Hi, I'm Jeremy Liew. I'm a partner at Lightspeed Venture Partners. I've been there since 2006. Some of my notable investments would include Snapchat, BeReal, Affirm, and Blockchain.com.
I'm 29 now. What advice would you give your 29-year-old self? Do you know what you want to do with the rest of your life? I think there's a lot of folks who think, 'Well, I know what I want to do, but before I can get there, I've got to do these things first.' And I'd say, no, you just go do the thing you want to do and go directly to that thing. The more experience you have doing that thing, the better you will be at that thing. If you had said, 'No, I don't know what I want to do,' that opens up a whole different path. I ended up working with people who I just greatly admired. I felt like I was learning a lot from them. Whenever they went somewhere, I just sort of blindly followed them because often times they go to a new place where they don't have a bunch of people they trust, so you can slipstream behind them in a pretty interesting way. Both of those things can help accelerate your career in a way that you might not otherwise be able to do.
What is a failure that you are most grateful for? Gosh, so many failures. Which one should I talk about? Let me talk about the time I got fired.
I'm your host Sheree, I'm Jean, and we're the Tiger Sisters. What goes through the mind of an investor who has invested in companies that have gone on to become multi-billion dollar companies such as Snapchat, Affirm, and Blockchain.com? In this podcast episode, we have an interview with Jeremy Liew, and then we also discuss the takeaways at the very end where Jean and I go through our biggest learnings in a business school case study method. We'll get to the interview right after this break.
Hey guys, quick break to let you know that we now have merch on sistersmatcha.com. We have sweatshirts and t-shirts that we designed ourselves. Go check it out and please rate us five stars on Spotify and Apple Podcasts. These ratings are so important for the distribution and survival of Tiger Sisters podcast. Thank you for your support.
Hi, I'm Jeremy Liew. I'm a partner at Lightspeed Venture Partners. I've been there since 2006, focused on consumer investing. Some of my notable investments would include Snapchat, BeReal, Affirm, and Blockchain.com. Prior to joining Lightspeed, I worked for a bunch of Web 1.0 companies for about a decade, so Netscape, AOL, CitySearch, IAC, and before that I was at McKinsey for a couple of years.
Awesome, so we'll just jump right into the icebreakers. So Jeremy, what are you listening to now in terms of podcasts?
Oh, I have been listening to Radiolab a lot. I like the way that they deep dive into sciency stuff.
Cool. Next icebreaker question is that when you were a kid, were you a rule follower or a rule breaker?
I was 100% a rule follower. Yeah, I was not a rebel. Not a rebel at home, in the classroom, just a rule follower.
Yeah, I would say so. I would probably say I was a rule exceeder. And what about now? How do you identify?
Um, I don't think of myself explicitly as a rule follower or a rule breaker. I don't really think about rules in that way, I guess. I sort of, at some point you get to make your own rules and you get to follow your interests in those directions, and it's a more comfortable place to be.
I love that. Making up your own rules for your own game. Yes. Okay, so the first question in this series is that I'm 29 now. What advice would you give your 29-year-old self? Do you know what you want to do with the rest of your life?
Do you? Do I? Yeah, yes, I do. Okay, then that's great. Then it's easy. So just go do that thing, right? I think there's a lot of folks who think, 'Well, I know what I want to do, but before I can get there, I've got to do these things first.' And like, you know, if I want to be an investment banker first, I've got to get operating experience, or sorry, if I want to be a venture capitalist first, I've got to get operating experience at a startup so I have the right set of experiences to become a successful investor later on. I'd say no, you just go do the thing you want to do and go directly to that thing. The more experience you have doing that thing, the better you will be at that thing. So just keep going. If you had said, 'No, I don't know what I want to do,' then that opens up a whole different path. But luckily you do know what you want to do, so you're good. Nike was onto something, just do it. Yeah, I mean, if you know, then it's pretty straightforward. Yeah, you can go directly there and you don't have to pass through any intermediary steps. Yeah, if you don't know what you're doing, and I didn't know what I wanted to do, then I don't have general advice for people. I can only tell you what worked for me, which was I, at a number of places in my career, I ended up working with people who I just greatly admired. I felt like I was learning a lot from them, and whenever they went somewhere, I just sort of blindly followed them. And that has a couple of effects. First is if they're really as good as you think, then they tend to make pretty good, well-informed choices about what they're going to do next. Secondly, you can slipstream behind them because often times they're going into a new place where they don't have a bunch of people they trust around them, both trust personally and they trust professionally. And so you can slipstream behind them in a pretty interesting way if you're going with them into those sorts of situations. And so both of those things can help accelerate your career in a way that you might not otherwise be able to do.
