theCUBE + NYSE Wired: Crypto Trailblazers | Bill Platt, Alchemy

Clips
More from theCUBE + NYSE Wired: Crypto Trailblazers

Bill Platt

COO

Alchemy

play_circle_outline Introduction of Bill Platt, COO of Alchemy, on the Crypto Trailblazers program.

play_circle_outline Bill's reflection on his career and desire to advance finance technology.

play_circle_outline Comparison between cloud disruptions and current disruptions in Decentralized Finance (DeFi).

play_circle_outline Bridging Traditional Finance and Decentralized Finance: Merging Reliability with Programmability and Speed for a New Financial Future

play_circle_outline Analyzing Venmo vs. Blockchain: Fundamental Differences and Recent Events Impacting Crypto Transaction Reliability

Info

Bill Platt, Alchemy

Bill Platt

COO Alchemy

search

John Furrier

>> Hello, I'm John Furrier, host of theCUBE, with Dave Vellante here at the NYC CUBE Studios. This is our Crypto Trailblazers program, part of the NYC Wired program and community. We have Bill Platt here, COO of Alchemy. We had the founder on earlier in its previous episode. This is the series where we break out the leaders and unpack the issues that are driving the revolution of DeFi, TradFi. Again, we've seen this movie with the cloud before, now we're seeing it in the decentralized finance infrastructure. Bill, thanks for coming in. We appreciate you.

Bill Platt

>> Thank you for having me. I appreciate being here.

John Furrier

>> We were riffing and talking before we came on camera of your career at AWS and you were a cloud OG, Engine Yard. So you had that whole cloud wave. And you see a similar trajectory and you kind of came out of retirement from AWS. You've got a lot of success there. Joining Alchemy, it's like, okay, tell us why. And do you see the same disruption that cloud had here?

Bill Platt

>> Yeah, certainly. So I jokingly say I'm like Taylor Swift. I've had many eras of computing. I started out programming mainframes when I was younger. Believe it or not, worked with some MIT professors on vector databases when I was like 24 years old. And then I worked at Sun Microsystems through the internet era and then saw the dawning of the cloud when I was working at a startup and then became part of AWS, and then the last couple of years at AWS, even agentic AI, right? So I've seen all of these lovely eras of computing. I also, after 12 years at AWS, did a lot of things, but I also decided it was time for reflection. I took a break and I started looking at life and also at what's happening in technology. And I realized maybe also jokingly, like Taylor Swift, I realized it's me. Hi, I'm the problem. I've actually ushered all of these transformations in technology and helped many people in the financial services industry move forward on their technology infrastructure, but we hadn't really affected the finance part yet, and I hadn't directly actually participated in moving finance forward in any way. And so as I reflected... And as I expected, some people called me, are you interested in doing certain things? And I said, "I would like to learn a little bit more about what this company, Alchemy, is doing and really understand decentralized finance and understand. Can we?" And I do believe it's true, there's an inflection point here where architecture change can help reshape an industry.

John Furrier

>> Yeah. Dave and I talked about our podcast thing, last two podcasts we've had, we talked about how people are, young and old, leading in. Because if you know systems and you've seen the errors, you look at this market and your eyes wide open, "Wow, this is a really big opportunity." And Amazon's early days, you were there, basic building blocks. It's compute, basically servers, storage, and networking, and queuing. Basic stuff. Then higher level services came out. That brought in new brands like Dropbox, Airbnb, the list is endless. The charts are out there. But it improved scale and productivity, lift and shift, move from a data center to the cloud. You're getting at something different where the structural financial infrastructure change. What does that look like to you? Because there was structural change with the cloud that enabled a certain set of scalable value and value extraction. But finance with DeFi, it's an infrastructure, almost IaaS and PaaS kind of movement, but different. What is your view on the opportunity, and then what does the structural inflection look like?

