The Exchanges

Q Can I ask, when you think about constraints in this way, is it purely financial constraints or are there other types of constraints?

A No, no, no. Um, uh, it's, uh, you know, like if you, let's take an engineering, uh, uh, like project. Um, time is, is one of the best constraints. Time boxing. Like in Shopify, like every project goes to multiple phases. There's a prototype phase, there's, uh, like there's a proposal phase before that. If proposal is accepted, you go to prototype phase, build, and then releasing. Um, the, um, prototype phase Is ideally, uh, time constraints and people constraint. You, you, you can only do it with three people and you can, you've gotten maximum of It's six weeks. It should actually be less. Like it's, it's sort of meandered upwards. Um, um, and, uh, um, in fact it should probably be a week. And then next week you should start from scratch to try something else, um, and do this multiple times. That, that is like some of the best code I ever wrote. I wrote, um, with, uh, my, uh, my first engineer we hired, Cody, who was a CTO of company at the time. We did this in pair programming and, um, we set our self-constraints like, Every single day we have to throw away all source code other than the unit tests. Um, and we would ship the feature the day we could implement the entire feature in that day. And, um, you really, really, really, really get good code if you do that. So, um, in fact, I think one of the best applications of creativity, um, in the world of creation is actually creat…

AI assessment note: “No, no, no. ... time is, is one of the best constraints.”

Q Do you agree the best CEO is the best resource allocators? When you think about constraints, that's immediately what I think too. It's a commonly held saying, as we both know, do you agree with it?

A That's a great question. Hmm. I think if you take a really, um, I think the term resource is so poorly defined that it might be true by, uh, default, um, because, um, like you, you can look at everything in plan and think about it as a, as a, as a limited resource, um, like attention, book hours, like, um, uh, team size, like roadmap items, backlog. Here's what I think. I think, um, I think it's a heuristic by which. People can explain to themselves why certain people are great CEOs, but it's not the reason. It's not causal. It's correlated. Um, I think the causal, causality runs by, I think the best CEOs are people who have, um, a more correct model of the world than everyone else, um, are good communicators, uh, or have a lot of trust or are hugely enabled in some other form to, uh, Um, have other people follow them in such a way that, you know, that, uh, um, you build against that model and, um, have such a high betting average that the world that they are building against, um, is gonna come true by the time the projects are done. If that happens, more than coin flip, then, uh, I think you have a company that's out-competing status quo, out-competing industry, out-competing orthodoxy, um, Because that ambient understanding of the world is just not that good. Like, again, most things we think are true or wrong, um, and, or at least wrong in important ways. And so, um, as, if …

AI assessment note: “it's not the reason. It's not causal. It's correlated.”

Q Like, I, you know, I, I. Can I ask you a divisive question? Why, why do you go after the short sellers? I saw this tweet ago and I was like, go Toby.

A Woof. I don't, I don't go after short sellers. I think short, short sellers are an important part of the market. I, I, I go after one specific one who's like a, like short and distorted bad faith actor, who is like, like this, this end to a left fool, who is like just unbelievable, that guy. So I, this is just like, Hey, if you make a principle decision that a company is overvalued and, uh, like you, you short the stock, you make the, you, you play the game of Wall Street. Again, I don't really care about the game of Wall Street, uh, very much. Um, but like you're playing the game and I like the scheme exists and therefore you like, if that's the best way to play it, like play it. And if it has, um, converged the stock market value on the fair market value of the business, then actually you, you, you, you, you, you, you're providing a service. If you post bullshit of a company and like make like grandstanding videos of false accusation using basically just like the ability of you being able to legally make shit up while you have a short position against the company, um, and the company being bound by SEC communication rules, not being able to really respond to this. If that's kind of something you prey on, you're just a prick. So that's like, I'm, I'm, I'm, I'm, I'm, I guess I'm pro short selling in the abstract. I'm against pricks.

AI assessment note: “I don't go after short sellers... I go after one specific one”

Q Do you, do you think you have the luxury to do that today, being a public company, given the expectations that Wall Street puts on you?

A Oh yeah, of course. Um, yeah, it's just like, dude, like it's, there's nothing better in the world than being a trusted public company. I mean, the second best thing is being a trusted private company. Um, it's much worse to be an untrusted public company than an untrusted private company. So like it's a stack ranking is like, unfortunately, People become a trusted private company with their investors and have to actually take a step down to being an untrusted public company by going public. This is why people don't go public. They just can't deal with that, which is funny because like it's actually required to then get to the best spot in the end. Um, so which is literally the reason why we did it 10 years ago when we were tiny because it just felt like cool, like let's go make the careers of public market investors who bet on us and then we go and they make lots of money and then they become on our side and now they trust us.

AI assessment note: “Oh yeah, of course. Um, yeah, it's just like, dude”

Q Do you, do you think you have the luxury to do that today, being a public company, given the expectations that Wall Street puts on you?

