For years, people have told me Emilie Choi is one of the top COO’s in Silicon Valley. She’s spent nearly nine years at Coinbase, most of it as Brian Armstrong’s number two. But she almost didn’t get the job.
She made a lateral move to join the Corp Dev team in 2018. Then the market crashed a few months later. When the COO seat opened, the board told Brian to go hire a big name.
But he promoted Emilie instead.
We talk through how to be a COO, how to hire a COO (or promote one), why 1-on-1's should have criticism, how adding process actually helps you move faster, tactics for avoiding meetings, and why starting a committee at Coinbase can get you fired.
We also talk about some of the hardest moments on the job. Competing against FTX during its rise and fall, suing the SEC (and winning), and how they pioneered going direct and controlling their own narrative.
A great listen for anyone who just became a COO, or any founder about to hire one.
Support this Episode’s Sponsors
Monaco: The revenue engine for startups.
Flex: Sign-up for Flex Elite with code TURNER, get $1,000 here.
Clerk: Everything you need for authentication.
Numeral: The end-to-end platform for sales tax and compliance.
Inquire about sponsoring future episodes here.
👉 Stream on YouTube, Spotify, and Apple
Timestamps to jump in:
0:00 Process helps you move faster
5:14 Writing is thinking. Don't ship AI slop.
10:20 Agents are the next customer
13:50 Why starting a committee gets you fired
16:56 Tobi Lütke auto-deletes recurring meetings
19:48 How to roll back your recurring meetings
23:30 Collaboration is a useless company value
24:56 No politics at work
28:17 Shivers down her spine meeting Brian
31:43 What a great COO actually looks like
33:39 The weekly 1-on-1 where they criticize each other
36:14 Internal promos beat an outside COO
38:20 How to find candidates with grit
40:34 Talent density means managing people out
43:31 How to tell someone it isn't working
45:44 Coinbase: intelligence with humans around the edge
50:04 Regulators want AI 10x better than humans
52:46 TradFi embracing crypto made it boring
54:47 How Coinbase evolved beyond Bitcoin
58:32 Fred Wilson was the only Coinbase bull at YC Demo Day
1:00:36 Embracing capitalism when it's least popular
1:02:53 Gambling, prediction markets, and NFTs
1:06:29 Suing the SEC after FTX blew-up
1:09:30 Two choices: die quietly or go direct
1:13:06 Does going direct actually change outcomes?
1:17:28 Narrative whipsaws on Meta, Anthropic
1:18:52 How to operate throughout cycles
1:21:04 Off-sites are out, “Surges” are in
1:22:50 Why Coinbase stays remote-first
1:24:31 Jeff Weiner, the best operator she's seen
1:27:17 Sit at the fricking table
1:28:28 You don't have to talk in every meeting
Find Emilie on X / Twitter and LinkedIn.
Referenced
Coinbase is a mission focused company
Bain's RAPID decision framework
👉 Stream on YouTube, Spotify, and Apple
Transcript
Find transcripts of all prior episodes here.
Turner Novak:
Emilie, welcome to the show.
Emilie Choi:
Thanks for having me, Turner.
Turner Novak:
I'm excited for this because so many people have said, "Emilie's one of the best COO's in Silicon Valley, you've got to have her on the podcast," and now you're here. So I think a lot of this conversation is going to be just how to be a COO.
Emilie Choi:
I get this question so much from founders who are thinking about hiring a COO, and from newly minted COO's, how they adapt, how they are successful with their founder. So it's funny, I get a lot of the same questions all the time.
So I think this is a great forum to be able to talk about all these things.
Turner Novak:
Yeah. I think we're going to do sort of all your top Q&A type responses. We're going to try to fit a lot of them in here. I think one of the most interesting ones is you actually think you should have a lot of process.
So most startup founders would say, "Let's do the opposite of the big companies. Let's throw out all the process. Let's move fast." So what is an example of having good process?
Emilie Choi:
So when you first found a company, I can ask you in the hallway, "Turner, can you just make this decision on the spot?" And that feels glorious. And then, if the company is very successful, it grows very fast, and all of a sudden we start to hear, "Okay, it's harder to get in touch with you, it's harder to get these decisions made. I don't know the process," anything like that.
So my whole thing is, it's kind of counterintuitive, but if you actually implement those processes from the beginning, you're going to be able to run much faster as you grow bigger and bigger and more complex.
So one example is decision-making. We have a process called RAPID: recommend, agree, perform, input, decide. There's somebody assigned to each of those different roles. These are for complex decisions, right?
Turner Novak:
What's an example?
Emilie Choi:
So, should we enter this country that is very risky? We're going to make sure that the person who's recommending is the right person. We're going to make sure that there's no more than two agrees, and usually those will be people who are directly influenced, like legal or finance or something like that.
Perform, who's actually going to have to implement this? Input, anybody who might have very relevant input. And then there's always one decide.
So we're not a consensus-driven culture, we're not a collaboration-driven culture, in many ways. We're saying there's one person who has to take this data, synthesize it, and make the final decision, because we have a one DRI culture.
The reason this is really important is because I know how to get decisions made at the company. I might not like the decision that is made. But I know what the process is, and I'm not tortured in agony figuring out, how do I actually get this over the line either way, spend a year trying to get this? Yes. Tell me either way.
And I think it's incredibly valuable. So if everybody knows that from the early stage of the company, it can last. It's something that is a cultural component of the company that everybody knows how to scale on. So those are the types of things we care very deeply about, and we actually think they help us speed up.
Turner Novak:
So for each of the steps in RAPID, there's different people that are involved? So you could potentially say there's 10 people influencing this thing. I mean, that sounds like a nightmare, potentially.
Emilie Choi:
In a written document, it's not a nightmare. We actually try to err on the side of making that decision within 48 hours of receiving the full input. So again, it actually should speed things up.
It is a nightmare when people violate the principles of it, and are overly inclusive, and they're like, "Oh, let's get 20 inputs so that we feel like we've touched every person who could in any way be shaped by this." That's where mistakes are made.
So we actually try to hold a high bar and say, "Hey, you have to kind of adhere to these different things," or, "Why is everybody agreeing with this RAPID?" Actually, if everybody's agreeing with it, it means it wasn't a complex decision. Those types of things. We like to push on all of these different levers.
Turner Novak:
So how do you know what's the right amount of people to involve in something like this?
Emilie Choi:
Well, in general, there's one recommend, there's no more than two agrees, there's a couple perform and input, and there's one decide. So it's a 5 to 10 type of group size.
One other thing, in an age of AI, written artifacts are actually quite valuable, right? So we can look back and say, "Wait, Turner, I thought you said this in the meeting." "No, I said this in the meeting." Oh, actually, we're going to look in the document. We're going to look, and we're going to see.
And we're going to understand, in a Ray Dalio kind of style, what decisions you actually turned out to be pretty correct about, and which ones you didn't turn out to be correct about.
Turner Novak:
Is there anything where this has happened at Coinbase, and it's significantly improved or changed things, like one of these RAPID decisions?
Emilie Choi:
Well, I think every decision, every complex decision goes through this process.
Turner Novak:
Oh, really?
Emilie Choi:
So every person... And by the way, I'm sure there's many people in the company who would be like, "Ugh, another RAPID," and I totally get it. But there is value and virtue in codifying these things, in having the written artifacts, in understanding what made a decision a certain way, what arguments we made, and whether or not that turned out to be a correct decision or not.
Again, in an age of AI, here's another recent one we did.
So I'm big on this anti-AI-slop kind of thing, because one of our great engineers keeps talking about how writing is thinking. So use AI to help accelerate things, but don't use it to outsource your thinking.
So a team was writing a RAPID on entering a very risky country, but one that had a lot of upside. And we went into the meeting, and honestly, the doc was not up to par. It was very dense, clearly written by AI, lots of em dashes, and lots of focus on risk.
And Brian and I were both in the meeting, and we were like, "This is... Why would we be into this? It just doesn't seem like a good decision."
So we put our inputs in, we leave, and later on in the day I was having a meeting with one of the product people who had written the recommendation, and he was kind of in a funk. And I was like, "Tell me." And he was like, "I don't really think you grokked how much of an opportunity this is for us to enter this country."
And he was like, "That was a horrible meeting, in my opinion. There were a bunch of people just... it was torture." And I said, "Do you think the RAPID was framed in the way that we would have truly understood the opportunity?" And he was like, "Maybe not."
So I actually, later that day, rewrote the RAPID with AI, but with my thinking, and I framed it the way I thought he was actually trying to present it. We went back in the room the next day, and we approved it.
And so I actually do think it's incredibly important that you're not just trying to kind of shove something in and feed it all into the agent and hope it magically makes your argument. You need to think about the lens of decision-making, and what the considerations are. Judgment matters more than anything, I think, in this new era.
Turner Novak:
So what did you do that changed the way that it output this thing, in the way that the judgment actually kind of came through? Are you chopping things off? Are you saying, "Summarize this, shorten this"? Are you saying, "You're thinking too much about the risks here. We need to talk more about the upside"? How do you redo something like that and make it more effective?