Something I'm realizing now is that people are indexing on now, like when they start a company, looking for a co-founder, it's all people they trust and people that they've worked with previously and have a good working relationship with. So yeah, totally, it's finding the people you admire and sticking with them because they can mentor you and bring you up in a different way.
I would say that's a slightly different angle. If you're starting a company with somebody, then the trust needs to go in both directions. If you are following someone, then the fact that they can trust you is a lot more important than... it's not bidirectional in that way. There's a hierarchy to it. And so, you know, that allows you to jumpstart things a little bit more when you're junior, whereas it's obviously much more peer-based if you're thinking about starting a company with somebody.
Fair. That's a good distinction. Yeah. So could you please share with us what your first job was and how it shaped your work ethic? And this could be your first job, like as a kid.
I studied mathematics. I went to the International Math Olympiad on the Australian team in my last year of high school, and I'd always assumed that I would go on to study mathematics and become a math professor somewhere. And I ended up running into this guy who was also on my team, called Terence Tao. And he's probably regarded as the brightest living mathematician today. But I didn't know that when I met him. I saw a 14-year-old kid when I was 18, and he was just so much smarter than me that I suddenly realized, 'Wow, this is what good looks like.' I thought I was good, but this is actually what good looks like as a mathematician, and I'm just not there at all. So I ended up getting my degree, and since I'd seen how far off the standard of excellence was, I was like, 'I got to do something where I can be better at it than this.' So I ended up going into business because the bar is just a much lower. I didn't realize that that was actually the best mathematician of our time. Well, I knew it was like, 'Wow, this random 14-year-old is so much smarter than me, I should do something else.' So I ended up going to McKinsey, and I spent a couple of years there out of undergrad.
So what is a time or what is a failure that you are most grateful for, and what did you learn about yourself?
Gosh, so many failures. Which one should I talk about? Let me talk about the time I got fired. I was working for USA Networks, it's now called IAC, and I was VP of Strategic Planning working for this guy Dara Khosrowshahi, who was the CFO of IAC at the time. He's went on to become CEO of Expedia and now the CEO of Uber. And he was a great boss, like an amazing boss, and I learned a ton from him and I was really grateful and he gave me a lot of opportunity. And so he was the EVP of Strategic Planning or something like that, he became promoted to the CFO. This guy Dan Marriott was brought in to be the new head of Strategic Planning. He's actually a guy that I'd worked with in the past at CitySearch, all part of the IAC group of family. I didn't give him anywhere near enough respect. I was this snotty 28-year-old, 29-year-old who really thought that I was ready for more than I was and didn't like the fact that this guy had been brought in on top of me when I used to have this direct line to the CFO and now there was someone between me and him. And I wasn't respectful and I didn't know how much I lacked in my ability to... I couldn't understand why I couldn't just stay in this role and not have this person between me and the CFO. And I was just very unaware about the presence that I was missing, about the broader scope of experience and intuition and judgment that I was missing. I figured that I could run spreadsheets as well as anybody, and that was the job, right? Of course, it was just a tiny sliver of the job. And over time, my lack of respect for Dan, which he 100% deserved and which I absolutely, if I had been more self-aware, would have realized, led me to a situation where it was not tenable for me to continue in that role. And so I ended up leaving IAC. I was lucky in that another guy at IAC, the president of IAC, so Strategic Planning at IAC basically meant which company should we buy next. So over the time that I was there, we bought Expedia and Hotels.com, we bought Match and CitySearch and Ticketmaster, and we sold the USA Networks, Bravo, Sci-Fi channel, TV studio, movie studio. So it was quite a transformation. So it was a great set of experiences. And the president of IAC ended up having all of these companies that we acquired reporting to him, and he ended up leaving to be the CEO of AOL in 2002. And so I was able to tag along with him and end up as his Chief of Staff, SVP of Corporate Development there. So it turned out that I had a nice sort of everything just fell into place for me, which was very lucky. I had a place to go that was exciting and challenging, gave me opportunities, but I didn't recognize the breadth of experience and capabilities that I needed.
