Moneyball

I grew up in the Bay Area during the Moneyball era and fell in love with statistics at an early age. I was obsessed with the idea that statistics could help find inefficiencies in baseball. When I went to Stanford, I studied math as an undergrad and statistics as a grad student, which also provided my introduction to statistical and machine learning.

That love of stats has stuck with me in my work as an investor. I think in terms of probabilities when I’m working through outcomes. And I see my job as doing everything I can to shift the distribution to increase a founder’s odds of success. When you’re starting a company, you’re going against these large behemoths, which can be very intimidating. But if you look at the inefficiencies and try to capitalize on them, then you have the potential for a promising startup.

Strategic technology

I saw the impact AI could have on society early in my career, both in terms of problems that humans can’t solve (like real-time fraud detection or recommendation systems) and in automating some human tasks within larger workflows. I joined an AI-focused incubator in 2018 following the deep learning wave and saw this impact first-hand by applying AI to traditional industries like agriculture and logistics. Over the past few years, we saw a dip in enterprise adoption of AI, followed by a massive surge after the launch of ChatGPT. I believe the new wave of AI startups will fulfill the promise I originally saw in the technology. Over the next decade, AI will become central to the development of new software. I want to stay on the front lines and partner with companies that are finding new ways of building offerings around foundational models.

I invest in cyber for similar reasons. Cybersecurity presents a critical battleground, not only for enterprises but geopolitically as well. And increasingly sophisticated AI-powered offensive cyber from nation states and other bad actors makes continued defensive development all the more important.

Bridging the gap

Great technical or product chops are essential for founders. But so is the ability to recruit and inspire others. To get a sense of that, I prefer to meet several times, in person, with founders to see if we’re a good mutual match. Seeing how their ideas evolve over time gives me a good sense of their velocity and how strong of a leader they could be.

Aviv from Kodem is a great example of this. He worked at a large cybersecurity company for six years as a security researcher. He felt like he had an ability to build game-changing technology that could help governments and enterprises protect themselves from cyberthreats. In our conversations, I got the sense that he was passionate about Kodem and saw the big picture impact it could have. Because that passion was infectious, I knew he’d make a strong recruiter and inspirational leader.

Jerry from LlamaIndex is another example – meeting him as he built a leading open-source project and before he started a company convinced me he would be a great community builder and sharp commercial thinker.