Emilie Choi:
In this case, it was about popping the opportunity, which I think had gotten muddled and obfuscated in the former draft, where it was only about the risks.
And so obviously, in a company where we're highly regulated, we always have to frame the risks in the right way, and we have to make sure we're not only outlining the risks. We're outlining the financial risks and all of those different bands.
But we also have to make sure we're talking about the opportunity and what that could be in terms of total revenue, customers, etc.
Turner Novak:
I think one interesting thing about Coinbase is you guys are usually early to making sort of a controversial or contrarian move, doing something other people maybe aren't doing yet. Is there anything you're doing inside Coinbase right now, or anything you guys are thinking, that you think people might kind of switch to in the next year or two?
Emilie Choi:
In terms of our product set, we're making a deep focus on agentic finance, and we think that the future huge opportunities in terms of customer sets for payments and trading are going to be agents, not humans.
The timing, of course, is anyone's best guess, but we just know that machine-to-machine payments, trading, all these different things are going to happen through agentic finance. And so we want to be there at the forefront of it. We want to test it. We want to build it. We want to figure out what works for these customers and how we can be at the forefront of all that.
Turner Novak:
What does it look like? Because I hear people say, "Oh, AI trading, AI payments, AI finance." How does that actually turn on in practicality?
Emilie Choi:
Yeah. Well, today it's obviously quite small, and there's a lot of service payments between agents that are very small. And so this is...
Turner Novak:
So what does that mean?
Emilie Choi:
It's just agents paying each other for microservices, just to be able to ship things.
Over time, I think what we're starting to see, and what we'll play around with a lot, is how do we enable agentic trading on existing AI platforms? How do we offer agentic trading on our own platform? Is this going to manifest in a direct-to-customer interface? Is the customer going to want a headless kind of experience? There's all sorts of different things that we're currently exploring.
I think the fun thing is we're just in the beginning stages of it. And so this is, to me, the most exciting part of being in this space, where you're in a frontier space and you want to just...
We obviously were early in stablecoins. We had a partnership, and we continue to have a partnership, with Circle on the stablecoin front. Stablecoins are obviously a big form of payment with agents and in this agentic economy, USDC. And so us being able to power these payments through agents, I think, is going to be quite powerful.
Turner Novak:
I've always wondered this with stablecoins and agents. Why do you use stablecoins versus, I don't know, just not stablecoins? What's the benefit?
Emilie Choi:
The primary benefit is a stablecoin is a programmable form of money, and so you have to believe that in an agentic world, you're going to want to be able to program the interface with different task controls, etc., that are more frictionless.
Turner Novak:
And you can't do that with a US dollar or traditional kind of banking rails? How do you think about what the stablecoin interface or platform does differently?
Emilie Choi:
Well, we don't know, right? We're going to say, maybe it's going to say that you're going to put in a control around if it hits a certain dollar threshold, and you want it to pull back if this, this, or this. That is inherently embedded in a programmatic function for a dollar.
It's a machine-to-machine payment or trading, and you want to be able to embed certain functionality into it. We don't know what all those permissions might be today, but we know that they're going to be important.
You don't just want a $1 bill. You want something that you can say, "If this, then that."
Turner Novak:
So one thing I wanted to ask you about that I thought was kind of... I feel like it's probably one of your biggest posts ever on Twitter. You posted that it was a fireable offense to basically create a committee at Coinbase.
Kind of another one of those things, big companies, a lot of committees. Why is it a fireable offense to start one?
Emilie Choi:
So we take a lot of inspiration from Elon, who always talks about how it's easier to add than to delete. And we operate in a very regulated industry, where it is very easy to build up committees upon committees upon committees. And by the time you look back, you're like, "What? Wait, what was this? What was the intention of this? And how do we even close this down?"
So what we were finding was that a lot of people were creating committees just for the sake of creating committees, and in general, we're not big believers in consensus or any of that. So we're believers in DRI cultures, where somebody is accountable for driving the whole thing forward.
Turner Novak:
What is DRI?
Emilie Choi:
A directly responsible individual.
Turner Novak:
Hmm. Okay.
Emilie Choi:
So if somebody is directly responsible, we want them to be accountable for a decision. There are certain highly regulated parts of financial services where you need a committee for risk management. Totally understand. But in general, we want it to be absolutely required by law, as opposed to us forming it.
So anybody who does start a committee without getting Brian's and my explicit permission, they will be fired. It's kind of the ethos of our company, which is around agency, around being directly responsible, around not creating layers of bureaucracy or consensus building to get something done.
So that's why it's such a serious thing for us.
Turner Novak:
So there could be you, me, Brian, a couple other people in our committee. The thing we're doing doesn't go well, and we can all kind of hide behind, like, it was actually kind of Emilie and Brian, but you shouldn't fire Turner because, you know, whatever. So it enables you to kind of hide, and remove your sort of...
Emilie Choi:
Remove your accountability.
Turner Novak:
Responsibility.
Emilie Choi:
Yeah.
Turner Novak:
Okay. And so on the flip side, maybe integrating back with the RAPID document, are people getting ownership more through this kind of RAPID process as well?
Emilie Choi:
We push it more and more. So the decision-maker should be the DRI in many ways. They should be the person who is directly accountable for making that outcome happen.
I find that what happens in very complex processes is that we'll have a meeting or something, and there will be 20 people on a call, and we'll see three or four people from the same function on this call, often second-guessing one another, often, you know, whatever. And at the end of the day, we're never going to get anything done that way.
Tobi Lütke, who's on our board, he actually... By the way, now we're getting into meetings, and we're in no-meeting week at Coinbase, which I love. Tobi actually does these very, very extreme things where he will auto-delete recurring meetings randomly at the company, at Shopify, just to kind of show, like, if you... And this goes back to the add-delete thing. It's easier to add than to delete.
So if he just deletes it, he'll see, did they really, really need that recurring meeting? And if they really need to add it back later, that's totally fine. But if they didn't need it in the first place, it shows once the meeting is deleted. It's a very freeing thing.
Turner Novak:
Yeah. And there's this phenomenon in VC where people wanna do these recurring meetings, to catch up, to share deal flow. And I always was like, "This is kinda dumb." If you wanna share something, just text me.
Emilie Choi:
Yes. Email me. You know, I find that generally in life too, right? With these recurring meetings. First of all, I dread them. When I look at them, I'm like, "Oof, what are we gonna talk about?"
As opposed to, a great example is our board members. I think there's a school of thought where you should have recurring meetings with the board members to keep in touch with them.
In my experience, I have the best relationships with our board members when I can, in the moment, ask them something that is top of mind for me that they have unique expertise on.
Turner Novak:
Yeah, versus waiting two weeks because of the board meeting.
Emilie Choi:
Yes. It just doesn't... It's like, "Oh, we're just all gonna magically catch up before the board meeting, and everything that is relevant is gonna be spoken." No.
It's, I have this burning question about a compensation structure at our company, and Fred Wilson happens to have been through this with several of his companies. I'm gonna call him right now.
And I think we have a much deeper relationship because of that, because he can uniquely add value in that moment, and I'm not wasting his time, and he's not wasting my time. So yeah, I'm a very big fan of that.
Turner Novak:
And even with these, like, oh, I should probably talk to my board members once a month, just make sure I stay in touch with them. I've made stuff with AI where you have people you wanna keep in touch with, and it just tells you you haven't talked to them for two months, and you just do it, right? You don't even need a meeting.
Emilie Choi:
Yeah.
Turner Novak:
Right? I mean, I used to actually do that with my calendar. I'd throw something on just to remind me to do some things. You can automate a lot of that stuff now.
Emilie Choi:
It's fantastic. And you can achieve this via text with the board, too. It's like, "Have you seen this company? Have you..." You know, whatever it is.
Again, it raises my cortisol to have this bunch of recurring meetings with different people. It just doesn't solve anything.
Turner Novak:
So if I'm sitting there, and I have a ton of recurring meetings, what have you seen work best on rolling those back?
If I'm the founder or if I'm the CEO of a company, and I'm like, "We gotta get rid of these recurring meetings," what would you recommend somebody do if they wanna start to go down that road of removing some of those layers?
Emilie Choi:
We prompt in the meeting. We're like, "This is a very expensive use of time. Did we need this time together, or could we have solved this async via Slack or something else?"
So we try to... And it's not a blame thing. It's like, "By the way, did we actually need this time right now?" We have no meeting weeks.
Turner Novak:
Yeah, I was gonna ask you about that.
Emilie Choi:
Yeah. So again, this is no meeting week for Coinbase, which I love, which is like we wanna focus on shipping, and we wanna reduce the meeting tax and toll on the people who are shipping.
And so if we force this... It's kind of a lighter-weight version of the Tobi extreme, just delete the recurring meetings thing. So we just try to find different ways to do this.
And the other thing my chief of staff did recently, which I thought was great, was he did do an analysis via AI of the point about how many people are in meetings. Do we have duplicative people within the same function on each of these meetings? And then being able to analyze, oh, this function tends to overrepresent at meetings.
We need to be able to obviously take good notes in the meeting and then disseminate them after, and not have to have everybody have the tax of being in the actual meeting, and not feeling FOMO for having missed the meeting.
