YC Tech Talks: Defi and Scalability with Nemil at Coinbase (S12)

Cool! Thanks everyone. I'm super excited to talk. My name is Nimail. I'm at the head of crypto at Coinbase, and I'm excited to talk to you today about DeFi and scalability. Um, but in part of talking about that, I'll talk about the landscape for crypto and why today is just such an exciting time to be able to get into crypto.

Um, so to kick it off, um, what I loved seeing when I was a YC founder was seeing the journey, especially the early years about the founders, um, who had been successful in Y Combinator. And so when I saw this, it was recently a tweet by Fred Urson, one of the co-founders with Brian Armstrong of Coinbase, um, walking through the first version of Coinbase and where we are today: one of the largest, um, downloaded apps in the world.

It's been quite a journey over the last decade to get to where we are, but we're a small fraction of the way in the crypto industry to where we need to be to allow this technology to be accessed by everyone around the world. So quick uh, background on Coinbase, and I think many of you may be aware, but generally speaking, like the core things that we offer are we have a whole series of different users on our platform; they trade crypto, they hold crypto, um, safely on our platform.

We have a whole host of uh, different uh, employees, um, and we're rapidly growing. Um, and we're a Y Combinator company; we started in the summer of 2012. I was actually in that batch. One of the reflections that I had was unlike, I think, some of the founders here today, I actually shut down my startup. But while I was doing that, I was accepting Bitcoin for my startup, and I got super excited about this idea of Bitcoin scripting.

Very few people talk about it today, but it was an early originator of the idea of smart contracts. And then when Ethereum came out, I got super excited about that, built a whole host of security best practices, audited all a number of different protocols in the space, and then launched my own DeFi protocol.

The journey at Coinbase has been really fantastic. One is that I used to run USDC; it's today about a $30 billion stablecoin that's available around the world. We launched Rosetta, a way for different blockchains to interoperate together, and we also announced the um, different types of open-source funding to help grow the crypto ecosystem.

The core things I think I want to drive home in the chat today are like, one is that the price of Bitcoin is one of the most powerful onboarding mechanisms in the history of technology. So basically, what happens is your price cycle: the price of Bitcoin goes up, a lot more people get excited about it. They start signing up for crypto wallets; they start signing up for trading platforms, and then they go deeper and deeper down the rabbit hole.

You've seen that over the course of the maybe three or four cycles of the last decade that we've gone through. But it really has driven hundreds of millions of people into this space, um, in a way where all this technology is today still very rudimentary. It's hard to use, um, it's not super scalable, and yet the demand has just been off the charts.

So I think that's one observation: that Bitcoin and the growth of this asset ecosystem has really driven people into crypto. The second thing I think that I always explain to people is often people think about crypto as finance and when I, or, or payments is the other quick thing—the idea that it's going to solve remittances.

I think the core thing I just want to drive home to all those excited against the space is not just payments, it's not just cryptocurrencies, it's not just finance. And I think Devon Vinson from OpenSea is a really great example of that. NFTs are a behavior which has some analogs in the pre-blockchain world, but in a lot of ways, it's brand new behavior, much like the internet led to things like social media that we could barely have imagined at the time.

So, that's a huge shift in how we think about this: is that anything that can be owned can be moved, um, and can be held on the blockchain ultimately. And you've heard this from a few other folks that talked earlier today is where we're headed is billions of users and millions of developers, and we're a small, small, small fraction of the way there.

We might have 150-200 million users in the world today, and I probably would argue there are tens of thousands of developers, when for web development there might be tens of millions of developers. The big shift we're slowly getting is that if on the internet we have this whole OSI model—it's how we set up the stack of how we build apps on the internet—we're slowly bringing that together in the crypto model.

And it's an exciting time because early on when I started in crypto there was just a layer one layer, then slowly added layer two; we added base contracts and apps on top of that, and that's what we're gonna talk about a little bit more today.

So I think the core maybe career advice for the folks who are looking to get in, especially for the first time, is one feedback I get from a lot of people is like, is it too late? The price of Bitcoin is this; the price of Ethereum is this; NFTs exploded. I feel like it's too late. And I just want to tell you that a lot of us, like, what we see every day is a tiny fraction of what the rest of the world is actually using, right?

So I would say it's not too late; this is actually the perfect time. I actually would argue that I probably got in too early. Um, but the core shifts that have happened, of course, in the last few years are one: just powerful primitives.

So we have smart contracts; we have the ability for layer one platforms that allow all this other stuff without you having to build your own layer one platform. Just like we don't have to build TCP/IP to ship a website, there's wide open space like the apps that have been created—we're still scratching the surface. We tried one form of NFT digital art, but there's a host of other things that we need to explore.

And that, um, not least then I think there's critical technical and product problems. And this, as an engineer, gets me very excited: the scalability, the security, usability, and I'll walk through a few more of these later.