A Oh yeah, of course. Um, yeah, it's just like, dude, like it's, there's nothing better in the world than being a trusted public company. I mean, the second best thing is being a trusted private company. Um, it's much worse to be an untrusted public company than an untrusted private company. So like it's a stack ranking is like, unfortunately, People become a trusted private company with their investors and have to actually take a step down to being an untrusted public company by going public. This is why people don't go public. They just can't deal with that, which is funny because like it's actually required to then get to the best spot in the end. Um, so which is literally the reason why we did it 10 years ago when we were tiny because it just felt like cool, like let's go make the careers of public market investors who bet on us and then we go and they make lots of money and then they become on our side and now they trust us.

AI assessment note: “Oh yeah, of course. Um, yeah, it's just like, dude”

Q like, I spoke about, you know, doing my work on you beforehand. I heard about the wonderful marriage that you have. Like, Toby, you have a fantastic partnership with your wife. What makes a great marriage, and how do you sustain that with being CEO of a public company? I mean, I can't even do it with a girlfriend in a podcast, Toby. How the fuck do you do it?

A I met Fiona, like, super early in life, um, and we ended up being, like, so, we're in many ways opposite people, and, uh, but, like, also have such a strong bond of, like, intellectual curiosity, and, like, we, I mean, we grew together from basically being kids. We, I think we, we, we started dating above 20 years old, which was more than 20 years ago. I mean, honestly, I think, um, very often being like the CEO of a company as a team production, um, and, and, and, um, Fiona does not get the credit she deserves. Like she encouraged me to even start a company in the beginning. So like, like, it's, you know, often you can play this game of like how many variables would have to change to lead to a completely different outcome. And it's like, it is always exactly one, which is Fiona for me. Like, so, um, she's like, she's never been on the payroll of a company, but she's always been like, Someone I talked about all the like, you know, challenges we've sometimes divided and conquered by reading, like there's like 15 books to read on to, because we need, like we as a unit need to understand this thing better. Um, okay, let's go halfsies. Actually, she, she reads faster than I do. So, um, she usually gets to a stack faster. But what sustains a great marriage? I mean, it has, it is to have common projects, but, uh, different lives a bit as well. Like, uh, there's like, there's just lik…

AI assessment note: “what sustains a great marriage? I mean, it has, it is to have common projects”

Q How do you know when to stop doing something? What I mean by that is you said every new piece of information is a delight. Say you engage with a strategic decision, and the information is revealed, is it strong enough to make you turn back? You don't have enough data. How do you have enough data or know when enough is enough and you've got to turn around?

A Yeah, I think there's a, Colin Powell did a, had a good sort of juristic for that. If you have, like, between 40 and 70% of that information, you should probably make a choice. Um, before that, you don't have enough, and afterwards, it's too little. I think that's largely right, but, like, I think it's missing a dimension on reversibility cost. Um, like, if, if, um, if a reversibility of a decision, um, and sort of the total body of work on it is, uh, is low, then you should change your mind very early, like, after you Like, basically, you know, the way this is gonna go, and, ah, if a whisper is high, you wanna be very, very careful, like, measure twice, cut once, um, so.

AI assessment note: “If you have, like, between 40 and 70% of that information, you should probably make a choice.”

Q one right, but he said the biggest problem that I have today is every day I walk in and people's heads are down because they're looking at their stock options going, ah, we've just lost another X percent. How do you think about maintaining morale in the face of volatile markets? You said earlier about being that great communicator. I was fascinated about that when I heard him say it.

A Yeah, it's interesting. Yeah, it's never been my experience. Uh, like, uh, Shopify has, um, appreciated a lot of value at some point and then lost a lot of value, and, um, yeah, this was sort of extreme during COVID, but, like, before that, it also, like, went up and down, left and right. I've always told the company, hey, um, the stock price is not the company. Like, it's, like, we, we all work on something called the fair market value of Shopify, which is, like, the sum total of products and skill sets of, like, of the people. Um, Most of the value of, like, of a SaaS company is, like, extraordinarily hard to value from a GAAP perspective. Like, Shopify's ability to hire the best engineers in the world. Like, what is that worth? Like, it's, I think it's tens of billions as a competitive advantage. Not an asset on any asset sheet. Never, never should. I don't want to value it. Like, I think that would be difficult. But it is one of the major assets we developed, right? Like it's, it's, it's, it's a, it's, it's a reputation. It's the fact that, you know, people are doing super impactful work for some, like for people they care about. Um, so we are building the fair market value for companies. It's unguessable. Um, um, the stock market is playing a daily guessing game. It's like, guess the fair market value. And, um, mostly, actually most people don't play that game. They play t…

AI assessment note: “I've always told the company, hey, um, the stock price is not the company.”

Q Can I ask, what captivated you so much? You mentioned Six when you got your first computer. What was it about it that made you so interested?