Yeah, and it's helpful to understand the gap between where you are and what you need to be, because then you can start working toward closing that gap. Yeah, thank you for sharing that. It's very clear the reflection that's gone into and a lot of the thinking that's gone into this.
Didn't feel good at the time, right? Of course, at all. Felt quite shitty at the time, you know.
Thank you for sharing that. And I feel like maybe that's where Touchy Feely also helps as well.
Touchy Feely is great. Yeah, yeah.
The next question is about hot takes. So what is a hot take that you have or a belief that you have that might be more contrarian in the VC or business tech world?
So I can speak for myself, not as a representative of Lightspeed or any of the other positions that I hold. I think that there's a good argument for AI companies to be nationalized. So there's a couple of dimensions on that. The first is, and I would say that I hold this strong opinion weakly because I'm still learning a lot about the fundamental structures of the AI ecosystem and I think we're still waiting to see what some of these implications are and how quickly they'll play out. But I think that we are going to see over the course of the next presidential administration significant impact from AI on jobs. I think that a lot of the white-collar jobs that historically have been the path to the middle class in the US are going to get... the productivity is going to get dramatically increased for people who are relatively senior in those roles, and that's going to mean that there's going to be less opportunities for people who are more junior in the roles because that is the mechanism through which productivity was usually enhanced for senior people, right? Like you have this sort of team that reports to you and they're analysts and junior developers. It could be in law firms, it could be in accounting firms, it could be certainly in developer roles. But anytime when your work is primarily digital, like your input and your output is digital, you're not touching a physical thing, you know, fixing a car or doing surgery on a human, but you're mostly working on a computer. So I think we're going to see massive productivity improvements and I think that's going to result in, for at least entry-level jobs, a significant reduction in those. And then I think that will propagate as AI gets better and better and you start to see agentic AI and so forth. And I do think that this is going to happen over a four-year time horizon. We start to see that in the Biden four years. So if you're a company, whether you're an OpenAI or Microsoft or any of the foundational model developers, your only responsibility is to shareholders. And so you're excited about the prospect of replacing a $150,000 salary with a $20,000 AI agent, whatever, because that's just revenue to you. But the externalities of that, which is significant job loss, are not borne by the shareholders, so you don't have to worry about it. Those are going to get borne by society. It's going to get borne by the government. And so nationalization would be one way to internalize some of those externalities when we see, when I think we're going to see significant change in the composition of the labor market. That's one reason. A second reason is more of a geopolitical one. So I think a lot of people are talking about China and its work to develop AI. We heard at the conference today that the sort of spread is they're between six months and two and a half years behind the US, but gaining faster right now. And they're gaining faster for a couple of reasons. One is that they have much greater ability to add power. But second is they have much more access to data. But the third reason is I think that they have greater talent density. There's less foundational model teams in China and they can actually be brought together. And so the greater your talent density, typically the greater your level of innovation. Talent density for AI in the US is actually spread out over quite a large number of foundational model companies and actually it's being increasingly spread out as people spin out and niche down, and specifically OpenAI, right? And go and join or start other companies. So you're seeing some of that talent density actually getting spread out a little bit more. If you could bring all that talent density back together again, I think you could actually see a re-acceleration of innovation within... well, not to say that it's been slow, but a lot of the progress has been about more compute, more data. And there has certainly been some innovation in algorithms as well, but by bringing all those people together, I wonder if we couldn't actually re-accelerate our overall development as a country.