Turner Novak:
Yeah. That's also another great AI use case, is just taking notes in meetings.
Emilie Choi:
Exactly. Exactly. And then everybody has them, and AI can disseminate the action items and so on.
Turner Novak:
One thing I know you've pushed back on is this concept of collaboration as kind of a company value. You definitely see that a lot of, like, we work cross-functionally across teams. We all collaborate and work on projects together, etc. So why do you push back on that?
Emilie Choi:
We were coming up with an updated version of our company values at one point, and we take deep inspiration from Ben Horowitz's shocking values, shocking culture values, that kind of thing.
Turner Novak:
What is that? I've never heard of that before.
Emilie Choi:
It's this idea that if I tell you that our value is collaboration as a company, you're gonna be like, "You could be anybody. You could be Salesforce. You could be Google. You could be anybody." It doesn't tell me anything about you.
Whereas if I tell you one of our key values is mission first, no politics at the office, you probably would have a sense that Coinbase is the company.
Turner Novak:
Yeah, I might guess Coinbase from that.
Emilie Choi:
Yeah, yeah. So in other words, be known for something. And don't be known for some generic value that doesn't really tell you much about anything.
Turner Novak:
So this would be like customer first or something, or like teamwork or collaboration?
Emilie Choi:
For us, talent density is another big one. And I think we were, like, the earliest there about just relentless focus on talent density. And that means managing people out aggressively.
So I think it's good for a company to be known for something. And again, in 2020 when we did the mission first, no politics at the office, it was a very shocking thing.
Now I feel like it's almost more mundane because everybody realized it was the right thing to do, but back then we got eviscerated for it.
Turner Novak:
Yeah. What was that like when that happened? Because I mean, I remember it, and I remember there's probably two subsets of people. There are people that were like, "This is evil. This is terrible. I can't believe you'd do this."
And there was the other subset that was like, "This is amazing. This is so smart." So what was that split like, and how did you make the decision, and what was just kinda like the feedback from people?
Emilie Choi:
It was a very heated time in the world, and Brian, I could tell, just had this really big bee in his bonnet about, he felt very strongly that we should go out with this no politics at the office thing.
And to be totally honest, I was super scared about it because it just felt like not the right time. It felt too heated.
Turner Novak:
Why did it feel too heated?
Emilie Choi:
Because everybody and their mother was an activist about everything. It was just one of those times in the market.
And he wrote a blog post, and he's like, "I'm ready to go out with this." And there's times when I know you can't fight city hall. I actually very much agree with Brian on this. I don't know if it's exactly the right time, but I think we're doing it whether or not I want to.
And so one of the things I talked to him about was we didn't actually have a contract with our employees, when they were joining Coinbase, to let them know this is how we felt. So it felt like we should have an offer to those employees who don't agree with this.
We're all adults. And if you don't agree with this, and we're telling you this now, we're making it very clear right now, you can part ways with us, and we'll give you severance, and we'll all kind of part as friends because now it's clear. So anybody who joins now must know and must agree with these. And if they don't, they shouldn't join the company.
So that was actually a very... It was like a nice operational thing to think about, because I feel like the people who left during that time, about 5% of our employees, felt very good about the terms on which they left.
We shipped the blog, and it's funny, if you look at it now, I think the thing that stands out to me is we didn't lead with the mission first. It was a little embedded. And so I think Brian, looking back, was like, "We should've just said it the first thing," right?
But quickly people picked up on it, and there was, as you said, a swath of people who were like, "This is the way," and they're gonna look like Nostradamus in a couple years.
And then there's people who got very upset. And looking back, I encourage you to look back on some of these tweets, some of which have been deleted, and some of the rhetoric around it. There were CEO's who got very upset.
There were people who had a lot of big followings who were very, very upset with us. And it was interesting to me just how emotional they were about it.
And again, I think if you look now back at all of this, Brian was quite prescient about it, and it doesn't feel like a big thing anymore.
To me, part of the reason that I've been at Coinbase for nearly nine years is because I love working with somebody who is so brave and who really wants to just say the thing. And I think it's embedded into our culture, and it's one of the principal reasons I love being at Coinbase.
Turner Novak:
And you said, I think you've said this, that you got shivers was, I think, the way you described when you first met Brian. So what was that meeting like?
Emilie Choi:
Yeah. So I'm the luckiest person in the world 'cause I've just had the best jobs ever, and I've gotten to work with the best people ever. I fully acknowledge this.
But I had been at LinkedIn for nine years, and I was kind of just ready for a change. I had been there when it was 400 people. I was there when it was 11,000 people. I was there when it was private, public, owned by Microsoft, the whole gamut. And I was just kind of open to thinking about a new area.
And I met Brian through a contact at a16z, and I walked into the room, and I remember I just had shivers down my spine. I was like, "This..." We were talking about crypto, the future of crypto. It felt electric. It felt like there was something really powerful. And even though I wasn't obviously deep in crypto at that time, I knew I had to be a part of it.
I was like, "I just wanna jump in." And I took a lateral career move to run corp dev for Coinbase.
I've always been into working with founders. That's why I love M&A. And I come over, and Brian had assembled, like, a new leadership team of, like, 15 people.
Turner Novak:
And was this 2017?
Emilie Choi:
2018.
Turner Novak:
2018, okay.
Emilie Choi:
So I signed in 2017 when the market was like this, and then it promptly went down as soon as we joined. We were like, "Oh."
Turner Novak:
Oh, man.
Emilie Choi:
But that actually... So this newly formed leadership team of 15 people had never worked together, had never worked in crypto before, and the market just started tanking basically halfway into 2018. So tension erupted.
Turner Novak:
Yeah. And you guys were all outsiders, so I'm sure there's people like, what are they doing over there?
Emilie Choi:
We're all outsiders. What the hell did I just get myself into? Yes. And I know people from LinkedIn were like, "What in God's name did Emilie just do? This is... This poor girl."
And what I did was, my whole hack in life is I surround myself by very smart people that I'm inspired by. And, you know, the new people are miserable, frustrated, fighting, quitting, being fired. It was like chaos. It's like a recipe of what not to do.
And I'm looking at some of the couple smartest people in the company. Heads down, happy as clams. Couldn't be happier that the market was down, 'cause they were like, "All the mercenaries get out, and we can just focus on building."
And so I was like, "I'm just gonna suck it up and take my cues from them, 'cause they seem to be very smart." So put my head down, kept working my butt off. As different problems emerged, I would volunteer to help fix them.
And at the end of that time, our COO moved on. And the board was like, "Brian, you can go hire some big name out there."
Turner Novak:
Yeah. And you probably should bring a big name in to, like, right the ship. Cement everything.
Emilie Choi:
Yeah, yeah. And he was like, "Nope, I'm elevating Emilie because I know that she can do the job because she did the job in one of the worst times."
And so that was... We've had this huge evolution of our relationship over time, but it's one of the things that really cemented our relationship, because we've been through so many ups and downs together, and he had that trust in me after, you know, all of those reps together.
Turner Novak:
So what do you think a good COO looks like? Maybe you can just describe yourself. Maybe there's the qualities that, you know, maybe you wish you had, you're working on.
But when you think of the best COO, what does that look like? And maybe a company like Coinbase, a public company. Maybe it's a Series B stage startup who's starting to scale. What does the best COO look like?
Emilie Choi:
So I always work backwards from what does the CEO founder look like? Because in general, I'm a big believer in founder CEO's. I think that they stand the test of time. There's a method to their madness.
And I always say to people, I'm like, "Think about what you are uniquely good at as the founder CEO and what you don't like to do, and that other person should complement you. It should be a yin and yang kind of thing."
Turner Novak:
Really? But that's probably hard to admit, and there's somebody that's, like, better than you at certain things. I feel like some founders maybe have trouble with that.
Emilie Choi:
But I think some founders, by the time they get to a certain point where they need a COO, they've realized this, right?
So they're saying, "I hate people management," or, "I love product and engineering and design, but the other core functions, I really want them to be in great solid shape, but I want somebody else to manage them and lead them."
And so that's what it roughly is for Brian and I. He runs product, engineering and design, and I run the rest of the different functions at the company that we call core, whether that's security, whether that's customer service, whether that's marketing, comms, BD, ventures, corp dev, etc.
And it works really well. It doesn't mean we don't second-guess each other either. I've definitely had moments where I question a product strategy, and I will get up in his biz about it, and he will get up in my biz about how we're doing a compensation policy or something, or a legal policy or something like that.
And so we kind of have checks and balances on each other. One of the things I feel very lucky about, about the beginning of our relationship, is that in general, I don't believe in leadership coaches. I think all of them suck. But we did have somebody who helped us forge some new paths in terms of understanding one another, understanding what frustrates us or irritates us, and having a way to talk about that with one another.
In the early days, we were getting very methodical about the way we did these little systems with this coach. But today, every week when we have our weekly one-on-one, we have a section for juicy feedback, and it means we kind of have to have something.
It could be something like Brian saying, "I actually didn't like the way you talked about this thing in the all-hands," or something like that. Whatever it is, it's okay. We can talk about it. We trust each other so much that we get better by telling each other these things.