A Oh man, I, I just like, I, I know this sounds so wrong now, but like, even just, you know, first time sitting in front of it and just typing, um, like any letter and seeing it appear on the screen was just magic. Um, and, um, remember that was like, I mean, see, before that there was, oh, this was 80, Early eighties. So mid eighties. Um, and, um, you know, typewriters were also like in a small town. So like, you know, if you would have been in Silicon Valley, you would have encountered computers more. Um, but I wasn't. And, um, there was this sort of just magic. And then you're like, okay, well, how does a computer actually do that? And then, um, I'm curious about everything. I hate black boxes. I don't, I really, really dislike black boxes. It's actually like, um, Like a parroting principle in our house. Like we talk about this with my, with my kids. It's like, we do not like black boxes in the house. Everyone, everything is understandable and everything is way more interesting than it seems. And so, so I, I just wanted to understand these machines. And, um, again, like when you were like, maybe six is a bit early, but like when you're a 10 year old boy growing up in a, in a, in a, in a somewhat rigid, um, uh, schooling system like Germany's, um, like We, you know, had hours of Latin a day and these kind of things. Um, like, I mean, you have a lot of things. You might even hav…

AI assessment note: “typing, um, like any letter and seeing it appear on the screen was just magic”

Q like, I spoke about, you know, doing my work on you beforehand. I heard about the wonderful marriage that you have. Like, Toby, you have a fantastic partnership with your wife. What makes a great marriage, and how do you sustain that with being CEO of a public company? I mean, I can't even do it with a girlfriend in a podcast, Toby. How the fuck do you do it?

A I met Fiona, like, super early in life, um, and we ended up being, like, so, we're in many ways opposite people, and, uh, but, like, also have such a strong bond of, like, intellectual curiosity, and, like, we, I mean, we grew together from basically being kids. We, I think we, we, we started dating above 20 years old, which was more than 20 years ago. I mean, honestly, I think, um, very often being like the CEO of a company as a team production, um, and, and, and, um, Fiona does not get the credit she deserves. Like she encouraged me to even start a company in the beginning. So like, like, it's, you know, often you can play this game of like how many variables would have to change to lead to a completely different outcome. And it's like, it is always exactly one, which is Fiona for me. Like, so, um, she's like, she's never been on the payroll of a company, but she's always been like, Someone I talked about all the like, you know, challenges we've sometimes divided and conquered by reading, like there's like 15 books to read on to, because we need, like we as a unit need to understand this thing better. Um, okay, let's go halfsies. Actually, she, she reads faster than I do. So, um, she usually gets to a stack faster. But what sustains a great marriage? I mean, it has, it is to have common projects, but, uh, different lives a bit as well. Like, uh, there's like, there's just lik…

AI assessment note: “what sustains a great marriage? I mean, it has, it is to have common projects”

Q Can I ask you, what do you do when the value is ambiguous? And I think about, obviously, I create content. When you're debating on a new content strategy, the, the kind of terminal value is actually ambiguous. It may work. It may not work. How do you think about dedicating time when ambiguous terminal value is a potential outcome?

A I think, I think it just, you, you, well, time is your finite resource. Attention is your finite resource. So, so you kind of have to, you, you have to create like an evaluation function that you run over your, your ideas, right? Like, I think, um, you know, effort is a big, um, uh, factor. Uh, also reversibility is a big factor. Like if something is easy to reverse and not a lot of effort, um, to, to, to try, give it a go, because the fact that you think the value is ambiguous is a hole in your model of a planet Earth. Right. Um, so, so filling that hole means that you, like afterwards can avoid similar paths because you will encounter the same idea over and over again. Um, and you're developing a piece of content, you know, like, um, you, you, this could be, you, you could make a decision to commit yourself to a, uh, deep research, uh, subject about, you know, like having many, many meetings, many conversations, maybe reading some books, or it could be a, um, uh, You, you could say, well, I'm trying to, trying this new channel, and maybe I'm going to reuse some, uh, some, some, some content, or maybe it would work in this channel, or something I can do in an afternoon, and then you time box it. So this is usually when, when, when I'm, when my objective is to find out if something is of value, I try to figure out with everything I already got, and an afternoon of time, how can…

AI assessment note: “if something is easy to reverse and not a lot of effort... give it a go”

Q Are you proud of him as a Canadian for, like, bluntly standing up in an elegant and independent way?

A I can't say I'm proud of him for that. It's the greatest, it's one of the greatest speeches I ever heard. I don't think it's a full credible witness to the reality on the ground. I, I just, like, I, I also love a speech that is an ambition for, for, for a country. I personally disagree with, uh, uh, the stance to the United States. I, um, I think Canadians Uh, massively overfit to, um, niceness. Like it's kind of a best thing and a worst thing, uh, in a country is like to, to, to prioritize niceness. Um, because it's like, I mean, it's certainly pleasant, but the problem is it leads to, it leads to, uh, unkind lies, right? Like it's like, it's, it's means like you're lying by omission. You're lying, um, by saying things are fine when they aren't, right? Like it's, it's like, this is, this is a, you know, can you speak the truth is a much more important test. So Unfortunately, the way that intermeshes with the Trump administration is, is, is really brutal, because it creates this sort of positive feedback loop inside of Canada, which is, because he is so unbelievably not nice. Um, in, in any which way you look at, there's a Trump, Trump derangement syndrome in Canada that is just stunning. 60% of people, more than 60, I think the numbers have gone up since, and people think that the United States is the largest risk to, uh, Canada, like, much larger than, like, Russia or China w…

AI assessment note: “I can't say I'm proud of him for that.”