Bill Platt

>> Sure. So the structure today of what happens in decentralized finance, there's a couple of aspects of it that are really ripe for adding to. So decentralized finance infrastructures have some really wonderful properties. They have the ability to be, you can be fast, you can settle things up automatically. You can deliver programmability to money, right? This is wonderful. But traditional infrastructures, like the things that I came from AWS, have incredible reliability. They have incredible regulatory advantages, and we want to merge those two things. And the inflection there, and the reason why I think it's so exciting is that until recently, things like AI weren't being able to be applied to things like the blockchain, but now we believe they are. And I'll just give you an example. In our case, we recently released a thing called Cortex, which is our intelligent infrastructure engine. So we can assemble the components necessary through automation. We can allow for agentic autonomous financial systems to be built on top of it because it's built for that purpose. It's two and a half times faster, it's 50 times less expensive, and it's 500 times more throughput. That allows this transformation to really take hold, just the way it happened with the cloud when you finally had the building blocks set. So we are a level of abstraction above, technically above the layers underneath, but we're doing that in a decentralized way, which has some of these wonderful properties.

Dave Vellante

>> So what you said is profound because you're talking about moving finance forward versus moving the IT department of finance forward. Could it happen without AI, gen AI and agentic?

Bill Platt

>> It certainly could happen without that. It would take a lot more human horsepower in the beginning and be a little slower to happen, but it could. Now, one of the wonderful properties about this shift in the ability to generate agentic AI, agentic workflows is that you can build a company to build a decentralized finance app or to build a FinTech app. You can build that with twice as fast with one-tenth the cost. And that's the same fuel that helped that grew the cloud, that investors could cough up a whole lot less money because they didn't have to buy infrastructure themselves and order to move forward. And now you can build a new decentralized finance app in weeks with a few people, which is really powerful for unlocking innovation.

Dave Vellante

>> So what is the traditional, because you talked about merging essentially the resilience of the financial services industry with all the benefits and of course the novelty of blockchain. What has to happen to the underlying scaffolding within the financial services industry? Or is it sort of greenfield DeFi firms that are obviously going to disrupt this? But how do those two worlds come together?

Bill Platt

>> Yeah, great question. So part of the reason why I'm here at Alchemy and what I'm excited about is bringing some of the discipline and the innovation from a company like AWS into the decentralized finance world. What we're seeing today is you're seeing more aggressive and ambitious traditional finance companies, JPMorgan for instance, already moving into FinTech. And FinTechs are leveraging greater and greater amounts of decentralized finance. Because that overlap is growing, the folks on the... So you've got all these advantages of traditional finance infrastructures, but they really need the programmability, and the speed, and the 24/7 availability that decentralized finance offers. So we see it as integrating the two.

Dave Vellante

>> So I wonder if you could explain something because I get this question a lot because we're obviously big fans of crypto and have been following this industry for quite some time. People say, "Well, what's the difference between Zelle and Venmo? I mean, I can do that already. I can transfer..." I say, "Okay, but the underlying architecture and the ability to do things in a decentralized way anywhere in the globe." But can you explain, and from your point of view, what's different and what the future looks like?

Bill Platt

>> I'll do my best. Some of the differences? So I'll actually use an example of outages, okay? So recently, there were two significant events that occurred around the world. One on October 13th, one of the largest liquidation events in the history of crypto occurred, maybe the largest.

Dave Vellante

>> I think the largest. I think it was, yeah.

Bill Platt

>> And interestingly, lots of people were stuck when that happened. They couldn't actually transact. Secondarily, on October 20th, AWS had maybe its largest scale event ever. Now, I've been on the other side of those events, and I know it takes an awful lot to cause that, but it does also show you that the best of the best can still have these. Interestingly, in both cases, Alchemy continued to operate fully to deliver its infrastructure and the core APIs and the core streaming services that are necessary for blockchain and for crypto. And I say that to you because really that is one of the big advantages of taking a true decentralized approach to how you're going to do infrastructure. We built this thing called Cortex, which is our intelligent infrastructure engine. And through it, we also are able to provide against, we really believe this philosophy is the right one, multiple cloud providers across multiple regions with no single point of failure. And that is the thing that decentralized thinking is really good at. And that's an interesting property to bring back to, say a JPMorgan, who wants to have a faster ability to do things, but also still needs that core number nines have to be there or they can't do it.