Turner Novak:
So it's like constructive criticism. You try to find something that they can do better, and you just bring it up to them and say, "Hey, just by the way..." Or...
Emilie Choi:
Yes, or, "This bothered me this week." Like he said to me before, "I think you're being a little too hard on this person in product about this thing."
And I just want you to realize he's taking it hard. Whether you want that or not, he's taking it. So it's great insights for the way that we operate together.
Turner Novak:
It sounds like really the best fit is you figure out, so starting with the founder CEO, what are they good at?
Where are the holes? You want to make sure the COO has a pretty good plug-in. It's almost like a puzzle piece, where the differences between them kind of fit well together.
Emilie Choi:
Yeah.
Turner Novak:
And then they're probably able to talk to each other.
Emilie Choi:
Yeah.
Turner Novak:
Openly, and be honest with each other.
Emilie Choi:
Yes. And it's magic when it works. But just to be clear, it's very hard to find that perfect match.
One of the things I tell a lot of founders is I would go try to find some great VP ops or something like that, and go see how it all works first. Does this person match your style? Does it feel like you are additive together, that you trust one another, that you can get to that right place?
If so, then carve out that path for them and elevate them. But if not, it's less high stakes. Humans are very nuanced, and so finding that exact right person at the right time is always going to be very challenging.
Turner Novak:
So it seems like probably you're not going to find a COO making a COO move.
What's an appropriate way to go, say, the VP of ops? How would you figure out that they're ready, or that they would be good to take a step up?
Emilie Choi:
In general, I love internal promos when they work. We've had a couple recent ones. Crypto's had a tough last year. We've had some attrition.
We practice what we preach, so our CLO left and we elevated the person under him. Our CHRO retired and we elevated the person under him as well. What you have when you have those people in-house who are ready, who have apprenticed under those people, is that they have a hunger, and they also have a knowledge of the company that outsiders might not have.
So in general, all things being equal, I'll prefer an internal promo over an external hire, if we have that person who can be groomed to that.
If not, we're going to run the search. And in general, I love a biz ops profile for these things. My general thing is these people who have some type of foundation of consulting or something, where they understand how to do good presentations, how to do good data analytics, how to run these things down, and are generalist athletes rather than superior at just one thing.
We've had these fascinating, wonderful people. One of our heads of biz ops back in the day came in when we had a bunch of gaps in HR and recruiting. We had him go fix those. He was like, "You know what's weird? I'm just loving this recruiting thing. I'm realizing the talent foundation is what sets us up for success of the company. Can I go run recruiting?"
So you have these people who just love problem solving and want to go after it. That's, to me, the magic sweet spot of what you're looking for in these types of people who can eventually be COO.
Turner Novak:
How do you suss out if someone's a good problem solver? If you're interviewing me and I'm like, "Oh yeah, I love problem solving..."
Emilie Choi:
Yeah.
Turner Novak:
What would you ask me, or try to get me to talk about, to prove I actually do and that maybe I'm good at it?
Emilie Choi:
So for anybody director level plus, we have them prepare a presentation, and we ask them very pointed questions. One of the first questions we will ask you is, "Tell us about a time personally or professionally that you've demonstrated grit and resilience." It's a very important thing to be successful at Coinbase because we have so many ups and downs.
Turner Novak:
Yeah.
Emilie Choi:
So very instantly you can tell if there is an authentic story or not. There's one woman who talked about divorcing her husband, her dog dying, everything falling apart, and her having to get back up. It was a very vulnerable kind of thing. And then we had somebody who was talking about killing a spider, and it was just super meaningless.
So you get the whole gamut, and you can see who actually thought and knew what drove them versus not. But similarly, we have very specific questions based on their function, related to how good they are at problem solving. Something for a compliance person would be, how would you think about automating all of our compliance queries given the complexity of what we're dealing with, and things like that.
And we want to see how they think, and how they communicate. I mean, this sounds like a generic value, but I actually think our most important value is clear communication. We're pithy. We need people to be pithy about the way they communicate.
And it's not only verbal. It's about, back to my first point, written. How can you communicate very clearly in a written form? Because in a world of AI, your written artifacts are going to be incredibly important for us to be able to derive big insights from.
Turner Novak:
Because you might word something in a way that three years later the AI is still referencing some wrongly worded thing.
Emilie Choi:
Yes, yes.
Turner Novak:
I run into that all the time. When I'm using Claude to do something, it'll bring up something that's on the internet...
Emilie Choi:
Yes.
Turner Novak:
...about me, and I'm like, why do you keep doing that?
Emilie Choi:
No, totally.
Turner Novak:
I've told you to stop doing this.
Emilie Choi:
No, I'm like, why are you over-indexing on this one concept or something like that? It drives me nuts sometimes. Ugh.
Turner Novak:
One thing that you just mentioned, and I think you've mentioned a couple times, is the importance of talent density at Coinbase. So what does that look like?
Emilie Choi:
So I came from LinkedIn, and it was hammered into me that talent is the operational foundation for everything if you want to be successful as a company.
And so we focus very much on recruiting the right talent. I look at people that I've worked with in previous careers who were phenomenal at the companies that I used to work for, who would be absolutely horrible at Coinbase.
Turner Novak:
Really?
Emilie Choi:
One example, as I mentioned, is clear communication. It's shocking to me how many very successful people are not as pithy as we require them to be at Coinbase.
And so I'm very much looking for a certain type of person that I know is going to be able to run fast and hard at Coinbase, who can deal with the ups and downs of crypto winters and summers, who can see the frontier technologies being the most exciting things, who aren't just into incrementalism.
Those are the things that we're looking for when we're recruiting great people. And then when they're at Coinbase, how do we set them up for success? Because we're in a regulated industry, we do get a lot of people from TradFi.
So we just hired this wonderful person like a year ago, a senior from TradFi. And one thing I remarked was that I hadn't really seen this person show up in Slacks, and that's actually where I love to get insights from about people, about the way they think.
And the manager said to me that he'd never used Slack at his old company. So he did it in meetings or emails. It was this really interesting thing. And I was like, well, we should make sure people understand we're watching them on Slack.
We want to see what their insights are. We want them to lead with insights on Slack. We want them to be able to do work through Slack, and that's a key to them being successful at our company. So just helping onboard and get people to the place where they're going to actually be successful at the company.
And then also just performance managing, thinking about people who might not be a great fit at the company for whatever reason, and being able to have those conversations early. So just like, "This isn't right for you. This isn't right for us. Let's have a very adult way of parting ways," and just saying, "This isn't going to be the right thing for any of us."
So we think about talent throughout all of its different ways. And then obviously on the compensation front, we want to make sure we're rewarding people who are crushing it. So all of these different dimensions matter deeply in a talent density kind of function.
Turner Novak:
So one thing I wanted to ask you about, maybe you hit on this a little bit, about managing out people.
So let's say you and me are sitting in a meeting. I've had a couple chances. I'm probably not a good fit at Coinbase anymore. How would you do that conversation with me in a respectful way, so that I start making the transition out?
Emilie Choi:
Yeah. I mean, I've obviously had this conversation many times, and it's generally like, "This isn't working."
It's probably not a fun conversation. "This isn't working. Let's figure out a way to make this as painless as possible. Let's make sure there's a good transition, that you've got everything in a place where people can take it up, and that you feel like you're leaving in a good place and we're all leaving on good terms."
There's that Moneyball scene where it's like...
Turner Novak:
He just comes in, it's very much a business decision, and then like 10 seconds later...
Emilie Choi:
Don't, yeah. Don't talk about, oh, where are your kids going to be able to go to school, or something like that. To actually be kind and humane is just to cut to the chase and say the thing.
Back in the day I worked at Yahoo, and we did the acquisition of Flickr, and Stewart Butterfield had this great thing about horses smelling fear. So whether you're in the performance review or whether you have to terminate somebody or something like that, me beating around the bush, just starting with three nice things about you and all this kind of stuff, and me being nervous in the way that I'm communicating with you, instead of just being kind and compassionate and just cutting to the chase and saying the thing. I think that's a much better way to operate.
So it's what I've always lived by.
Turner Novak:
Yeah. I feel like if I'm sitting in that meeting with you, I probably know. And when you're saying all these things to make it maybe seem nicer, at the same time I'm probably, it's not going to do anything.
It's not going to change anything. It just took longer.
Emilie Choi:
Yes.
Turner Novak:
And I'm probably also confused because I'm like, "Well, you said I was so great, and I'm doing all these great things," but then I'm getting laid off. Or...
Emilie Choi:
I'm making you suffer more in that moment.
Turner Novak:
Yeah. That'd be tough.
So one thing I want to talk about. I think Brian described kind of the new Coinbase. I think it was May 5th, was the day. It was over the summer. He was just like, "Coinbase is now an intelligence layer with humans around the edges."
So how does that change how, and how have you guys been thinking about Coinbase as a company now with AI at the center?
Emilie Choi:
So in general, what I love to think about in an AI native company is how we can do even more, how we can ship better products, intelligent products for our customers. And what AI can also do for us in terms of enhancing automation, all these different things. Ultimately, what then matters in terms of the people I hire is how much great judgment they have about the way that we're implementing these things.