Q Toby, I'm gonna do a quick fire round with you. So I say a short statement, you give me your immediate thoughts. What have you changed your mind on most in the last 12 months?

A One thing I'm wrestling with is I had to change my mind. I, I thought, um, Straight out of school, uh, uh, engineers would have a really big advantage with no priors and like immediately sort of being AI native. And, uh, many of them are exceptional and are speed running their careers in ways. And there's like even younger people at universities, like we had an intern was 13 from Waterloo, which is like, still blows my mind. Um, and he was not even in his first semester and, um, his mom has to accompany him for classes, which is like, it's like, I love this. And, and By the way, he's very good. Like, it's just unbelievable. Um, so, um, you know, we are seeing like incredible, uh, like sort of career trajectory increases that I think are coming here. Um, but, um, I had to change my mind here because it, uh, because like initially I thought just having no price would be better because we had to reinvent everything. Turns out the way, uh, I think authentic programming actually works. You give an initial problem, you give a task, it might come for a ticket. Afterwards, an engineer is like constantly talks with, um, uh, the AI to like steer it in the right direction. Um, all engineers are massively underestimating how important the steering is. It's, um, it's, it's, it's incredibly important. It's actually That it's, it's just the same as programming. It's just very high level. Um, …

AI assessment note: “I thought... Straight out of school, engineers would have a really big advantage”

Q Does that excite you as someone who loves the craft?

A Oh, I love it. It's so wonderful because I, I just like, I love the craft and I can, the craft is not lost. Like also love mechanical watches, which make no sense, right? Like, but like they're like, they're selling more than ever, right? Because so like people always overestimate, again, this is what it, this is what growth is. Growth is not, um, subtraction of things or replacement. It is just adding. There's more in the world, right? Like, um, so We now have AI as well onto all the other wonderful products we can consume. So anyway, um, I can sit down and write code when I want to do this. Sometimes I do want to do this. Um, often it's part of a steering, right? Like I write the part of a code that concerns itself with the data structures and how the data is persisted on disk, which unfortunately that's sort of a German school of engineering, um, which is different from the computer science-y The third of, uh, North America. Um, I think North America is wrong. Um, I think you should think about how data is persisted on, on, on disk above everything else in your databases or so, and then you, you, you, you can plan the system with this. I, I, I still follow this and, um, I, uh, so work on this either by describing or by literally just like making the sort of data access layer. Everything else can be wipe cold on top because the, I, I know result would be excellent as long as …

AI assessment note: “Oh, I love it. It's so wonderful because I, I just like, I love the craft”

Q Toby, I'm gonna do a quick fire round with you. So I say a short statement, you give me your immediate thoughts. What have you changed your mind on most in the last 12 months?

A One thing I'm wrestling with is I had to change my mind. I, I thought, um, Straight out of school, uh, uh, engineers would have a really big advantage with no priors and like immediately sort of being AI native. And, uh, many of them are exceptional and are speed running their careers in ways. And there's like even younger people at universities, like we had an intern was 13 from Waterloo, which is like, still blows my mind. Um, and he was not even in his first semester and, um, his mom has to accompany him for classes, which is like, it's like, I love this. And, and By the way, he's very good. Like, it's just unbelievable. Um, so, um, you know, we are seeing like incredible, uh, like sort of career trajectory increases that I think are coming here. Um, but, um, I had to change my mind here because it, uh, because like initially I thought just having no price would be better because we had to reinvent everything. Turns out the way, uh, I think authentic programming actually works. You give an initial problem, you give a task, it might come for a ticket. Afterwards, an engineer is like constantly talks with, um, uh, the AI to like steer it in the right direction. Um, all engineers are massively underestimating how important the steering is. It's, um, it's, it's, it's incredibly important. It's actually That it's, it's just the same as programming. It's just very high level. Um, …

AI assessment note: “I thought... Straight out of school... engineers would have a really big advantage”

Q Now, I thought we'd start with something a little bit off the bat, to be honest, Toby. And so I thought we'd start with, think back to when you were a child. What did you want to be when you grew up?

A So I got lucky. I got really, really, really lucky. My parents got, I got a computer for me when I was, for the house. Um, I was six years old. Um, it was like a, like a basic computer, like a, sort of, If you think C-sixty-four, then you sort of get the right idea. It's a German product called an Amstrad. Um, and, um, I knew that there was nothing else like that would be more interesting than, like, I, I, I, um, I, I learned programming before I knew that was a thing or a term or that it wasn't was called some, uh, programming, uh, just from, uh, you know, following listings on magazines and so on. So I, uh, I very, very, very early made the decision. I must have been like 12 years old or so that, you know, there's just no chance I'm going to do anything other than program all day. No, I failed. In this particular frame of reference, um, uh, it, it was only a, uh, a period of my life where I did that all day long. Um, but I never stopped programming. I, I, I, so it's, it's been with me ever since.