Dave Vellante

>> You're raising the bar on well architected.

John Furrier

>> It's a new era. You mentioned Amazon's outage and the new architecture, and the developers are a big focus of what you guys are targeting, the new developers. There's entrepreneurial activity in finance. Entrepreneurs are creating new products. You mentioned some of the things like DeFi, staking, all these new financial products. It's almost like a product management opportunity here. But Amazon had a term called primitives.

Bill Platt

>> Yes.

John Furrier

>> Nikil and I were talking, founder of Alchemy. He said, "Crypto is a new primitive after computers and the internet." I've had one other founder saying, "The internet is like fax machines, it's going to go away. But decentralized is a little bit different." So if primitives are going to be important, they were critical for the developer in the cloud game. So what does that look like for you guys? Obviously you've got the platform. What will developers do with the primitives? Is there set building blocks that need to be locked in? What's your thoughts on the role? What is a primitive in financial infrastructure?

Bill Platt

>> So continuing the analogy a little bit, so the primitives that we're talking about today are when you have a transaction, it's going to be made up of some components that represent a transaction. For instance, I want to swap a currency or I want to swap a digital asset for another digital asset. And so you need to have highly available APIs that understand what it means to go get the right piece of data, bring it back, allow you to transact on that, bring that back to the blockchain, and ensure that it's actually transparent, that it's actually correct, and that, in fact, the transaction occurs at a speed that it really is required. And so those primitives include super low latency interconnections. We're talking about sub-50 millisecond interactions all the way from the endpoint to the chain and back. That primitive itself is something that we work very hard on just to make sure that that's really, really substantial. Another great primitive is can you fetch a large amount of data easily and both quickly, but will it also be correct? And correctness being a huge primitive. So it's those types of primitives that make up the new form of how you compute against decentralized finance.

John Furrier

>> I bring up the fax machine over the top haymaker because the trends that we're seeing in this AI-enabled infrastructure on the, say the NVIDIA side and the AI build-outs, and on DeFi, is that it's a different kind of OS. I've heard Jensen Wong say KV cache is the OS for supercomputing. That's their network. That's networking. So I just posted on LinkedIn today, networking is the new OS. So if you think about things like timing, if you're going to have low-level latency on trades, how do you manage the timestamps? Because you're decentralized, is that a primitive, like an atomic clock or some sort of quantum clock? Or is there going to be a new thing?

Bill Platt

>> I don't necessarily think an atomic clock as the primitive here, but I do think that-

John Furrier

>> Well, timing, right?

Bill Platt

>> Well, the timing is the big deal. The timing is the big deal. And some really good sophisticated engineers continually are working on this. Where the AI fuel comes from is that when you have a system that has to learn over time how to keep those things in sync with brand new applications, which people are just inventing right now, new ways to take advantage of, for instance, a stablecoin, right? In order to take advantage of a stable coin, people are trying to build applications that, on one side, are interacting with the more traditional finance world. On the other side are trying to create yields, really useful yield. My children really want eight point, seven point yields on the small amount of money that they've been able to accumulate. And so that timing thing is really important. But also the reason why we built an infrastructure that's ready to serve agents is that as the agents can learn these application spaces, they can do that really, really tricky thing that's just to keep all this timing correct, keep all the data flowing correctly, and allow people to build above that kind of hard to do level of, is the actual timing perfect, by saying, "It is perfect, so you're good to go. Go ahead, call me and I'll give you the right information."

Dave Vellante

>> So if I understand it correctly, you've built an abstraction layer to simplify so that those primitives, you don't have to get down and dirty on those. So you've got tooling to do that. And as a result, you've got challenges in terms of performance, SLAs. So maybe you could describe a little bit how you're able to build in that abstraction layer, which sometimes causes overhead, and yet at the same time deliver the type of performance in SLAs that people demand.

