"I think there's tremendous alpha in the largest cap."

Alex manages $17B and runs one of the best performing hedge funds over the last 3 years.

"There's a huge structural underweight of the largest tech companies in the world.

We also realized that a lot of our performance over the years was from some of the largest companies.

There's a belief that there's no alpha in large cap, so the largest pools of capital are underweight this.

They have a lot of small and mid managers that are stock pickers because it's intuitive that large cap can't have alpha.

In tech, the leader usually grows bigger and wins and develops very high market share quickly, and there's great competitive advantages.

So this is gonna lead to massive profit pools and massive market caps.

Most endowments are betting against this because they're completely underweight this.

I think there's tremendous alpha in the largest cap because a small cap, it just takes one person to figure out it's good and move it up.

But it takes a hundred diversified PMs to realize Google's not a loser, it's a winner.

And can we figure that out before 95% of those generalist PMs, and we've been able to do it."

Another great example of the alpha in seeing things firsthand.

Alex explains the difference between spotting consumer and enterprise trends:

"Andy Grove says when you have strategic inflection points, you can't trust the data.

And strategic inflection points are about intuition and anecdotal evidence.

We invested in the mobile video game S-curve for so long.

The screens were small and the processing power wasn't good, so you had all these casual games.

But then I was in China and I saw this little 12-year-old boy with a huge phone playing an awesome video game.

I'm like, "It's now coming to the phone." So it's visual.

Enterprise is hard because you can't see it.

So we go to the Gartner IT Symposium, 30,000 American CIOs go there.

We saw this happen with Splunk - the room where they were explaining it was standing room only.

And with AWS, we went there and the grand ballroom was completely packed, and that was at nine o'clock. And at 10 o'clock the grand ballroom was completely packed.

So you could see the demand exploding before it happened."

Alex Sacerdote on the de-commoditization of hardware:

Three years ago, "the sole supplier of the Google TPU server" was trading at 8x earnings

"For the past forty years, nothing has changed in the data center.

The whole industry basically commoditized every part. There was no innovation.

Now you go to AI, the workloads are growing 10x every year, and they're pushing every single aspect of this hardware to the physical limits of what it can do.

Not only are you creating tremendous unit growth, but we call it the de-commoditization of the hardware industry.

Celestica was a contract manufacturer. This has been a disaster industry since 1999... And then we noticed they were the sole supplier of the Google TPU server three years ago.

To do an AI server, it's two or three hundred thousand dollar piece of machinery, whereas an old server was five thousand dollars. If it breaks, you just throw it away.

If this thing breaks, the whole thing goes down.

So you become a commodity like supplier to like selling a critical part on a plane. You'll never get swapped out.

It turns out these are excellent businesses...The stock was trading at eight times earnings.

We see that up and down the supply chain.

Corning, they make the fiber...this Microsoft data center they just built, there's enough fiber to circle the world four and a half times in that one thing.

And it's higher margin, and the fastest-growing part of their business.

In the power supplies, every Nvidia chip or rack uses 50 to 125% more power. That drives the ASPs of Delta and Advanced Energy.

I can't believe these stories when I hear. I'm like, 'Wait, so your ASPs are gonna like go up 40% for the next four years in a row, and it's higher margin?'"

he's wrong on open source models but an interesting perspective none-the-less.

Another interesting factor he doesn't seem to consider is the utility cieling, which is that once you improve coding as much as we have currently, improving it even further reaches a ceiling of human limitation on humans speccing things. So a better model isn't really going to impact coding that much further, only because SOTA models have already increased a coder's output by 20-30x. Maybe you'll get like 1 or 2% improvement with better models. It does start making an impact in other industries like cybersecurity (ie, mythos) but that requires modeling it differently.

I think 20k per engineer is not in our future.