AI assessment note: “there's just no chance I'm going to do anything other than program all day”

Q know, say, take COVID, and think about how shipping and delivery changes in the world of COVID, you have the ability now to call up Ryan at Flexport, or the CEO of UPS, or you name any of the big providers, and say, I'm learning how this changes, can you please explain to me the different moving parts? And you would get access to knowledge that that podcast wouldn't give.

A I think that's right. Although I'm pretty sure if you go, you would find within the timeline, uh, Ryan being on a podcast, uh, and, and, and talking about this very topic, right? Like, because it's, it's, it's, it's an obvious topic to ask him. And, um, I think there was a bunch of like write-ups about it. I, I, again, I don't want to belittle, you, you, you, you're, you're clearly right. I just, I, I actually don't use this as much. And I think this, there's a particular reason I grew up in an outsider city. Again, like I come from a town which Was started, I think, by Julius Caesar. Like, it was like 2000 years old when I grew up, uh, was being celebrated. And I think that was also the most interesting thing that ever happened there. So like, it's, it's just like, um, uh, like I, I was observing, uh, groups of people doing, um, high tech usually, or, or working in, in highly competitive places, figuring out new things from afar, my entire life. Um, I, I still am not in Silicon Valley. I live in Ottawa, Canada, which is not where I have coffee with, um, uh, Ryan from Flexport. Um, so, uh, so, so my skill set is probably, um, has sort of developed to try to, uh, go and find pieces of information to, the, the, the, the answers to the questions I have is often white papers or books, um, uh, often books that are Written very long ago. Um, and, and, and, and go to those, uh, source…

AI assessment note: “I actually don't use this as much. And I think this, there's a particular reason”

Q you said about a 10 to 20% productivity decline. I'm too interested. Because you've got, ah, I think, bluntly, there's a 10 to 10 productivity decline when you're remote. People take Friday afternoon off, Monday morning they're slower in, you name it. But then you also talked about, and have done before, like, reducing the number of meetings, cancelling meetings. Are people more or less productive in person or remote?

A Uh, I think people are more productive in person. Um, it's just what we are built for, like, it's like, like, in our entire brain, uh, is kind of, like, there's a lot of types of works that are, um, that can be disrupted by, like, like, that are harder to do when you're, like, surrounded by a lot of people in a busy place or interrupted constantly, so, like, for some types of works, uh, work, it's actually really, really beneficial to, uh, you know, work remotely. Um, so, so it's, it depends a lot on the job. There are certain things that remote work is not good at, which, um, you will always have to, you know, either create a great culture around or great process or like meet in person for, like, um, like this sort of free idea flow brainstorming session is not great over, uh, video. There's something, there's sort of like magic that manifests only sort of in the, in the middle of a table, like sort of floating over the conversation. About like the sort of shared picture and everyone's trying to figure out what could we do rather than, um, how to get there. How to get there was great remote. Um, what is possible is not that good. I think you can proxy it also to sort of wartime, peacetime, if you like that language. Um, like wartime, anything is harder to do remote. Um, but I mean, they invented airplanes, right? So like, it's like, it's not, you know, like The cool thing is g…

AI assessment note: “I think people are more productive in person.”

Q Do you not think you have to abdicate responsibility, but share accountability? And that's what sucks about leadership. You didn't do it, but when it goes right or wrong, in good and bad, it's often on you.

A No, you can't, like, it is impossible to, uh, um, it's impossible, like, you can share responsibility, but you never not end up being responsible for it. Like, I'm responsible for everything that ever goes wrong in the company at some point, maybe because I hired someone. Um, but, um, um, I also hold the responsibility there for, for, for, for, for, for, for all outcomes. This is why founder-led companies are also like a little bit of a different piece because There are tricks by which you can diffuse responsibility and, and, and, and reject. I mean, the entire, uh, consulting, world of consulting is basically one large way of outsourcing responsibility for their choices. Um, uh, uh, that's the main product they are selling. Um, and, um, uh, people might disagree with that too, but like, um, that would be a fun, different conversation for another day. Um, but I don't think you should do that. Like, I think you should stay responsible for everyone, and you should bring what you uniquely can bring to all, like, conversations where it's of value, and you should be very, very clear that you're looking for the best rather than being right, which is not, again, which is also not micromanagement. Like, it's, like, this is actually just We call it micro leadership. It's, it's like helping editing the final thing. A great company is looks, great products look to the users as if they com…

AI assessment note: “you can share responsibility, but you never not end up being responsible for it.”

Q you know, in the, you know, German school system and hating black boxes. I always think we're kind of all functions of our histories. I'm secretly an armchair psychologist who actually never got a psychology degree, um, or any degree at all actually, Toby. Uh, but my, my question is just, What do you think you're running from when you reflect on that all being a functions of our history?