Bill Platt

>> Yeah, it's interesting. A little bit of going back in time to relearn some lessons, but then build it into some agents, so it's kind of fun. And by no means am I saying this is done. This is all work in process. Everything we're talking about is why I came here because we have to drive this forward, and we know that there's some really important properties that we have to keep making better.

Dave Vellante

>> And you know it's technically feasible. You can see-

Bill Platt

>> It's definitely technically feasible. It does help that AI just makes it a lot easier to write code and test code so that you can go quickly on this. But fundamentally, at a certain level, you have to make some architectural choices of how close to the metal do you have to make this action if you're going to keep the latency that low. And that might not sound very AWS cloud-like, but it's actually what AWS does way down below that nobody has to touch, but it is happening, choices every day of how to architect both hardware and software to do that. And that's some of the really cool innovations that are happening in the decentralized finance space is even at the system-by-system level, is how to serve this blockchain in a fashion that's really fast, really reliable, and SLAs can be met. But it does require us, and we have built some architectural shifts all the way down to right next to the metal in order to make sure that the system is fast.

Dave Vellante

>> And can scale.

Bill Platt

>> Exactly.

John Furrier

>> Well, great to have you on. I wish we had more time. We'll certainly do a follow up. Love to keep in touch. Love the engineering, because they're engineering going on. And again, it's a whole 'nother infrastructure. Give a quick plug for some of the stats.

Bill Platt

>> Sure.

John Furrier

>> How many customers, countries you're in, billions of transaction volumes you're doing? How many APL? Give me some numbers. Give us some plugging.

Bill Platt

>> So we're currently powering 70% of the crypto apps. We're supporting 80% of the stablecoins, including some of the biggest ones, the Tethers, the Circles. This is the enterprise of that space. And stablecoins plus AI is the killer app for the moment, right? It's really, really important to the world. We're doing over 100 billion plus transactions a day right now. We think that's probably 10X less than it will be in a year.

John Furrier

>> Billion events, not dollars.

Bill Platt

>> Yeah. Events. Transaction events.

John Furrier

>> Okay.

Bill Platt

>> Yes, correct. And supporting our latest improvements in this infrastructure delivered by Cortex are supporting 2.5 times faster, and, like I said, 500 times the throughput of any previous decentralized finance.

John Furrier

>> And your view on apps, and you got the apps, is Robinhood the next AI native-like transformation? They're transforming very fast.

Bill Platt

>> Yes.

John Furrier

>> You're seeing Stripe come out with their own chain. I mean, everyone's got payment rails, settlement rails are coming.

Bill Platt

>> Yes. So I do think that there's a great set of innovations occurring at the payment level right now. If you've heard of x402, this is a really big step forward. People like Google supporting agentic payment systems. This is a very big step forward at the top of this stack, allowing for application to be built, which allow us to, like I said, to build autonomous financial services with humans attached, which creates a great financial community of the future. And so that's why we're building all of this to make sure that that actually can happen, and we're supporting some of the best in the industry in order to do it.

Dave Vellante

>> And your point about stablecoins in the killer app is because of liquidity, is that right?

Bill Platt

>> Definitely. Liquidity and allowing the traditional folks to feel more comfortable. All the regulations, GENIUS Act, all these things are helping create-

John Furrier

>> Yeah. For sure.

Bill Platt

>> more or less create the environment, the right climate so that we can grow.

John Furrier

>> Indeed. What about Venmo? It's just another digital payment mechanism on your app. If you're an app provider, it's just like, "Oh, I got custody issues. They go away under the covers. I'm just buying with my crypto."

Bill Platt

>> Exactly.

John Furrier

>> And anywhere in the world, by the way,

Bill Platt

>> Exactly.

John Furrier

>> US dollar backed.

Bill Platt

>> It's actually-

John Furrier

>> Good for the US.

Bill Platt

>> really amazing, right?

Dave Vellante

>> And it's essentially any asset that you can tokenize.

Bill Platt

>> Any asset. I mean, BlackRock wants to tokenize everything, right?

Dave Vellante

>> Sure.

