this episode:
> the #1 place anthropic spends compute isn’t model training or customers… it’s the research team. he explains:
- research team uses compute to build a better model
- better model uses LESS tokens so the same amt of compute can serve MORE people in the future.
“we could’ve earned billions more in revenue but then we would’ve lost the model race”
> anthropic’s the only lab to train AND inference their models across 3 different chip architectures.
krishna: “one morning we will run an inference block on trainium and by the afternoon it’s used to train claude”
the unique advantage = they don’t rely on a single supplier.
> krishna says people don’t fully comprehend how advanced each model version gets. agents, long-horizon tasks, cost effectiveness improve ALL AT ONCE
“we went from $9B to 30B in months” - that wasn’t just because the chatbot got better.
bonus:
krishna uses claude to generate the companies monthly financial report. it does it in 30 mins and is 95% accurate.
the highest token-consumer on his team is the tax guy.
there’s so much more i want to share but honestly just watch the interview
@patrick_oshag you killed it
Patrick OShaughnessy@patrick_oshag
Krishna Rao is the CFO of Anthropic, and this is his first podcast appearance. He joined the company two years ago when run-rate revenue was about $250M. Today it is $30B. He has helped raise ~$75B and is responsible for the procurement and allocation of compute. I feel lucky we get to hear what it is like to sit inside a company this consequential at a moment this pivotal. We discuss: - The cone of uncertainty - How he allocates compute across Trainium, TPUs, and GPUs - What investors misunderstand about model companies - Why the returns to frontier intelligence keep rising - Platform vs application and where Anthropic builds its own products - How Anthropic uses Claude internally I have asked my closing question about the kindest thing more than 500 times. Krishna's answer is one I have never heard before. Enjoy! Timestamps: 0:00 Intro 2:38 The Compute Canvas 6:51 The "Cone of Uncertainty" 11:58 Why the Returns to Frontier Intelligence Are So High 16:45 Recursive Self-Improvement 20:20 Scaling Laws 23:30 Sourcing $100 Billion in Compute 28:05 Platform vs. Application Strategy 32:52 Pricing Dynamics 38:48 How Anthropic’s Finance Team Uses Claude 43:24 Raising Capital & Overcoming Investor Skepticism 52:32 Public Perception, Risks, and Government Regulation 57:25 Mythos Release 1:12:33 What Could Derail the AI Revolution? 1:13:47 Biotech and Healthcare 1:15:31 The Kindest Thing
English
Last Friday “I feel silly for keeping my daily briefs in .md for so long. I should have created a command center”
Last Sat “Claude Cowork Artifacts “one click for me to connect to your tools and create a command center”
This Friday…
Thariq@trq212
HTML is the new markdown. I've stopped writing markdown files for almost everything and switched to using Claude Code to generate HTML for me. This is why.
English
Exactly how I felt about this, and so was the interview he did on Lenny’s too! So much clarity
carlisle@0xcarlisle
This was so extremely good my god. I wasn’t planning on listening to it since I’m on a trader and investor kick, but you could pull and try to deploy a thousand bits from this to your life. I will actually probably listen to this again.
English
@lennysan @_catwu @AnthropicAI It happens every time. Every single time there’s an episode that really resonates with me above the rest, they mention loving history
English
How Anthropic’s product team moves faster than anyone else
I sat down with @_catwu, Head of Product for Claude Code at @AnthropicAI, to get a peek into their unprecedented shipping pace, how AI is changing the PM role, and how to be the right amount of AGI-pilled.
We discuss:
🔸 How Anthropic’s shipping cadence went from months to weeks to days
🔸 The emerging skills PMs need to develop right now
🔸 Why you should build products that don't work yet—then wait for the model to catch up
🔸 Why a 95% automation isn't really an automation
🔸 Cat’s most underrated AI skill (introspection)
🔸 What Cat actually looks for when hiring PMs now (hint: it's not traditional PM skills)
Listen now 👇
youtu.be/PplmzlgE0kg
YouTube
English
Roudy B 🇭🇹 retweetledi
This Mets fan feeding his dog a hot dog during the game 😂
English
4/18/25. Elite talent scout Boosie Bad Azz
Boosie BadAzz@BOOSIEOFFICIAL
IF WARRIORS DONT WANT KAMINGO THE HAWKS WILL TAKE HIM ✔️DUDE 🔥🔥🔥
English
Roudy B 🇭🇹 retweetledi
Roudy B 🇭🇹 retweetledi
Roudy B 🇭🇹 retweetledi
Ant on going for blocks:
"I just be watching Dwyane Wade clips. He would beat shit at the rim so I just be trying to beat shit at the rim that's all. If Rudy ain't down there, I'm trying to beat that shit. I don't care who it is"
English
@ohnohedidnt24 Me after finally using Claude design today (and linear agent too)
English
KAT enjoyed the same Gatorade too lmao
Oh No He Didn't@ohnohedidnt24
Ant after drinking the Gatorade: "Fuck with this one" 🤣
English
Roudy B 🇭🇹 retweetledi
This is how I feel about Bill Simmons
Nate Jones@JonesOnTheNBA
Since he first broke out during the 2008 Tourney, Steph has been must see TV. Nearly two decades of being a high entertaining/box office player. The thing I respect about guys like LeBron and Steph is that two decades in they still take pride in delivering that kinda experience.
English
Oh great and powerful @DarioAmodei - builder of minds, father of Claude. I humbly request you leave payroll to us at Deel.
We are but simple folk who process paystubs and chase compliance deadlines. But if you do come for us, call me first 🙏
English
The day this app makes me people pay for their takes rather than the other way around will find its way in the history books
English
Roudy B 🇭🇹 retweetledi
if a lebron and bronny two man game isn’t cool to you, idk man
English
Roudy B 🇭🇹 retweetledi
Every B2B software company is (or should be) building a "headless" version of their product. One that can be used by agents.
But "headless" doesn't mean "brainless".
You don't just wrap your existing APIs into an MCP server and call it a day.
The companies that succeed in the agentic era are those that take a thoughtful approach to *designing* an agentic user experience (AUX).
Yes, that will likely involve APIs, MCPs and CLIs.
But the difference will be in the *ergonomics* of the interface. We need to figure out *how* agents actually want to use our products/platforms. Because if all they wanted to do was use them like humans do, we have "computer use" for that.
I'm personally very excited about this new agentic world when it comes to B2B software.
HubSpot is all-in on building the #1 agentic customer platform.
Just posted this in a private Slack thread with the HubSpot exec team:
Being agentic is not just about agents running *on* our platform, it's about agents *running* our platform (being able to operate it). That's how you take AI from being a simple tool to a savvy teammate.
English
@geoffintech Not included but important to call out, the best way is to have the MOST AI pilled CPO & CTO in tech drive the change everyday.
English
Roudy B 🇭🇹 retweetledi
