Kevin Novak joined Uber as the 21st employee, seventh engineer in 2011, and in 2014, he was the head of the company’s data science. He spoke proudly at that time, but like all the good things, it was running the course and at the end of 2017, after achieving what he wanted in the company, he left.
At first, he took the rate of angel investment, worked he had begun to focus on weekends and evenings, in the end building a portfolio of more than 50 startups (including Fintech pipes and standard cognition of the checkout company).
He also began advising startup and venture companies – including global playground, Costoya Ventures, Renegade Partners and Collective Data – and after falling in love with work, Novak this year decided to launch its own business clothing in Menlo Park, CA., called the capital of Ventura Rackhouse. Indeed, Rackhouse just closed his debut fund with $ 15 million, anchored by the first Uber Engineering Head, Chambers Curtis; Steve Gilula, former chairman of the spotlight, and the Candal Capital fund fund. Many Novak VCS knows also investors in funds.
We followed Novak last weekend to chat with that new vehicle. We also talked about this term of office in Uber, where, was warned, he played a major role in creating a surge price (although he preferred the term “dynamic price.”) You can hear more complete discussions or check the quotes from it, edited lightly For length and clarity, below.
TC: You plan to become a nuclear physicist. How do you end up on uber?
KN: As a scholar, I am learning physics, mathematics and computer science, and when I arrive at graduate school, I really want to teach. But I also really liked programming and implementing the concept of physics in the programming room, and the Nuke Department had the greatest allocation of supercomputers, so finally driving a lot of my research – only the opportunity to play on computers while doing physics. So [i] learned to be a nuclear physicist funded indirectly through research which eventually became a boson higgs. When Higgs was found, it was very good for humanity and was truly terrible for my research budget.
A friend I heard what I did and sort of knew my skills and said, like, ‘Hey, you have to check this Uber taxi company that it’s like a limousine company with the application. There are very interesting data problems and very interesting math problems. ‘So I finally applied [even though I was committed] Cardinal sin of the startup application and wearing a suit and binding with my interview.
TC: You are from Michigan. I also grew up in the Midwest so appreciative why you might think that people will wear a suit to an interview.
KN: I got off the elevator and friends who pushed me to register like, ‘What are you wearing?! But I was asked to join as a computational algorithm engineer – a title that preceded the data science trend – and I spent the next few years living in the world of engineering and products, building data features and. , . turn our ETA machine, basically predicting how long it will take uber to get to you. One of the first projects I was on a toll road and tunnel for finding out the tunnel where Uber passed and how to build time and distance was a common point of failure. So I spent, like, three days directed large excavations in Boston to Somerville and returned to Logan with many phones, collecting GPS data.
I have to know a lot of random facts about uber cities, but my big claims for fame is a dynamic price. , , And it turns out it is a very successful foundation for strategies to ensure ubers available.
TC: How does that happen, when you tell people that you find a surge price?
KN: It’s a very fast litmus test to find out like the enthusiasm of the underlying person for econ behavior and finance. Wall Street crowds like, ‘God, it’s very cool. “And then many people like,” Oh, thank you, yes, thank you, amazing, you bought a round of the next round drink. , , , [Laugh.]
But data is also an incubation space for many initial specific projects such as uber pool and lots of ideas around, okay, how will you build a shipping model that allows people who are different from the request of travel collected? How do you collect it efficiently in space and time so we can get the right match rates that [so this] profitable project? We do a lot of work on the theory behind the delivery model of hub-and-spoke uber eat and think through how we apply our learning about traveling for food. So I got the first person’s perspective on many of these products when literally three people wrote notepad or riffing on the laptop at lunch, [and that] finally continued to become this big and national business.
TC: You’re working on uber shipments for the past nine months in your career with Uber, so there when business with Anthony Levandowski exploded.
KN: Yes, it is a very interesting era for me because of more than six years in, [I have developed attitude] ‘I have done everything I want to do.’ I joined the company of 20 people and, at that time, we approached 20,000 people. , . And I rather miss a dynamic little team and feel like I hit a natural termination point. And then Uber 2017 happened and and there was Anthony, there was Susan Fowler, and Travis had this terrible accident in his personal life and his head was clearly not in the game. But I don’t want to be a known person because it caught fire in the worst quarter of the company’s history, so I finally spent the following year basically keeping the band together and trying to find out what I could do to keep anything small part of my company ran intact and motivated and empathic and good in everything.
TC: You go at the end of the year and it looks like you’ve been very busy since it, including, now, launched this new fund with outsiders support. Why call it Rackhouse? You use the Jigsaw brand venture capital when you invest your own money.
Kn: Yes. A year [into an angel investment], I have formed LLC, I “mark my portfolio to the market, sending quarterly updates to me and my wife and my wife. It is one of these exercises which is the accumulation of how I train the manager, because I think you grow most efficiently and succeed if you can develop some skills at a time. So I tried to find out what was needed to run my own back office, even if it only moved my money from my current account account to “investment account,” and wrote my own portfolio update.
I was really excited about the possibility of launching the funds faced by my first externally with other people’s money under Jigsaw banners too, but there were actually funds in England [named Jigsaw] and when I started talking to LPS and said ‘See’, I wanted Doing this data fund and I want it to be the initial stage, ‘I will get a call from them like,’ We just saw that Jigsaw did this series in Crowdstrike. ‘I realized I would compete with other jigsaw from the perspective of the mind, so I thought before everything went too big and crazy, I would make my own brand.
TC: Do you throw one of your angel offers into new funds? I see Rackhouse has 13 portfolio companies.
KN: There are some that I agree to move forward and warehouses for these funds, and we are only through the technical to do that now.
TC: And the focus is on engine learning and AI.
KN: that it’s true, and I think there are extraordinary opportunities outside the traditional field of industry focusing that, as far as you can find like a strict AI application, it will also be significantly less competitive. [Offer] that doesn’t fall in the strike zone is almost as much as [venture] the company is the game I want to play. I feel that the opportunity – regardless of the sector, regardless of geography – bias against domain expert.
TC: I wonder if it also explains the size of your funds – you want to get out of the strike zone of most venture companies.
KN: I want to make sure that I am building funds that allow me to become active participants in the earliest stage of the company.
Matt Ocko and Zack Bogy are my good friends – they are mentors, in fact, and small LPS in funds and talk to me about how they started. But now they have one billion-plus [dollar] in assets under management, and he’s the people I [like to come back] are two people who are moonlighting and getting ready to take risks and [company collective size data] basically has price. themselves out of the formation and pre-superior stages, and I like that stage. That is something where I have a lot of useful experience. I also think it’s a stage where you come from where the domain expertise, you don’t need five financial quarters to get confidence.