John Furrier

>> Well, Bill, they're lucky to have you with all your experience. Thanks for coming on. You're certainly a trailblazer on the cloud, now crypto. Thanks for joining the program.

Bill Platt

>> Thank you very much.

John Furrier

>> I'm John Furrier, with Dave Vellante. We are here at theCUBE's NYSE Wired studio, part of the programming community. We're building, of course, theCUBE and the NYSE coming together to bring all this great programming to you. Thanks for watching.

Bill Platt, Alchemy

search

John Furrier

>> Hello, I'm John Furrier, host of theCUBE, with Dave Vellante here at the NYC CUBE Studios. This is our Crypto Trailblazers program, part of the NYC Wired program and community. We have Bill Platt here, COO of Alchemy. We had the founder on earlier in its previous episode. This is the series where we break out the leaders and unpack the issues that are driving the revolution of DeFi, TradFi. Again, we've seen this movie with the cloud before, now we're seeing it in the decentralized finance infrastructure. Bill, thanks for coming in. We appreciate you.

Bill Platt

>> Thank you for having me. I appreciate being here.

John Furrier

>> We were riffing and talking before we came on camera of your career at AWS and you were a cloud OG, Engine Yard. So you had that whole cloud wave. And you see a similar trajectory and you kind of came out of retirement from AWS. You've got a lot of success there. Joining Alchemy, it's like, okay, tell us why. And do you see the same disruption that cloud had here?

Bill Platt

>> Yeah, certainly. So I jokingly say I'm like Taylor Swift. I've had many eras of computing. I started out programming mainframes when I was younger. Believe it or not, worked with some MIT professors on vector databases when I was like 24 years old. And then I worked at Sun Microsystems through the internet era and then saw the dawning of the cloud when I was working at a startup and then became part of AWS, and then the last couple of years at AWS, even agentic AI, right? So I've seen all of these lovely eras of computing. I also, after 12 years at AWS, did a lot of things, but I also decided it was time for reflection. I took a break and I started looking at life and also at what's happening in technology. And I realized maybe also jokingly, like Taylor Swift, I realized it's me. Hi, I'm the problem. I've actually ushered all of these transformations in technology and helped many people in the financial services industry move forward on their technology infrastructure, but we hadn't really affected the finance part yet, and I hadn't directly actually participated in moving finance forward in any way. And so as I reflected... And as I expected, some people called me, are you interested in doing certain things? And I said, "I would like to learn a little bit more about what this company, Alchemy, is doing and really understand decentralized finance and understand. Can we?" And I do believe it's true, there's an inflection point here where architecture change can help reshape an industry.

John Furrier

>> Yeah. Dave and I talked about our podcast thing, last two podcasts we've had, we talked about how people are, young and old, leading in. Because if you know systems and you've seen the errors, you look at this market and your eyes wide open, "Wow, this is a really big opportunity." And Amazon's early days, you were there, basic building blocks. It's compute, basically servers, storage, and networking, and queuing. Basic stuff. Then higher level services came out. That brought in new brands like Dropbox, Airbnb, the list is endless. The charts are out there. But it improved scale and productivity, lift and shift, move from a data center to the cloud. You're getting at something different where the structural financial infrastructure change. What does that look like to you? Because there was structural change with the cloud that enabled a certain set of scalable value and value extraction. But finance with DeFi, it's an infrastructure, almost IaaS and PaaS kind of movement, but different. What is your view on the opportunity, and then what does the structural inflection look like?

Bill Platt

>> Sure. So the structure today of what happens in decentralized finance, there's a couple of aspects of it that are really ripe for adding to. So decentralized finance infrastructures have some really wonderful properties. They have the ability to be, you can be fast, you can settle things up automatically. You can deliver programmability to money, right? This is wonderful. But traditional infrastructures, like the things that I came from AWS, have incredible reliability. They have incredible regulatory advantages, and we want to merge those two things. And the inflection there, and the reason why I think it's so exciting is that until recently, things like AI weren't being able to be applied to things like the blockchain, but now we believe they are. And I'll just give you an example. In our case, we recently released a thing called Cortex, which is our intelligent infrastructure engine. So we can assemble the components necessary through automation. We can allow for agentic autonomous financial systems to be built on top of it because it's built for that purpose. It's two and a half times faster, it's 50 times less expensive, and it's 500 times more throughput. That allows this transformation to really take hold, just the way it happened with the cloud when you finally had the building blocks set. So we are a level of abstraction above, technically above the layers underneath, but we're doing that in a decentralized way, which has some of these wonderful properties.

