Twitxy Saved

Leo

Verified account This account is verified. Learn more

@leoporuro

fundception, pm @ThetaInvest
Joined October 2009
624 Following 568 Followers 7.3K Posts

Latest posts 5

Leo Verified account @leoporuro
On sourcing alpha: I've grown skeptical that AI-powered sourcing engines are a real source of it.

These tools have commoditized the very alpha they were built to capture. If a deal surfaces through an engine, it's surfaced for everyone. It is arguably worse as you risk adverse selection: the teams most legible to AI-driven discovery are often the ones who've optimized for being found, not for building.

Real sourcing alpha (historically) always comes from other founders and from differentiated networks where you have genuine access.

I don't think these tools are useless. They've raised the floor on sourcing coverage across the industry and every serious fund should have them. But that's exactly why they can't be a source of alpha.

So the real work now is in two places. First, solving for adverse selection: building relevancy and scoring layers that filter out teams who are good at being found rather than good at building. Second, process: making sure the right resources flow to the right opportunities at the right time given the volume of deals is so high.
Leo Verified account @leoporuro
Dispersion ✨❤️✨
Ruben Verified account @rdominguezibar Oct 25
Founders Fund II is in a league of its own

a $220m fund from 2007 that has created $37.4bn

that's 183.5x per dollar at risk. The median across 408 ranked funds is 1.3x

1️⃣ Founders Fund II, 183.5x, @SpaceX
2️⃣ Bullet Time Ventures I, 53.3x, @Uber
3️⃣ Founders Fund III, 52.2x, @PalantirTech

more than 3x any other fund on the list. INSANE

Source: @JoinOdin via @hkanji

Is this skill or one SpaceX-sized bet? @foundersfund
View quoted post on X
Leo Verified account @leoporuro
This is the first quarter that I can clearly see the AI productivity gain in our fund managers: LP letters for Q3 have been sent way earlier than before.

I guess that Q1 and Q2 were periods of experimentation and by now, all processes for data capture, valuations, aggregation of news/data have been incorporated.

The challenge now is (1) reading like a human and (2) prioritizing relevancy/substance over trying to sound smart. Original ideas are scarcer by the day!