Totally. That's really interesting. I mean, it's also a different mindset and mentality too. I feel like in China, or in Eastern countries, so much about it is the collective, and here in the US it's so much more about the individual and the individual company and like who is winning out. And so I see it as quite distributed today. So I wonder if there can be something done to bring everyone together, and maybe that's the power of open source as well, is that everyone contributing to the same pool of ideas.
Potentially, yeah. I mean, if you just sort of, people are talking about the next generation model is going to cost about a billion dollars to train, and people are already talking about the generation after being 10 billion, and maybe the one after that being 100 billion. And whether that's, you know, historically it's been 18-month cycles, maybe it starts to extend out a little bit. But if you look at the overall US software market, it's $600 billion. And you could say, well, the market expands because now you're actually going to go after salaries, and so now you've got a much bigger TAM. But if you think about a hundred billion to train a model, right, and you say, okay, well let's say there are five foundational model companies, right? Today there's more like 10 or 15 in the West, but let's just say that some consolidation, there's only five. So that's $500 billion to some level replicating work, right? Because if you look at foundational models' performance, they're actually pretty tied with each other, with typically the most recent release being the leapfrogging to the top because they're all following a very similar path. So 5 times 100, 500 billion. If the software market's $600 billion a year, then at some point you say, well, can you do a next order of magnitude? It starts to get harder and harder to see how that all works. So fascinating. If you start to converge from the 15 to the five to two or maybe even one, which would be the nationalization model, right? Then you remove some of this duplication, you get again some of the talent density, aligned incentives, and so forth. So you know, that's a... like I hold this opinion weakly, you know, it's an interesting thought as your representative of Lightspeed or anything else, but I think when you zoom out from the perspective of what makes sense as an investor to what makes sense at the level of the country, you sometimes get different answers.
Fascinating. It's really, really cool. I appreciate you sharing that because I hadn't heard that before and it's interesting to think what could be. Cool. So we're going to move on to... we have two more questions left. This one is more of a fun question. So it's about travel recommendations. Okay, completely different, we'll do a little 180 here. But so for travel recommendations, like what is a place or city that you really enjoy visiting when you're off from work?
So I stepped back from making new investments at Lightspeed at the end of 2021, and we took a family gap year of international travel. We left the summer of '22, I came back summer '23, and we spent the first couple of months in a month in Tanzania and a month in Kenya. And that was truly fantastic. Yeah, would recommend, strong recommend. And yeah, I think I would say that I wouldn't want to live there, but in terms of a vacation destination to experience safari and really realize that at the end of the day we're all just bags of meat that could be easily... you don't think of yourself as prey at all, you just think of yourself as the apex of all animal kind in our day-to-day lives. But you go see that, like, okay, we are just bags of meat, could just be easily eaten, and it's only through our brains that we got to the level that we are.
Amazing. Did you end up climbing Kilimanjaro?
No, that sounds like a lot of work. I was there on vacation.
Okay, that's fair, that's fair. Well, I don't know, you go to the gym and all that stuff. I just thought maybe it's like a bucket list item or something. Cool. And so the last question that we have is on conference. So for conference 2024, what is something that you're looking forward to or the value you feel that you're getting out of this coming to this conference?
A lot of tech conferences that I go to, you're mostly there for the networking, right? You want to figure out, okay, who's this person, what can they do for me, how can they help me with whatever agendas that I have. What I've enjoyed about conference has been that some of the things that are obvious to people in other fields are new to me. Particularly, I think, around some of the framing around some of the political stuff. So again, this is obvious to people who are paying more attention, but I really would have thought that AI is higher on the agenda of, or should be higher on the agenda of whoever wins the next election. And what I heard was actually the most important thing that's going to get all the attention is taxes, because the Trump-era tax cuts are going to expire in 2025 and they have to be... it's urgent, maybe not as important, but certainly urgent, and that's going to get all the attention. So just some of these, again, for maybe for many people this was sort of obvious, but for me it was not. And so getting these perspectives that are a bit more orthogonal to my day-to-day, I find it very helpful.
Yeah, definitely. It's the worlds that we inhabit, and then people inhabit different worlds, have different algorithms, and when you bring them together, there's a lot of cross-learning that happens. I think that's oftentimes where you get the most benefit.