Dave Vellante

>> So what you said is profound because you're talking about moving finance forward versus moving the IT department of finance forward. Could it happen without AI, gen AI and agentic?

Bill Platt

>> It certainly could happen without that. It would take a lot more human horsepower in the beginning and be a little slower to happen, but it could. Now, one of the wonderful properties about this shift in the ability to generate agentic AI, agentic workflows is that you can build a company to build a decentralized finance app or to build a FinTech app. You can build that with twice as fast with one-tenth the cost. And that's the same fuel that helped that grew the cloud, that investors could cough up a whole lot less money because they didn't have to buy infrastructure themselves and order to move forward. And now you can build a new decentralized finance app in weeks with a few people, which is really powerful for unlocking innovation.

Dave Vellante

>> So what is the traditional, because you talked about merging essentially the resilience of the financial services industry with all the benefits and of course the novelty of blockchain. What has to happen to the underlying scaffolding within the financial services industry? Or is it sort of greenfield DeFi firms that are obviously going to disrupt this? But how do those two worlds come together?

Bill Platt

>> Yeah, great question. So part of the reason why I'm here at Alchemy and what I'm excited about is bringing some of the discipline and the innovation from a company like AWS into the decentralized finance world. What we're seeing today is you're seeing more aggressive and ambitious traditional finance companies, JPMorgan for instance, already moving into FinTech. And FinTechs are leveraging greater and greater amounts of decentralized finance. Because that overlap is growing, the folks on the... So you've got all these advantages of traditional finance infrastructures, but they really need the programmability, and the speed, and the 24/7 availability that decentralized finance offers. So we see it as integrating the two.

Dave Vellante

>> So I wonder if you could explain something because I get this question a lot because we're obviously big fans of crypto and have been following this industry for quite some time. People say, "Well, what's the difference between Zelle and Venmo? I mean, I can do that already. I can transfer..." I say, "Okay, but the underlying architecture and the ability to do things in a decentralized way anywhere in the globe." But can you explain, and from your point of view, what's different and what the future looks like?

Bill Platt

>> I'll do my best. Some of the differences? So I'll actually use an example of outages, okay? So recently, there were two significant events that occurred around the world. One on October 13th, one of the largest liquidation events in the history of crypto occurred, maybe the largest.

Dave Vellante

>> I think the largest. I think it was, yeah.

Bill Platt

>> And interestingly, lots of people were stuck when that happened. They couldn't actually transact. Secondarily, on October 20th, AWS had maybe its largest scale event ever. Now, I've been on the other side of those events, and I know it takes an awful lot to cause that, but it does also show you that the best of the best can still have these. Interestingly, in both cases, Alchemy continued to operate fully to deliver its infrastructure and the core APIs and the core streaming services that are necessary for blockchain and for crypto. And I say that to you because really that is one of the big advantages of taking a true decentralized approach to how you're going to do infrastructure. We built this thing called Cortex, which is our intelligent infrastructure engine. And through it, we also are able to provide against, we really believe this philosophy is the right one, multiple cloud providers across multiple regions with no single point of failure. And that is the thing that decentralized thinking is really good at. And that's an interesting property to bring back to, say a JPMorgan, who wants to have a faster ability to do things, but also still needs that core number nines have to be there or they can't do it.

Dave Vellante

>> You're raising the bar on well architected.

John Furrier

>> It's a new era. You mentioned Amazon's outage and the new architecture, and the developers are a big focus of what you guys are targeting, the new developers. There's entrepreneurial activity in finance. Entrepreneurs are creating new products. You mentioned some of the things like DeFi, staking, all these new financial products. It's almost like a product management opportunity here. But Amazon had a term called primitives.