A What I'm running from, um, well, I, um, boredom is something I like to avoid, um, ignorance. Um, I, I actually, um, it's more magnetic for me, honestly, like there's more magnetism in the direction of where I'm going than there is, uh, like, um, a reverse polarity in, in the place I'm coming from. I just, you know, there's certain, uh, schools of philosophy that do a much better job of, uh, articulating what I'm about to say, because this is sort of my own and home Homegrown life philosophy, but I just have a strong sense that the, the, the, the goal in life is, um, is, is, is really try to become the best version of yourself. I know it sounds like super fortune cookie, but like I, I describe it like I, uh, you know, it's, it's as if at the end of life you meet the person you could have become and the, the, the, the, the work of life is really try to minimize the diff between that person and the one that actually showed up. And, um, I find, uh, not understanding a thing feels, I, I, I, I, almost amoral to me. Like, it just feels like I, I, I, I, we've got these incredible brains, um, and, and, and it, we, um, these brains run a model of planet Earth, which we then constantly interrogate. And, um, I want that model to be right. Like, and, and therefore, Every single thing, every single time I, I get to dig into something, into learning something, I can update this model. And the…

AI assessment note: “boredom is something I like to avoid, um, ignorance.”

Q Final one for you. What's the best piece of advice you've ever been given? Like mine is, you're never wrong to do the right thing, and the right thing is often the hard thing.

A I think the best advice I've been given, but I haven't been giving it, but like it's, it's like sort of a rallying cry. I think that is just like, I, I kind of want to call us on that sentence just because it's memorable and I think people can't hear it enough is you can just do things. It's just like, At any point of time, the system is, exists to try to get to good outcomes. If you know what good outcomes are, you can step out of system and give it a go. The cost is lower than ever, and action causes information. And again, one of the best things that can possibly happen is you figure out something that, um, there's actually a lot of reason why the system discourages it. There's limits to this, of course. Um, there can't be, um, this needs to be a completely victimless, uh, um, endeavor. Like you can't have someone suffering for you trying a thing. Um, uh, that, that's, that's, that's very important. Um, um, no, people can't have negative consequences of, of, of these kind of things because then you're just like actually being an asshole, but because you're dividing people and prioritizing your own gain over Other people, and fundamentally you're engaging in a zero-sum thing, and in zero-sum environments, you should probably not run experiments, but you should do, if you are in a zero-sum environment, you should figure out how to get a positive sum, and that's much more often…

AI assessment note: “the best advice I've been given... is you can just do things.”

Q I mean this nicely. Why do you think this is one of the most interesting times? We have, like, deglobalization like never before, income inequality like never before, worst healthcare Systems like never before. Climate change like never before. I'm looking at the world quite scared, if I'm totally honest, Toby.

A Yeah, but I, I think this is, I think because, um, this is the, the overlay narrative of the media. Again, if you, if you, if you go at this from truth, um, I, I am like, I'm, I'm, I'm, I'm quite concerned about climate change, but I'm also a hundred percent, like, we know how to solve it. It's a tech, it's a technology problem. Um, wealth inequality is like, this is where most wealthy, uh, society has ever been. You, you can't compare the existence of, uh, any king to, uh, You have baseline experience of anyone in Western society. Uh, it, it just, it is like, you don't even need to go to like incredible, like, like, like potential for healthcare and all these kinds of things. Like no one until they recently had ever had a warm shower. Like it's, it's like, it's like, these are unbelievable luxuries we have. We just, we just have a built in hedonistic treadmill that makes us forget about it all. If, if, if, if you don't have to take my word for it, you, you go for something like, uh, Steven Pinker's book, Path to Enlightenment, or I think. And it just, it just, it just like drops like every single stat you've have seen quoted as going worse and shows you it actually gets, it's getting, getting better. Um, I'm extremely optimistic. This is like, it's, um, there is a effect in the world, which is, um, which kind of sucks, but it's gotta be temporary. Um, and it's, uh, um, I'm doi…

AI assessment note: “shows you it actually gets, it's getting, getting better. Um, I'm extremely optimistic.”

Q Do you believe in government intervention? I'm in Europe where we have a lot of government intervention.

A No, not really. Um, I, uh, well, I mean, okay, so let, let me rephrase this. Yes. Um, in, I am a fan of Friedrich List, who is a Prussian, long before James Hunt and Niki Lauda's rivalry, it was, um, Adam Smith, Lasse Faire, and Friedrich List and the Prussian school of, um, uh, uh, like, uh, economics. And so, um, I, what the Prussian school says differently to Lasse Faire is that, um, the role of government is very simple. It's to define games that have as externalities, um, Uh, societal, uh, thriving. And, uh, then completely get out of the game, uh, to then, um, uh, let the players in this game cause, through means of competition, um, extremely positive externalities to accrue to society. Which, you know, again, happens automatically, um, in, uh, uh, profit-seeking markets because, again, what a company does is it takes, um, the Potential, potential value that exists, which is infinite, absolutely infinite, coalesces into products that other people then vote for, and again, the voting mechanism causes then a larger supply chain and more of a supply of these products coming into society, and then, um, that creates a wealth, and then the wealth is, uh, part of the society, and any dollar that's created, um, from, in, in a place that brings new dollars into the system cycles seven times for, uh, uh, for society, um, Um, and, and, uh, in a, in a local way. So dollar flowing int…

AI assessment note: “the role of government is very simple. It's to define games”

Q Do you believe in government intervention? I'm in Europe where we have a lot of government intervention.