Totally. Yeah, is through these sort of insights from other industries.
Awesome. Is there anything else? Do you have any questions for me or any questions that you want us to ask?
Okay, awesome. Then that's it. Nice job. Okay, how do you feel?
Good. It's good. That was great, guys. I'm really excited about the discussion for this. It's so good. There's so much to say. And also I just want to say Jeremy is one of those people that he's so smart. Like I'm blown away by how smart he is and he's so good at expressing himself and explaining things. And there's a lot to talk about and we'll get straight to the takeaways right after this break.
Hey everyone, quick break to share something special. Sisters Matcha. We've launched limited batches of ceremonial grade, single estate, single cultivar matcha straight from the family farm Jean worked on in Japan. It's pure, authentic, and crafted with intention. Head to sistersmatcha.com to grab yours before it sells out. Make matcha your daily ritual for lasting energy and focus.
The first takeaway is when Jeremy talks about his advice for 29-year-olds, he has two pieces of advice. And the one that I really, really loved is work with people that you admire. This one just really hit home for me.
Yeah, I think he was even more specific about it, where he was literally like, if there's someone that you really respect that you work for, you should follow them where they go. He said blindly follow them.
Yeah, and he also used the word slipstream. That actually leads me to Jeremy's first piece of advice for 29-year-olds, where he said if you know what you want to do, go directly to that thing. Do not take any other steps. So I don't know if I fully agree with that, or I would say sure, that can be true, but there are definitely caveats.
Caveats. Do I have an accent? Caveat. I think there are certain circumstances, right? Like, it's nuanced.
It's definitely nuanced. It's nuanced. I get the sentiment behind his idea, but I do think there's, yeah, like sometimes you do need to optimize for other goals in the meantime. Like let's say you eventually want to be an entrepreneur, you want to be in a good financial situation when you're doing that, right? So maybe you can't just jump directly to being an entrepreneur. You need to first gain XYZ skills or build up your safety net over the course of two, three, or four years first before you can then go directly to what you want to do.
Yeah, I think it's a balance, right? It all depends on the person. It's so nuanced. Like, do you already have the skills needed up to a point so you can get to being an entrepreneur? If you are literally starting from zero, it's really hard to be like, 'Okay, I'm going to be an entrepreneur tomorrow' if you don't kind of have some of the stepping stones along the way to get there. I do hear his point because it depends on the person, but like if you want to be an entrepreneur and you're like, 'I have to do these three things first,' you could get stuck doing three things and get very complacent and then never do the thing that you wanted to do.
Yeah, I see why he's giving that advice because I think more often than not it's the opposite. People will delay and people come up with all these different excuses that are not really actual blockers but are just more reasons for them to delay not doing the thing that they truly want to do.
Exactly. I mean, Graham Weaver, a lecturer at Stanford GSB, who I've never heard of him, very famous, she only mentions him all the time. I mean, I was in his class. He's kind of like a cult following at Stanford for a good reason. I like when you mention your professors, it's fun.
You like it? Yeah, because it makes me feel like I know them and it makes me feel like I also went to Stanford. I talk about them every single episode, actually.
I know, unprompted. They're your friends. They're like literally amazing. But he's just like, the two most dangerous words anyone can ever say is 'not now.' Like that's a Graham Weaver-ism, where it's like, you know, 'not now, I'm going to wait until I get this promotion,' 'not now, I'm going to wait until after my kids are out of the house.' Like those words, if you continually say them to yourself, it just is not going to happen ever.
Yeah, and maybe like the first step is to clock how often you are saying it.
Yeah, and it comes in different forms, right? Yeah, yeah. Like another saying is like, you can't improve what you don't measure.
What you're not tracking. Yeah, that's something... that's a G-ism.
No, that's what PMs say. That's what product managers say.
Yeah, I'm going to take ownership of that. That's a Sheree-ism. If you can't measure, you can't improve it, so you have to figure out ways to measure it.
That's what product managers say.
I thought I invented that.
You definitely didn't. It's what product managers say when you're like running an experiment or like feature testing.
Yeah, that's why you need taxonomy.