Bill Platt

>> Yes.

John Furrier

>> Nikil and I were talking, founder of Alchemy. He said, "Crypto is a new primitive after computers and the internet." I've had one other founder saying, "The internet is like fax machines, it's going to go away. But decentralized is a little bit different." So if primitives are going to be important, they were critical for the developer in the cloud game. So what does that look like for you guys? Obviously you've got the platform. What will developers do with the primitives? Is there set building blocks that need to be locked in? What's your thoughts on the role? What is a primitive in financial infrastructure?

Bill Platt

>> So continuing the analogy a little bit, so the primitives that we're talking about today are when you have a transaction, it's going to be made up of some components that represent a transaction. For instance, I want to swap a currency or I want to swap a digital asset for another digital asset. And so you need to have highly available APIs that understand what it means to go get the right piece of data, bring it back, allow you to transact on that, bring that back to the blockchain, and ensure that it's actually transparent, that it's actually correct, and that, in fact, the transaction occurs at a speed that it really is required. And so those primitives include super low latency interconnections. We're talking about sub-50 millisecond interactions all the way from the endpoint to the chain and back. That primitive itself is something that we work very hard on just to make sure that that's really, really substantial. Another great primitive is can you fetch a large amount of data easily and both quickly, but will it also be correct? And correctness being a huge primitive. So it's those types of primitives that make up the new form of how you compute against decentralized finance.

John Furrier

>> I bring up the fax machine over the top haymaker because the trends that we're seeing in this AI-enabled infrastructure on the, say the NVIDIA side and the AI build-outs, and on DeFi, is that it's a different kind of OS. I've heard Jensen Wong say KV cache is the OS for supercomputing. That's their network. That's networking. So I just posted on LinkedIn today, networking is the new OS. So if you think about things like timing, if you're going to have low-level latency on trades, how do you manage the timestamps? Because you're decentralized, is that a primitive, like an atomic clock or some sort of quantum clock? Or is there going to be a new thing?

Bill Platt

>> I don't necessarily think an atomic clock as the primitive here, but I do think that-

John Furrier

>> Well, timing, right?

Bill Platt

>> Well, the timing is the big deal. The timing is the big deal. And some really good sophisticated engineers continually are working on this. Where the AI fuel comes from is that when you have a system that has to learn over time how to keep those things in sync with brand new applications, which people are just inventing right now, new ways to take advantage of, for instance, a stablecoin, right? In order to take advantage of a stable coin, people are trying to build applications that, on one side, are interacting with the more traditional finance world. On the other side are trying to create yields, really useful yield. My children really want eight point, seven point yields on the small amount of money that they've been able to accumulate. And so that timing thing is really important. But also the reason why we built an infrastructure that's ready to serve agents is that as the agents can learn these application spaces, they can do that really, really tricky thing that's just to keep all this timing correct, keep all the data flowing correctly, and allow people to build above that kind of hard to do level of, is the actual timing perfect, by saying, "It is perfect, so you're good to go. Go ahead, call me and I'll give you the right information."

Dave Vellante

>> So if I understand it correctly, you've built an abstraction layer to simplify so that those primitives, you don't have to get down and dirty on those. So you've got tooling to do that. And as a result, you've got challenges in terms of performance, SLAs. So maybe you could describe a little bit how you're able to build in that abstraction layer, which sometimes causes overhead, and yet at the same time deliver the type of performance in SLAs that people demand.

Bill Platt

>> Yeah, it's interesting. A little bit of going back in time to relearn some lessons, but then build it into some agents, so it's kind of fun. And by no means am I saying this is done. This is all work in process. Everything we're talking about is why I came here because we have to drive this forward, and we know that there's some really important properties that we have to keep making better.