So we talked before about a decision where we were thinking about entering this risky country. The way I frame it versus the way you frame it with AI, we might be able to write the document much more nimbly and quickly with AI, but it still matters how you framed it, and then how I perceive that and how I take that.
So do I take it as, we're entering this country and it's full of risks, and just be aware that the cost could spiral out of control if we don't watch this very closely? Or is it, this is a huge freaking opportunity of a country. We've seen this time and time again with other analogous traditional models in that country.
By the way, it comes with these types of risks. We've bounded them at this type of range at this moment, and we're going to come back to you with more data, but given the information we have to date, we'd like to proceed with future findings or whatever. So all of these things matter in this AI native world in terms of your perception, your judgment, the way you frame things.
And I think it's going to become more and more important in terms of the type of talent that we seek out, especially at the management and executive layers. Obviously, we could go and wax poetic about all the other things that are kind of obvious. You want smaller, nimbler teams. You want ICs. You want people who actually get their hands dirty and who actually know how to use this stuff, as opposed to people who like to delegate and not get their hands dirty and who are just moving things around on paper.
We really feel like things like distribution, trust, which is a core part of what we built the company on, data and data insights are going to be some of the most important things that we can have as proprietary moats during this new era.
And so we're heavily invested in jumping in, learning as much as we can as possible.
Turner Novak:
Yeah, I saw... on your earlier point, I saw this chart actually earlier today. It was the value of being autistic and having ADHD going up over time. It was like centuries, and each time it went up and up and up.
Emilie Choi:
Yeah.
Turner Novak:
Whereas, I don't know, 1500s, it didn't really help you at all.
Emilie Choi:
Yeah.
Turner Novak:
And then as new technology came, and the importance of math. And then the 2000s was basically peak. Like, if you have ADHD or if you're autistic, you can lock in and accomplish these really hard things.
Emilie Choi:
Yeah.
Turner Novak:
And then it was projected to nosedive, where basically it's not that valuable anymore because the...
Emilie Choi:
The machine...
Turner Novak:
The machine can do it. And so if you're... I was actually thinking about this while I was going up the elevator when I first got here. I was like, "Man, I feel like the importance of sales, relationships,
in-person things, AI's never gonna do that." Even the humanoids. I'm not gonna be sitting with a humanoid, building my relationship with a humanoid robot. Maybe I will, but probably not. But the value of those kinds of things actually goes up way more.
And then your ability to... I'm always thinking, how do I use AI to do something that I kind of don't wanna do, and just automate it, make it faster? And the locking in and getting the thing done,
Emilie Choi:
Yeah...
Turner Novak:
I don't even have to do it anymore. It's just spending 30 seconds spinning up a new Claude prompt to go do the thing.
Emilie Choi:
Yeah.
Turner Novak:
And that becomes way more valuable. It's how you interact with it.
Emilie Choi:
Yes. For us, again, compliance automation is one of the beasts, because we have so many different customers, and they have to go through KYC, and we have to do all of this different stuff. And it's not simple if-then queries or anything like that.
So AI is automating a huge part of it. And then at the end of that, there will still be some type of human judgment. And we have to make sure also that the... One of the... I think one of the most important things that might be overlooked right now with the AI stuff is, it doesn't just... it's not just about the automation.
It has to be 10x better. For example, for our regulators, they either don't care if we're using AI, or they don't like it, right?
They wanna know that if we are gonna automate a bunch of this stuff, that we're doing 10x better than we would if it were purely manual, purely human.
And so those are the types of bars that we're gonna have with our outward facing products as well, when you think about agent to agent banking and other transactions. There's a bunch of press that came out last week about JP Morgan not wanting agents to be in the mix of this stuff, because this stuff is so sensitive.
And so when we think about all these things, these interactions have to be fantastic, and they have to be 10x plus better than what exists today in a manual way.
Turner Novak:
Mm. Yeah, when you think about what a lot of, I mean, not just the financial industry, a lot of industries,
Emilie Choi:
Yeah.
Turner Novak:
it's just spreadsheets that are getting emailed back and forth,
Emilie Choi:
Yep...
Turner Novak:
maybe getting uploaded and exported from systems and, again, emailed back and forth.
Emilie Choi:
Yeah.
Turner Novak:
And then when you go back even maybe pre-computers, we actually couldn't even do some of that stuff. It was literally, we would hand write
Emilie Choi:
Yes.
Turner Novak:
some things down and mail it.
Emilie Choi:
Yeah.
Turner Novak:
And then, again, you would take this handwritten thing, and maybe you'd lose it
Emilie Choi:
Uh-huh...
Turner Novak:
or something. I had Toly, the founder of Solana, on, and apparently, I didn't know this, up until 9/11, they still every... I forget exactly what it was, but every single stock trade purchase had to still be done physically.
And they would literally mail it on planes to the central exchange. And then because of 9/11, they realized, "Ah, we can't do this anymore." 'Cause you might actually lose the file. And so some of these things take so long, but then the things that we are able to automate and use technology for just go so much faster and so much more efficient.
But it always blows my mind how there's so many things that are still done without using the new technology that we have.
Emilie Choi:
Well, and that's why we're so bullish on crypto as infrastructure, because we do think it digitizes everything. We have this thing about updating the system, and there are so many manual, horrible parts of what you do financially.
And when you think about things where it's 24/7, global, no middleman, all that kind of stuff, who gets to judge your financial transactions, there's so much appealing about updating that system. And we think that there... this is why I think TradFi is really just embracing crypto infrastructure.
You know, the OG crypto people in some ways I think are devastated how boring crypto has become, because you have all the TradFi banks and others talking about using the technology now. But it's obviously much needed.
Turner Novak:
Yeah. One thing I will never fully embrace on the crypto is the 24/7.
I feel like it is good to sleep. I feel like the market should be closed. I don't want it to be 4:00 AM, 3:00 AM on a Saturday night and have to worry. But I don't think I'll ever change my mind on this. I think the market should be closed, and there should be times where you don't have to think about it.
Emilie Choi:
But do you believe that for your agents? You like your agents working all night long, right?
Turner Novak:
Yeah, fair.
Emilie Choi:
Yeah. And I think people who trade, and people who are global, at some point they're going to want these markets to be accessible at any given time. So you might not be that prototype.
Turner Novak:
Oh, yeah, I'm definitely not that.
Emilie Choi:
But it's very appealing.
Turner Novak:
Well, I just only have friends, pretty much everyone I know who is really hardcore into crypto and does a lot of trading, you can tell they don't sleep.
Emilie Choi:
Yeah.
Turner Novak:
I'm like, "That's just not the life for
Emilie Choi:
me." Yeah, you're like, "This isn't good. This isn't good."
Turner Novak:
I like, you know, private market investments.
Emilie Choi:
Yeah, yeah.
Turner Novak:
10 years with no liquidity. It's like...
Emilie Choi:
You're a long-term guy.
Turner Novak:
Yeah, whatever happens, happens. Like, I slept for 10 years and
Emilie Choi:
Yeah...
Turner Novak:
and then we got all our money back.
Emilie Choi:
And we're okay. Yeah.
Turner Novak:
Yeah. Yeah.
Preferably, maybe it's like two years, right? The sooner the better, right? I think the super interesting thing about Coinbase is most people probably think it's just the Bitcoin wallet. That's historically where it started. I remember when it first came out, I was like, "Oh, that's the trusted place to keep your Bitcoin, to keep it safe."
I don't think that's all Coinbase is now. And I don't know how many people realize that. So what is Coinbase today?
Emilie Choi:
Yeah. So Coinbase is a place to trade any asset, any time, 24/7. And the novel thing that happened over the past year was we recognized that people wanna be able to trade truly anything at any given time, not just crypto.
So we launched the Everything Exchange, where we have traditional equities, where we have perps, derivatives, etc., options. We also obviously have a very trusted custodian. And I think your point about trust is actually quite important, because the thing that Brian and Fred Ehrsam, our co-founders, leaned into kind of counterintuitively in the early days was a foundation of trust and leaning into regulatory.
Because the roots of crypto are very libertarian, anti-authority, which is part of what we love. But they recognized that if we were gonna touch people's money, that trust was gonna be a foundational thing, and I think that has carried us very, very far. So we are the most trusted place and the easiest to use for all of these different assets.
I think another thing perhaps the market hasn't fully caught up with is we were very early in stable coins. We did this partnership with Circle in 2018, and USDC has become the number one regulated stable coin in the world. It's kind of amazing. And by the way, in the first couple years it wasn't showing much signs of life.
Turner Novak:
Why'd it take so long?
Emilie Choi:
It... That's... I mean, I think this is true of technology, right? Timing is everything. But this is also why we plant a lot of seeds as a company, because you just never know. And we thought that there was going to be a need in the market for a stable type of US-backed stable coin. But we didn't know what a catalyst would be or what would take off. It turns out that people did want a programmable dollar, it just took a long time.