A No, not really. Um, I, uh, well, I mean, okay, so let, let me rephrase this. Yes. Um, in, I am a fan of Friedrich List, who is a Prussian, long before James Hunt and Niki Lauda's rivalry, it was, um, Adam Smith, Lasse Faire, and Friedrich List and the Prussian school of, um, uh, uh, like, uh, economics. And so, um, I, what the Prussian school says differently to Lasse Faire is that, um, the role of government is very simple. It's to define games that have as externalities, um, Uh, societal, uh, thriving. And, uh, then completely get out of the game, uh, to then, um, uh, let the players in this game cause, through means of competition, um, extremely positive externalities to accrue to society. Which, you know, again, happens automatically, um, in, uh, uh, profit-seeking markets because, again, what a company does is it takes, um, the Potential, potential value that exists, which is infinite, absolutely infinite, coalesces into products that other people then vote for, and again, the voting mechanism causes then a larger supply chain and more of a supply of these products coming into society, and then, um, that creates a wealth, and then the wealth is, uh, part of the society, and any dollar that's created, um, from, in, in a place that brings new dollars into the system cycles seven times for, uh, uh, for society, um, Um, and, and, uh, in a, in a local way. So dollar flowing int…

AI assessment note: “the role of government is very simple. It's to define games”

Q talking about is how every great American company is using frontier American models and then going, well, that's too expensive. So let's use open source Chinese models to get as close as we can to the benchmark set by US frontier models. And it's like these CCP funded models are basically powering a lot of the US startup ecosystem in many ways. Do you think the Chinese threat is underestimated?

A I mean, I think it's both underestimated and overestimated, but like, uh, you know, I think the much more bigger threat is that, uh, uh, in a lot of countries, I think this happened in, uh, Spain already, and I think even in the United States, social media aside, like, the moment the government has the ability to, uh, decide at which age, uh, children can use, uh, uh, technologies, they will, um, it, like, it goes like taxes, right? Like, it's just like, AI has so many naysayers, like, we are gonna tell kids they can't use AI. Kids will not stop using AI. What they will do is they will download Chinese models, and you will never, ever, ever get another high school essay about Tiananmen Square as a consequence of that. So, um, is that truly what we want to create here where, um, uh, children are, um, incentivized to, uh, get a monoculture of, like, because models are trained, uh, um, and, uh, are LHF'd into, um, a perspective. Chinese models are I mean, especially if you switch them to Chinese and then ask something else to translate, it's a very collectivist worldview, which makes sense because it's a collectivist country. I mean, this is the predominant fight. Like, all politics is just collectivism versus individualism, right? Like, it's, it's, it's everything else is, uh, people, um, smokescreening this stuff.

AI assessment note: “I think it's both underestimated and overestimated, but like, uh, you know”

Q Does that excite you as someone who loves the craft?

A Oh, I love it. It's so wonderful because I, I just like, I love the craft and I can, the craft is not lost. Like also love mechanical watches, which make no sense, right? Like, but like they're like, they're selling more than ever, right? Because so like people always overestimate, again, this is what it, this is what growth is. Growth is not, um, subtraction of things or replacement. It is just adding. There's more in the world, right? Like, um, so We now have AI as well onto all the other wonderful products we can consume. So anyway, um, I can sit down and write code when I want to do this. Sometimes I do want to do this. Um, often it's part of a steering, right? Like I write the part of a code that concerns itself with the data structures and how the data is persisted on disk, which unfortunately that's sort of a German school of engineering, um, which is different from the computer science-y The third of, uh, North America. Um, I think North America is wrong. Um, I think you should think about how data is persisted on, on, on disk above everything else in your databases or so, and then you, you, you, you can plan the system with this. I, I, I still follow this and, um, I, uh, so work on this either by describing or by literally just like making the sort of data access layer. Everything else can be wipe cold on top because the, I, I know result would be excellent as long as …

AI assessment note: “Oh, I love it. It's so wonderful because I, I just like, I love the craft”

Q know, say, take COVID, and think about how shipping and delivery changes in the world of COVID, you have the ability now to call up Ryan at Flexport, or the CEO of UPS, or you name any of the big providers, and say, I'm learning how this changes, can you please explain to me the different moving parts? And you would get access to knowledge that that podcast wouldn't give.