Um, there's a growing user base; many more people are being onboarded every day into this ecosystem, and that's amazing if you're a startup founder, if you're an employee at a company. And then, at least for now, there's easy funding. Again, crypto goes through cycles, but um, this is a really powerful way if you want to start a company.

I remember how hard it was to raise my first round as a YC founder, and this is pretty amazing if you're a founder to at least get started. So here's just some things for you to think about as you think about whether you want to get into crypto. It's like these are the types of technical problems that we get to work on.

One is scalability: how do you onboard to billions of users? How do you onboard to millions of developers? How do you build the tools? So like NYC, Stripe is the one that I, when I started NYC, I was the company most excited about, and this is exactly something: how do we make the crypto developer experience like Stripe?

Another example is security. So you might hear about DeFi hacks, and we'll talk about that in a minute, but like this is a really, really critical problem in crypto: how do you make sure that funds are safe? Usability: so there's so much going on below the surface, and when I go to a website, I don't have to think about what's happening in the TCP layer; I don't have to think about what's happening, you know, an IP layer and the layers above that.

All I have to think about is I want to go to a website; I want to interact with it. Today we have not abstracted that away, and that's a really powerful and interesting engineering problem. And then I think one other one that's really interesting is interoperability: how do we get all these different systems, layer one, layer two to talk to each other?

So I think quick things I'll drive home is I'm from the speakers earlier today. I talked about DeFi, and I think the thing that's important to frame is why is DeFi so interesting. It's open; it's available around the world. That's a massive, massive shift in how we think about financial services. Up till the 90s you couldn't access a—you can actually have a bank go across state lines.

And yet, just like the internet did for information, crypto is doing for ownership and finance. It's composable. So if you're a developer, you're very used to composability. When I do my npm install, I don't have to build every single command. Not similar; that's what really powerful about DeFi.

It's programmable, and it's the first time we've been able to create what I call credible digital scarcity. That means that I can't—I can't just change a database and change a number somewhere to change a money supply or to change a rewards currency or anything like that, and that's a huge shift as well.

And so when I compare it, I think like one example of these is comparing the old-world lending companies and DeFi protocols together. It's like, you know, there's a process for a lending company; there's a process for a DeFi protocol.

And this might be the simplest version of a bank. I think I wrote this in 2016: it lets you put in money, it lets you pull out money. The really cool thing about this is as a creator of this—I'm the founder, YC, who's creating this—I can't touch the money. And that's a really profound shift in how we think about financial services.

One of the reasons we regulate banks so much is that the creators of that can run away with the money, and this really, really shifts that equation. And so if you think about it, this is a magical time to get into crypto—you’re a traditional example of like a bank of how it started.

You look at the default lending, and in a lot of ways, it's superior. There are other things that we have to figure out in this model, but I think this is a really exciting time. So I think the main thing I want to drive home in this talk is this: um, is that all of this excitement will not last unless we can solve one primary problem in crypto, which is scaling.

So if you look at scale, Bitcoin is 45 transactions per second; on the right-hand side, you can see what Visa has, right? And there's a tremendous amount of difference between that. The three main ways to scale a blockchain: the financial upgrade, the two pointers as an example of that; they're layer twos. A lot of these coming out, and then there's layer one competitors.

People are trying to refine that model, and so one of the challenges that we have to deal with as a company is that the user has some currency on one, um, thing—layer one, ETH—and they want to move it to somewhere else. And this is today a very complex process; they have to pay a lot of fees on layer one. It says 20 to 30; often it's 100 to 200 now.

Then you have to switch your wallet over; you need to be able to receive the funds, and often this might take up to a week to do. And so the vision that we have and the type of technical challenge we get to work on if you're interested is like things like how do you make this bridging just seamless and easy? The same way I don't have to think about this website's different from this website; I just have my assets and I want to use them wherever I want to use them.

And so to us, like a profound, powerful end-state vision is that a lot of these DeFi protocols that are looking for users, they need the ability to move funds easily across all these different options. Otherwise, what is going to happen is that a lot of this activity may just lead to really, really high gas prices that mean that retail users cannot participate in the process.

So, um, that's it on my end. I think the quick thing I'll say is I'm gonna be—we're hiring. Um, you can reach out directly to me; my email is right here. You can also come hang out with us and you know, jam on some of the problems. I would say compared to some of the startups, I think a startup is a super early stage; startups are like a super exciting place to work.

I love that experience. The exciting things about like where we are in our trajectory is that you get to work on some of the foundational problems for the space as well. So like open source—you get to work on how do I solve a layer two for the space? And my hope is that if we solve certain problems, there will be a hundred or a thousand more YC startups in this space because of the things that we've built that allow the entire space to grow.

Thanks everyone!