The other thing, when I first joined Coinbase, one of our board members said to us, "You know, you have a very volatile business. And what you're gonna wanna think about from an operating perspective is you're gonna wanna be able to cover your opex with subscription and services revenue. And then the volatility of what you get from crypto cycles, and the crypto transactions, is gonna be just gravy on top.
And then you're gonna have a business that you can actually truly manage on a multi-year cycle." And so we very consciously invested in things like Coinbase One, our subscription service for trading stable coins. And so subscription and services is now like 50% of our revenue, which is a really, really...
I mean, that was not an easy thing to build over time. So I think, yeah, I don't always think the market is fully caught up with where we are as a company. And that's fine. We're happy to be the underdog, but we think we have a very foundational business. When we think about the next couple years, we're gonna continue to invest very much in the Everything Exchange and being that place where people can trade any asset they want.
We're obviously super focused on stable coins. A lot of stable coin activity is in trading. We obviously think that there's gonna be a huge swath of activity happening in payments, so we're investing in that part. And then, as I mentioned, agentic finance is something that Brian's super passionate about, and so we wanna make sure we have a foundation there as well.
Turner Novak:
It's kind of interesting. Every single financial services company, it feels like you start in one specific thing and then eventually you just
Emilie Choi:
Yeah...
Turner Novak:
add more layers to it to where you do touch a lot of different things. And we've seen it not just in fintech, but even Wells Fargo, I think,
started initially it was like... I think they got their big break when the earthquake hit San Francisco in, or like, 1907, and that's when they first came to the US, right? You always have this really weird niche initial thing, like buying Bitcoin, keeping it secure, and then you expand and add all these different services and business lines over top of it.
Emilie Choi:
Yeah. It's funny, Fred Wilson, who's one of our most beloved board members, who's been on the board for like 13 plus years, he went to the YC Demo Day and saw Brian present. And then Paul Graham came out and said, "Which of the companies did you like?" And Fred said, "Oh, I like that Bitcoin wallet company, Coinbase."
And Paul said, "You're the only one who liked it." And Fred's like, "Great.
Turner Novak:
Yeah.
Emilie Choi:
I can get it for a better valuation." And this is actually what I love about Fred. He's not a herd mentality kind of guy, which is what I think many VC's are. He was like, "This is a great opportunity for me."
And he saw the whole future of, you know, there's gonna be other assets, there's gonna be other trading, there's gonna be this trust thing with the custodian that's gonna be important. So he saw the future with it and was able to invest early and be a phenomenal sounding board for us.
Turner Novak:
Well, yeah, it's interesting in financial services, trust is so important.
There are banks where they do things that you may not want them to be doing. Or even, I mean, you think about bank robberies, right? Or like, again, 100 years ago, the trains would get robbed, and the stagecoach stuff, people transporting the money and you lose it.
You really need to make sure that you trust the institutions you're working with.
Emilie Choi:
And it guides so many of the decisions we make about what we have to do with our products, even when it's more costly or painful.
Turner Novak:
I think one thing that we've kind of been seeing, generally, there's been sort of this pushback on capitalism, this pushback on free markets.
How do you feel about capitalism and free markets?
Emilie Choi:
We're unabashed capitalists. We are unabashedly in favor of free markets. We're unabashedly in favor of individualism, meritocracy, agency, the idea that you can create your own destiny. And again, I think this is very much inherent in the crypto ethos of you're in charge of your financial future, and this is why economic freedom is the guiding thing for us on our mission.
So we will double down on that when it is most unpopular, and we're very loud about that.
Turner Novak:
Why do you think it's become unpopular? 'Cause it's kind of like the foundation of the US, I feel like. I mean, really, a lot of the modern world is around capitalism.
But there's been pushback. Why do you think people are pushing back so much on it?
Emilie Choi:
Well, I thought it was interesting, that leaked memo from back in the day with Peter Thiel sending to Zuck, about...
Turner Novak:
I remember that, yeah.
Emilie Choi:
don't just dismiss it, actually try to seek to understand it.
Turner Novak:
Yeah.
Emilie Choi:
Which was essentially that younger people feel like they're being pushed out of any economic future. They feel like they don't have access to great jobs. They don't have great access to a good economy. They don't agree with things.
They're living with their parents when historically they would've been able to kind of buy a house and things like that, so.
Turner Novak:
Yeah.
Emilie Choi:
That's why I do think this whole notion of what crypto is supposed to be about, which is economic empowerment and being in control of your financial destiny, is very relatable in this moment. I think, of course, there can be dangers with respect to degenerate gambling type behavior. But I think there's this idea of Bitcoin and crypto being freedom in many ways, right?
Because there is no middleman. There is no... and there is no ability to inflate the supply, that kind of thing.
Turner Novak:
Yeah, it's always interesting when I talk to people that are... They almost treat gambling, like... I mean, whether it's crypto trading, NFTs, stocks, or just sports betting, it's like that's their economic opportunity.
If you just get that parlay to hit on NFL Sunday, then you don't have to work again.
It's just always wild to me that there's so many people...
Emilie Choi:
Who think of it that way.
Turner Novak:
That think that way. Yeah. And then on the other side though, there's people who, you know, you go to an NFL game, and you spend like 50 bucks on a bet of, like, this guy breaks 100 yards, and it just makes the game a little bit more fun. So it can make it more entertaining. And you're way more invested. And that emotion of the guy getting the touchdown, and you make $1,000, that actually can make it more fun too.
So it's this weird kind of balance between the...
Emilie Choi:
You can see both sides of it. It's funny. So we talk all the time about who's gonna win the midterms. Who's gonna win the Super Bowl? Who's gonna score a touchdown? And it's fun. It's really fun, and it's a huge part of what we do talk about on a daily basis, right? With friends and family.
Turner Novak:
Yeah.
Emilie Choi:
Or coworkers, fantasy football, and all these different things. So in so many ways, I feel like the sports market and the information markets are so... They're like a new form of news.
Many people don't even look at polling anymore because it's got so many error bars around it, and the information markets we're seeing with prediction markets are just more accurate, more fun.
So I think of it in its best form as being this new form of media consumption that I absolutely love. And I think that if we have good boundaries around the way that we operate these things, we can make sure that we're protecting users as well.
Turner Novak:
Yeah, how do you do that? Because a lot of people would say the era of NFTs is probably the most extreme example of this in crypto, where we were buying NFTs of rocks for a million dollars. There's no reason you should probably be doing that.
I don't know. I actually didn't really do much of it, so I didn't follow what you could do on Coinbase with all that stuff. But how do you guys incorporate that into the way that you're operating in the product, and making sure that you're doing the right thing?
Emilie Choi:
We think a lot about the risk controls around it. So if we notice that a customer has been doing something with certain frequency, or hitting limits that feel untenable, we will politely kind of prompt them and make sure that they know what they're doing.
So I think there's a lot of frankly easy guardrails to be put into place. But I honestly don't know if I see the difference between buying an NFT of something for a million dollars and buying a piece of art that you happen to uniquely think is valuable that you're gonna buy for a million dollars. That's your individual choice.
Turner Novak:
Yeah. So it's like if you're choosing to do that 'cause you think it's worth that, you should be able to do it.
Emilie Choi:
Yeah, and you love it.
Turner Novak:
Yeah. I mean, it's free market and capitalism. I can make fun of you for the NFT or the picture, the painting.
Emilie Choi:
Yeah.
Turner Novak:
It's your choice to do it. If you like it, you can have it.
Emilie Choi:
Absolutely.
Turner Novak:
One thing that you guys did, you actually... And you talked about the importance of working with regulators and working with all that stuff. So you guys actually sued the regulators. I think a lot of people said, "Don't do that. That's not a good idea."
So I don't actually know the full details around this, but what was that that happened?
Emilie Choi:
Yeah. This is another Emilie very scared moment. Brian was like, "We're doing this," kind of thing.
Turner Novak:
Oh, man.
Emilie Choi:
So post FTX blowing up, every politician and regulator was on our tail, and the irony is we had always been super compliant and operating compliantly onshore.
FTX and these other players were operating offshore because there were no rules that were clear in the United States. So we were doing the hard thing onshore, and then we got eviscerated by all these politicians and regulators. They were just trying to go after us.
Turner Novak:
Could you see what was happening? Like, did you know what was going on with them, and you were like, "We're not gonna do what they're doing. We're just gonna wait for it to end poorly, and hopefully we can continue it"?
Emilie Choi:
Kind of. Kind of. So our board members were like, "What in God's name are you doing so wrong? These guys can operate with 40 people and just mint money, and you guys have this big clunky infrastructure of people. What is wrong with you?"
We did. We were torturing ourselves. First of all, we knew that there was something funky going on with their proprietary trading and their exchange. 'Cause it was clear that there was no way they had a complete wall between the two.
So we were always kind of wondering why that didn't seem to matter to anybody. But-
Turner Novak:
I feel like when a chart's going up, people will just kinda...
Emilie Choi:
Yeah, people just overlook things. And there was aura around them. I mean, props to them. They figured that out, right?
Turner Novak:
Yeah, you and Brian need to get on beanbags and do some investor calls, and play some video games while you're on the call.