A I think that's right. Although I'm pretty sure if you go, you would find within the timeline, uh, Ryan being on a podcast, uh, and, and, and talking about this very topic, right? Like, because it's, it's, it's, it's an obvious topic to ask him. And, um, I think there was a bunch of like write-ups about it. I, I, again, I don't want to belittle, you, you, you, you're, you're clearly right. I just, I, I actually don't use this as much. And I think this, there's a particular reason I grew up in an outsider city. Again, like I come from a town which Was started, I think, by Julius Caesar. Like, it was like 2000 years old when I grew up, uh, was being celebrated. And I think that was also the most interesting thing that ever happened there. So like, it's, it's just like, um, uh, like I, I was observing, uh, groups of people doing, um, high tech usually, or, or working in, in highly competitive places, figuring out new things from afar, my entire life. Um, I, I still am not in Silicon Valley. I live in Ottawa, Canada, which is not where I have coffee with, um, uh, Ryan from Flexport. Um, so, uh, so, so my skill set is probably, um, has sort of developed to try to, uh, go and find pieces of information to, the, the, the, the answers to the questions I have is often white papers or books, um, uh, often books that are Written very long ago. Um, and, and, and, and go to those, uh, source…

AI assessment note: “I actually don't use this as much.”

Q Seven and a half, 8000. How many people will you have in five years time, do you think?

A I mean, my real hope is seven and a half to 8000, right? Like, uh, at, um, a hundred x productivity level, right? Like, um, I, I, I do think there's problems of scale. It would be sad to be in a company that is, like, opportunity limited, right? Actually, maybe I shouldn't say this. Most companies probably are opportunity limited. Um, they're probably not ambitious enough for, and I mean this in a product sense. Like, again, Remember, like, Shopify is, like, Shopify is a fairly broad mission. We want to make entrepreneurship more common, right? Like, and, um, so, um, inside of this ambition, we can do a lot. I just think we're gonna go into the golden age of entrepreneurship. It's by far the most AI-safe job. Um, it's by far the most AI-benefiting job. Like, think about that combination here. Um, so everyone who has a plan B of, like, some kind of ambition of, like, hey, I'm gonna make products is gonna be like, yeah, well, let's fucking go. All priors don't matter anymore because, like, doesn't, you didn't need to come from a family of entrepreneurs and having, like, been at dad's, uh, Office, or your grandma's, uh, printing shop, in my case, um, to understand that a business formation is something that real humans can do, and, and, and, and that gives employment and these kind of things. You can actually, I mean, first of all, you learn that anyway, and then all the various t…

AI assessment note: “my real hope is seven and a half to 8000”

Q What did you learn about building great teams from being in a band and learning music again?

A I, I learned a lot of lessons about leadership because like, um, I, I mean, this is, uh, most pronounced in jazz, uh, somewhat in blues as well, but like there's a basic contract. Like it's, Like artists, real artists will always talk about the tyranny of a blank canvas, right? Like this is actually not what leads to creativity. Creativity always comes from constraints. Constraining the space is a, is not just like, I've been in many meeting meetings where people say, I wish you could do this thing where it would be completely unconstrained by like, I don't know, budgetary time. And I'm like, yeah, but that's hallucination. That's not, you know, that's not, that's not, that's not craftsmanship. That's not, that's not engineering. I think the Older, um, crafts, um, have this totally figured out. Like in, in, in, in, in blues, you would, um, uh, you know, you would set a meter, like you would set a tempo, um, uh, you would presumably set a beat, and you would set a key. And then afterwards, people improvise over that. Um, different people might do a solo, different people, like, you, you're creating a new piece of music, uh, with every note. Like, it's, it's, it's, it's deeply creative. But the creativity comes out of the constraints of the set. And then, if you really know what you're doing, And this is where it gets really fun. You get to violate some of those constraints. Like…

AI assessment note: “I learned a lot of lessons about leadership because... creativity always comes from constraints.”

Q I spoke to people, you said surround yourself with other people on a similar journey. I spoke to Glenn before the show and he said on decisions, you're incredible, world-class. When it comes to dealing with sunk cost fallacy, what do you think he meant by this and how do you deal with sunk cost fallacy?

A A sunk cost fallacy is another thing. Um, I mean, it, it, it is a real, uh, uh, logical fallacy, like people over, um, um, the fact that you have been doing something for a while means you want to keep going because If you change your mind now, then you're basically, um, passing judgment on all your prior decisions. So, so, uh, again, if you're on a path of wanting to be right in the context of others, logical consistency is your primary optimization target. Like you want to have made a choice about something and, uh, you want to stick with this choice because your identity is wrapped around your beliefs, your shared beliefs, um, again, because you're prioritizing What other people think you give, you, you, you, you, um, one of the sort of, it's maybe too strong a term, but like, I can't help, but like think that if that's your operating models, you're actually outsourcing, uh, your self-worth to the people around you because you, you, you acquire positive affirmation by them to know that you're, uh, doing a good job, which I find is like really regrettable because that actually gives Too much power to people around you, first of all. Um, and second, they don't want that job. They don't, they have their own thing going, right? Like they, they, they're not, they're not interested in, um, uh, um, uh, moonlighting as your value system. Um, and so, um, I, uh, but if you're in this …

AI assessment note: “if you see people that seemingly have no problem with it, You will talk about them”