Dave Vellante

>> And you know it's technically feasible. You can see-

Bill Platt

>> It's definitely technically feasible. It does help that AI just makes it a lot easier to write code and test code so that you can go quickly on this. But fundamentally, at a certain level, you have to make some architectural choices of how close to the metal do you have to make this action if you're going to keep the latency that low. And that might not sound very AWS cloud-like, but it's actually what AWS does way down below that nobody has to touch, but it is happening, choices every day of how to architect both hardware and software to do that. And that's some of the really cool innovations that are happening in the decentralized finance space is even at the system-by-system level, is how to serve this blockchain in a fashion that's really fast, really reliable, and SLAs can be met. But it does require us, and we have built some architectural shifts all the way down to right next to the metal in order to make sure that the system is fast.

Dave Vellante

>> And can scale.

Bill Platt

>> Exactly.

John Furrier

>> Well, great to have you on. I wish we had more time. We'll certainly do a follow up. Love to keep in touch. Love the engineering, because they're engineering going on. And again, it's a whole 'nother infrastructure. Give a quick plug for some of the stats.

Bill Platt

>> Sure.

John Furrier

>> How many customers, countries you're in, billions of transaction volumes you're doing? How many APL? Give me some numbers. Give us some plugging.

Bill Platt

>> So we're currently powering 70% of the crypto apps. We're supporting 80% of the stablecoins, including some of the biggest ones, the Tethers, the Circles. This is the enterprise of that space. And stablecoins plus AI is the killer app for the moment, right? It's really, really important to the world. We're doing over 100 billion plus transactions a day right now. We think that's probably 10X less than it will be in a year.

John Furrier

>> Billion events, not dollars.

Bill Platt

>> Yeah. Events. Transaction events.

John Furrier

>> Okay.

Bill Platt

>> Yes, correct. And supporting our latest improvements in this infrastructure delivered by Cortex are supporting 2.5 times faster, and, like I said, 500 times the throughput of any previous decentralized finance.

John Furrier

>> And your view on apps, and you got the apps, is Robinhood the next AI native-like transformation? They're transforming very fast.

Bill Platt

>> Yes.

John Furrier

>> You're seeing Stripe come out with their own chain. I mean, everyone's got payment rails, settlement rails are coming.

Bill Platt

>> Yes. So I do think that there's a great set of innovations occurring at the payment level right now. If you've heard of x402, this is a really big step forward. People like Google supporting agentic payment systems. This is a very big step forward at the top of this stack, allowing for application to be built, which allow us to, like I said, to build autonomous financial services with humans attached, which creates a great financial community of the future. And so that's why we're building all of this to make sure that that actually can happen, and we're supporting some of the best in the industry in order to do it.

Dave Vellante

>> And your point about stablecoins in the killer app is because of liquidity, is that right?

Bill Platt

>> Definitely. Liquidity and allowing the traditional folks to feel more comfortable. All the regulations, GENIUS Act, all these things are helping create-

John Furrier

>> Yeah. For sure.

Bill Platt

>> more or less create the environment, the right climate so that we can grow.

John Furrier

>> Indeed. What about Venmo? It's just another digital payment mechanism on your app. If you're an app provider, it's just like, "Oh, I got custody issues. They go away under the covers. I'm just buying with my crypto."

Bill Platt

>> Exactly.

John Furrier

>> And anywhere in the world, by the way,

Bill Platt

>> Exactly.

John Furrier

>> US dollar backed.

Bill Platt

>> It's actually-

John Furrier

>> Good for the US.

Bill Platt

>> really amazing, right?

Dave Vellante

>> And it's essentially any asset that you can tokenize.

Bill Platt

>> Any asset. I mean, BlackRock wants to tokenize everything, right?

Dave Vellante

>> Sure.

John Furrier

>> Well, Bill, they're lucky to have you with all your experience. Thanks for coming on. You're certainly a trailblazer on the cloud, now crypto. Thanks for joining the program.

Bill Platt

>> Thank you very much.

John Furrier

>> I'm John Furrier, with Dave Vellante. We are here at theCUBE's NYSE Wired studio, part of the programming community. We're building, of course, theCUBE and the NYSE coming together to bring all this great programming to you. Thanks for watching.