Emilie Choi:
It was more like we were scratching our heads. We're trying to do this bottoms-up analysis of how we can operate more like them or something like that, and we knew we couldn't with the number of customers we had, and the amount of KYC and compliance we had to deal with. So something didn't add up.
So then fast-forward to them imploding. Everything fell apart for us. The market plummeted. Everybody was after us. We were persona non grata in DC. They hated us, Elizabeth Warren and all these others. Gary Gensler, the chair of the SEC, was relentless.
And long story short, he launched a Wells notice against us.
Turner Novak:
Yeah, what is a Wells notice? I've heard of it, but what is it actually?
Emilie Choi:
Basically, that we're operating illegally because of a variety of our practices. For example, he claimed that some of the assets that we had were securities, that we were claiming to be commodities, and things like that.
So we knew it was coming, and we had two choices. One was put our heads down and die, and the other choice was to just go out and make our case to the world.
There was a very inconvenient fact for Gary Gensler, which was we had gone public under the very auspices of the SEC with substantially the exact same business in 2021 that he was then suing us for in 2023.
And so one of our advisors, who's incredibly smart, was basically like, "Dumb that down and explain that to the world. Make your case directly to the world."
Turner Novak:
So we just worked with them to go public, to do exactly what we're doing, and now they're trying to sue us for what we just worked with them on?
Emilie Choi:
Yes. Yes. And I can't tell you the number of people, including bank CEO's, who were like, "You never sue your regulator, especially the SEC. You just... Do not do that. It is not smart. They're gonna go after you even harder."
Turner Novak:
Yeah, 'cause it just muddies things up.
Emilie Choi:
Yeah, and they can just claim that... They have the regulator. The judge is gonna believe the regulator, right?
Turner Novak:
Yeah.
Emilie Choi:
So we did do it, and what we did was we filmed Brian and our chief legal officer, Paul Grewal, and we just in a very plainspoken way said, "This is what's happening and this is our case to the world."
Turner Novak:
You posted it online?
Emilie Choi:
We posted it online. Honestly, it was pretty revolutionary at the time. This whole going direct thing, we're not the ones who originated it, but I think we've had very novel uses of it that were frankly very brave. This goes back to me loving working with Brian 'cause he's super brave.
And we put it out there, and it really helped us win hearts and minds, I think. Because it was just like, we're not trying to skirt the law. We've been operating with a huge legal team, and a huge compliance team, and a huge policy team for all this time because we're trying to actually have this great American company and be able to ship products from onshore while our competitors are doing whatever they want offshore.
And so, anyway, I think it was a huge turning point. I'm actually quite, I don't know, I'm kind of spiritual about these things, where I think these moments that come up in your company evolution, I think Fred Wilson calls them crucible moments. You basically can show if you're gonna make it or you're gonna take it.
And we're always like, "We're gonna fricking just go out there." Another example for me is we had this horrible thing happen a couple years ago where some very bad actors tried to bribe our BPO customer service agents to give customer information.
Turner Novak:
Oh, wow. That could've been really bad.
Emilie Choi:
And it was a huge amount of money, right? So they're trying to extort us. And once a year, I have a very sick feeling in my stomach about something, because we have to deal with these crazy things. And Brian's like, "We're gonna go direct. We're gonna go say, 'Not only are we not gonna pay you, we're gonna pay a bounty to whomever finds you.'"
And just turning the tables, and the way that we've done this, it feels so freeing. Back to this whole idea of agency and owning the moment and being individually accountable, I think it's such a huge part of our ethos. We can survive anything. We've got resilience like nobody's ever seen.
Turner Novak:
So I'm curious how that actually works in practice. 'Cause if we're doing some kind of closed-door conversation where there's just a couple people that are making some sort of decision, but there's all this public stuff going on around it based on Brian posted something, does that really influence what happens in some of those decisions? How does that play out?
Emilie Choi:
Well, meaning, like, when people are like, "You shouldn't do this"?
Turner Novak:
No, so let's say when you guys laid the case about here's what's happening with the SEC. You kinda framed the positioning. But you're still meeting with the SEC.
Emilie Choi:
We're still meeting with the SEC. Yeah, they hate us.
Turner Novak:
So how do the conversations change? Like, they probably hated you more because you did that.
Emilie Choi:
Yeah, they hated us more. They hated us more, and they iced us more.
Turner Novak:
But then how did the public perception around it, did that impact the decision ultimately, you think?
Emilie Choi:
Well, I do think it influenced things, because I think that it was such a telling thing to be like, wait, help me understand, so you guys are being sued, but you have the same business you went public for? That kinda makes sense to me.
And I think all of these things, as you know, with media, are layers. There's a public perception. There's us actually suing them in court and having our day in front of the judge, where they got to make a decision. And playing the long game on policy.
So one of the other things that we did, we had, again, we had a great advisor, and one of the pieces of advice was we need to have a grassroots organization, 'cause crypto people are really fervent, and they have a lot of mojo.
Turner Novak:
I was gonna say, it's like energy around this.
Emilie Choi:
They have a lot of energy. You don't see pro-data center people coming up in AI in town halls, right? You don't often see that, sadly. You see the anti-data center people showing up. But for crypto, we have a very strong base, and this guy was like, "You're not tapping into this at all. You have to make sure people understand."
We now have two million plus advocates who go to these town halls, and they actually speak, and they're very pro. And another thing, we have 52 million people in the US who own crypto. I think politicians just didn't know that. So making these things known was very important.
There's the different layers, and they all kind of come together to make this all happen. I think that's how you build a very effective policy and kind of media arm at a company when you're in these type of frontier spaces where people don't fully understand the whole thing.
Turner Novak:
So it's not just this whole going direct. It's not just like you create a Twitter account or an X account, and you just post and everything kind of magically happens. You have to use it almost as a lever in influencing other opinions that people have. It's certainly multiple layers throughout.
The media talks about you differently, your supporters, your customers, your fans, your suppliers, your customers, etc. It influences how all the different pieces interact.
Emilie Choi:
Yeah. And each time it's a different game. There was a time when The New York Times, post us going apolitical, they were relentless. They were trying to dig up any bad story about us. They had these horrible things. I still remember, again, the feeling in the pit of my stomach, where they're trying to dig up dirt on us.
And we did this novel thing, and it seems so mundane now, but it was so big at the time. I remember people were like, "Holy shit." This reporter had these things, and he was giving us 24 hours to respond, and we knew that many of them were bad facts. We knew that the data was bad. So we preempted him, and we posted the blog with the rebuttals before he could publish the story.
I mean, he was livid. He was quite-
Turner Novak:
Pissed.
Emilie Choi:
He was super livid. That breaks the rules of The New York Times, how you do this.
And our comms person at the time was like, "Oh, my gosh, I'm, like, ruining every relationship I have." So it's been this really kind of crazy, fun journey, because it's funny how in the moment you're like, "Oh, my God, this is the biggest thing in the history of the world," and then later you're like, "Oh, that was great. We did it. We said our truth."
And by the way, now everybody's following us. You see banks suing the SEC now. You see everybody going direct. I'm not saying we pioneered things, but I think we found very novel ways to do them, and for Coinbase, and I think the companies that do that will benefit greatly from that.
Turner Novak:
Yeah. And it's interesting how the current narrative, or the current cycle, just changes so quickly.
Emilie Choi:
Yeah.
Turner Novak:
Like, pick any topic. I feel like you could talk about how it's kind of whipsawed around. The Overton window always changes. I'm trying to think of something good right now on the spot, but like...
Emilie Choi:
Well, I, I... To me, Meta right now is fascinating, because if you look at the stories from like two months ago, it was like, "Oh my gosh, Meta bombed AI, Meta bombed the Metaverse."
Turner Novak:
Yeah.
Emilie Choi:
Like, they didn't get it, they did a terrible acquisition, blah, blah, blah, blah, blah. And then it tells you that you just have to hustle your ass off, because things can change. If you just keep focusing on shipping the best products and the best experiences and doing the right thing for your company, it'll all come around.
But I find this too with just the AI stuff. One day you're gonna see, it's like Anthropic has taken it all. It's over. And then it's like OpenAI, oh my God, their model's better, and oh my gosh, their numbers... It's just...
Turner Novak:
No, everyone's against...
Emilie Choi:
Anthropic.
Turner Novak:
...now.
Emilie Choi:
Yes, yes.
And you're like...
Turner Novak:
Whoa, these guys had...
Emilie Choi:
Yes. Like...
Turner Novak:
...it was like the mandate of heaven six months ago.
Emilie Choi:
Yes, yes. Like...
Turner Novak:
...it just always changes.
Emilie Choi:
I mean, it's fascinating to look back at this stuff and just realize that it's all a perception at the time. And if you keep grinding, you can just break through it all.
Turner Novak:
Well, so you've probably gone through some of those moments at Coinbase. So what... I mean, what is it like internally, and how do you... You know, you've made it through a couple times.
Emilie Choi:
Yeah.
Turner Novak:
I actually don't pay that much attention to crypto, so I don't know how, like, as closely as you do. So maybe you've gone through two, three, four of those.
How do you do that as a company?
Emilie Choi:
Yeah. I didn't know that I was accepting that mission when I came into Coinbase. I was so naive coming into Coinbase. I mean, LinkedIn was such an incredible, slow and steady wins the race, up and to the right kind of thing.
And that's what I tell anybody who's coming into the company. I'm like, "It's gonna be super volatile. And if you push through, you're gonna do very well. But if it feels like this is just not right for you, which is totally fine, don't come here. It's just not gonna be right for you."
But ultimately, it's about being mission first, not mercenary. And I think that you want people who are genuinely fired up about updating the financial system, and who wanna be here, and who are okay with there's gonna be some twists and turns. We're not gonna get everything right. Because with frontier technology, you're not gonna get everything right.
Like, I think Meta has proven this, you know, in leaps and bounds in terms of saying it's not the worst thing to mess up and to lose, as long as you win the long game, right? And so we've had our own little things. One example is in 2025. We have such a strong policy and legal arm.
I think we weren't ready for the moment of, in the new administration, that the market was just gonna rip, and we should have all the products and services out there just to ship them to our customers. And some of our competitors took advantage of that moment.
Over the past year and a half, crypto's been a little bit in the doldrums, and that's part of why we recognize, for the everything exchange, crypto's gonna still have ups and downs. So you wanna be able to offer any asset to any customer at any given time, so that there's a bull market in something at any given time.
Make sure you can capitalize on that.
Turner Novak:
It's like it's 5:00 somewhere.
Emilie Choi:
Yes. Exactly. I love it. Exactly.
Turner Novak:
And one thing you mentioned to me, this is like straight up, I should have probably asked you this way earlier, but we were talking about off-sites, company off-sites. How do you feel about doing company off-sites?
Emilie Choi:
Yeah. In general, I hate the concept of company off-sites, because I think they're like this feel-good moment where you go bonding, and you do the dinner, and you do these structured little segments. And then afterwards you're like, "Wait, did I ship anything during that time?"
So we very much pivoted to this whole idea of surges. Like, we're meeting in person for surges because we actually wanna ship things. Now, we will have the occasional off-site, because I think it's good for us to be like, "Okay, this is gonna be the company priorities in 2027, and this is how we have to layer them into the company, and these are the trade-offs that we're gonna make," and all that kind of stuff.
But for the most part, we're a remote first company. We have offices all around the world, but it's a remote first company. And so we wanna make sure that when we do get together, we're forcing actual shipping. And that we're excited to work together to go ship things, and to surge together to do that.
That's what I believe.
Turner Novak:
Interesting. Yeah. I feel like I get less done the more people I... Maybe I'm bad for this, but the more insulated I am, like when it's just me, there's no one else around me, I get so much done. But the more people and the more... Like if I'm at a conference, I'm getting nothing done that day.
'Cause it's just like 100 people. Like, "Hi." Well, maybe it's a different way of getting things done. Like, I'll catch up with 50 people. But I'm definitely more of those, like, I don't know, probably more introverted, so I'm not just going there commanding the room.
Like, maybe I use it, I treat it as a time to like, okay, catch up with 50 people, and then I got them all nailed out and done in one day versus... so it's like, I don't know.
Emilie Choi:
Similarly, that's why we still believe so much in the remote first culture, even though I know we could talk, we could do 10 podcasts about remote versus in-person culture and all that kind of stuff.
But there's a couple reasons. One is we believe very fervently that tapping into any talent across the world regardless of their location is a very big competitive advantage. And we have great examples of people who are all over the world who are proprietary in terms of their knowledge and skillset, and we just simply wouldn't be able to have them if we were in person.
Another, I think, in a world of AI, back to this whole thing of artifacts and codifying things in written documents, is we're really good at that because we're remote first. And so I think that in a world where you're gonna be using the AI agents to go disseminate the right insights and be able to build the right products at the right time because of this stuff, it's all in there, and it can just run with it.
Turner Novak:
Yeah.
Emilie Choi:
But to your point about being very effi... Like, I think we're very efficient, 'cause it's back to back to back to back, and we're getting our stuff done. And then if we have an external meeting, it almost derails from the day. It's like, "Oh my gosh, I have to go to this coffee. That's gonna be half an hour, and then another half an hour to get back," and all this kind of stuff.
Whereas I can jam that all in. I can jam those meetings in. I can jam that work in in a remote first way that is incredibly efficient.
Turner Novak:
Yeah. I like the way of thinking about it as a surge, not an off-site.
Emilie Choi:
Yeah.
Turner Novak:
Like, an off-site kind of gives vacation...
Emilie Choi:
Yes...
Turner Novak:
...vibes.
Emilie Choi:
It's like the office vibes. It's like a nice to have. Yes.
Turner Novak:
Whereas a surge, it's like, you know, you think of a surge of productivity, or of getting stuff done.
Emilie Choi:
Yes. A surge of shipping, yes.
Turner Novak:
Do you have... One last final question. Do you have a favorite founder, executive, famous person, like athlete, anyone that you've gotten a lot of inspiration from?
Emilie Choi:
So again, I'm so lucky. When I first went to Silicon Valley, I worked at Yahoo. I swear Yahoo was cool back in the day. And there were some of the best brains in Silicon Valley at Yahoo, and one of them was Jeff Weiner. And years later, I worked for him at LinkedIn. I think he's the best operator in the world.
Like, he just had this incredible mix of being able to get super into the weeds and spot things in the data that would tell you that a trend was good or alarming. And also be able to rise up and inspire the company. And I have a lot of sympathy towards this, because being the counterpart to Reid Hoffman, or being the counterpart to Brian Armstrong, the founder is gonna be a magnetic electric being with the crazy ideas and the ability to make mistakes in a way that is more forgivable or anything like that.
And so finding that right balance with that person is really challenging, and I think he did such a great job of it. Reid Hoffman often talks about how he had a founding moment with LinkedIn, but then he had another founding moment with Jeff. And I think that's a beautiful way to kind of have these evolutions.
So I learned so much from Jeff, including the thing I was talking about with the decision-making framework, and things about just how you scale a company in the right way when you're growing like crazy. And I still reference his old writings and blog posts about how to do this, because they're evergreen. Like, they stand the test of time. It's incredible what he came up with.
The other person at LinkedIn who was incredibly important to me was Kevin Scott. He's now the CTO of Microsoft. He was the head of engineering at LinkedIn. And I had so many moments with him, 'cause he was like a mentor to me. You know, I would be tearing up because one of the execs was mean to me about something in the room, and he was like, "Do you stand behind your work?" "Yes." "Then why are you crying?" Because he hurt my feelings. If you stand behind your work, you have to fricking stand behind your work.
And you can't let somebody make you feel bad. Only you can make yourself feel bad. And learning from people like that who kind of believed in me... I mean, Jeff, in the same way, would be like, "Sit at the fricking table. Don't sit in the back of the room." I realize now that they were just so generous in terms of their feedback. Like, they didn't have to give that feedback.
And so I just, like I said, I'm a lucky girl.
Turner Novak:
So you talked about the sitting at the table instead of in the back of the room. Is that something you think a lot of people struggle with, like, that concept?
And is that sort of like the, you get in there, you get more in the weeds, you take responsibility, or take ownership of your work kind of a concept?
Emilie Choi:
I think it's like, in my head, sometimes I think I was being deferential. And I think Jeff was saying, "You're an owner in this thing."
And it's very much inherent in what I was talking about in terms of agency and ownership, and these principles that Coinbase cares so much about, is just like...
Turner Novak:
Mm-hmm...
Emilie Choi:
...you're an owner. Sit at the fricking table and have an opinion and drive things forward, and don't be a passive participant.
And I think that's... All of these things have helped me be successful in these roles I've had ever since.
Turner Novak:
So is it like coming to an informed opinion? Like, you probably don't just have an opinion for the sake of having an opinion. You, like...
Emilie Choi:
Exactly...
Turner Novak:
...can talk about the opinion that you have.
Emilie Choi:
Totally.
Turner Novak:
And you're okay with maybe people think that you're wrong.
Emilie Choi:
Yeah. But...
Turner Novak:
...like, you came up with it of your own accord, from your own principles of why you think it's the right decision, or why it's the right opinion to have.
Emilie Choi:
Absolutely. Sometimes I'm gonna be wrong, and that's okay. But...
And to your point, I also think one of the things we talk about in these meetings, especially in a remote first way, is you don't have to say something in every meeting. In fact, I'm gonna value you more if you save your stuff for the stuff that you're really gonna be valuable for.
So that's another way that we assess performance. Is the person adding unique insights at the right time, and knowing when to not use the time and space for, you know, filler?
Turner Novak:
Interesting. Yeah. That makes sense. Well, cool. This has been a lot of fun. I mean, I feel like we...
Emilie Choi:
Super fun...
Turner Novak:
...we probably talked for like 90 minutes...
Emilie Choi:
Yeah.
Turner Novak:
...or something already.
Emilie Choi:
Yeah.
Turner Novak:
But...
Emilie Choi:
Flew by...
Turner Novak:
Yeah. Thanks again for doing this. This has been awesome.
Emilie Choi:
Thanks for having me, Turner. Super fun.
Find transcripts of all other episodes here.
