[INTRODUCTION]
Rob Hayes (00:00:02) - These are very complex things to go model. To simulate multiple molecules and how the electrons interact with each other and things like that actually takes a lot of compute power. And so this would be a marquee example of what you could do with a quantum computer is like cut the drug design, you know, like cycle down from, you know, months to maybe weeks or days.
Our technology is basically a vacuum chamber. So you see something about the size of a grapefruit. That's a stainless steel sphere. It's got some very clear windows into it, and we shine lasers into that sphere, and we trap our atoms in free space using what's called optical tweezers. There's probably, I don't know, 20, 30, 40 companies around the world that are investing in this now. I don't think you'll have that many winners at the end of the day. I don't think it'll be winner takes all. But as any industry matures, you always kind of consolidate down to 3 to 5 major players. And I think that's where we're all sort of jockeying for position to make sure that we position ourselves to be one of those winners at the end of the day.
[INTERVIEW]
Dave Tabor (00:01:11) - Welcome to the Stubbornly Young podcast for people in their 50s, 60s and beyond who want to remain engaged in the world and relevant to the younger people in their lives. I'm Dave Tabor, and that last part, staying relevant to the younger people in their lives, is why I've invited Rob Hayes to join me for episode ten. Rob is CEO and president of Atom Computing. There has been a ton of talk about quantum computing, the next milestone in computing power, the next quantum leap, right? Pun intended. You may have read about how quantum computing will propel artificial intelligence and enable the solving of problems even more complex than we've dared to approach with today's computing power, which is a lot. There's also fear that quantum computing can smash the current computer security measures we have in place now. I've been intrigued, and in preparation for this episode, I watched three videos that promoted themselves as a definitive source to explain and explain so I could understand what quantum computing is. And they all failed. I've asked Rob to join me on Stubbornly Young for my benefit and yours.
Dave Tabor (00:02:21) - And while we probably won't be able to explain how quantum computing works will be up to speed on its promise, maybe even its place in the future. And that's the idea that I and you will be able to discuss this with the younger people in our lives. So, Rob, happy to have you here on Stubbornly Young.
Rob Hayes (00:02:40) - Oh, wonderful. Thanks for having me.
Dave Tabor (00:02:42) - Yeah. And I hope you're going to accept this challenge of explaining it. But before you do that, and I'm sure you've been asked many times before you do that, how about a brief overview of Atom Computing and actually the problem that you're solving.
Rob Hayes (00:02:55) - Yeah. So Atom Computing is building quantum computing hardware platforms out of optically trapped neutral atoms. In short, we're building the next generation of computing platforms, out of technologies that really have been explored in academia for the last couple of decades. And we're one of the first companies to kind of bring them out as commercial platforms. And the reason we do this is to enable researchers and companies to develop applications and solve problems they just really haven't been able to do prior.
Dave Tabor (00:03:26) - All right. Well, that's a good overview. Now, you said optically trapped neutral atoms. I did not read about those when I read about bits versus qubits, subatomic particles and how things aren't ones or zeros. But what are you talking about?
Rob Hayes (00:03:40) - Yeah. So when we talk about quantum computing, we talk about what is kind of a once in a generation paradigm shift in computing performance. It's enabled by these new kind of building blocks that you just mentioned. Qubits. Qubits are different than bits. Bits, you know, classical computers have transistors. Transistors are a little switches. They're either on or off or one or they're zero. And to make a number or encode information, you just have to string a bunch of bits together into bytes and megabytes and gigabytes and that kind of stuff. Right? And qubits have this nice properties that they can actually hold a lot more information just one bit, one on or off kind of signal. They are represented by Bloch spheres and other, you know, three dimensional shapes.
Rob Hayes (00:04:23) - And you can encode much more information in a small number of qubits relative to what you can get in a small number of bits. And this allows quantum computers to run computational problems and explore a much larger solution kind of space, a much quicker, it’s fewer time steps, ultimately. So think of it as more parallelism, more range of data to be explored in a smaller amount of time. And that's where the power comes from.
Dave Tabor (00:04:48) - I get that you understand it. I'm glad you do. You know, it's okay for you to say no. A layperson cannot understand this. And like, I think about, you know, like movies about going to Mars or I watch stuff about like how, like, I can kind of understand the concept around like a slingshot and capturing an orbit and stuff like that. But no, I don't get anything that you just said. Is it impossible, seriously, for, is it okay? Should we just accept that this is true and it works?
Rob Hayes (00:05:17) - I don't think we should accept it, but I honestly think that's where we are today.
Rob Hayes (00:05:21) - Like I, I wish I had this nice canned 32nd answer on how does a quantum computer work and how it provides this revolutionary performance relative to a classical system. And it's a very hard thing to say in a way that people who really just haven't spent much time in it can grasp it in a short amount of time.
Dave Tabor (00:05:41) - And you know what? That's okay. I kind of entered this conversation thinking that that was the answer, that there are some things that you just have to be trained to understand, and this is probably one of them.
Rob Hayes (00:05:51) - I think that's fair.
Dave Tabor (00:05:52) - And that's okay. And I kind of thought this was where we'd end up and I'm okay with it. And so there are some challenges that you can solve. I read, too, about this notion of a traveling salesman concept as like a metaphor, the idea that, you know, if you're going to plot a route of a salesperson who has to hit ten or 12 or 20 locations within a period of time that a computer can easily route the optimal. But if those number of variables become hundreds or thousands or millions, right, then the way you have to solve it changes. Is that right?
Rob Hayes (00:06:23) - Yeah. That's right. And I think there's some very practical examples of this problem that exists in the real world in an everyday basis. So I like to think of like a, you know, the post office or a parcel delivery company where you have, you know, millions of packages going into the system from, you know, hundreds of thousands of different source addresses, and they're going out to hundreds of thousands of destination addresses. And the amount of paths that could take isn't infinite, but it's it's there's many paths. And the way people design those systems today is humans sit down and design loops that trucks run through neighborhoods to pick stuff up, take them to warehouses to transfer them to the next truck or the plane or the train to get to the next destination. And then they get delivered on loops again. And that actually works. It works every day, but it's not optimal, right? If you knew what packages are coming in, where they're coming from, where they were going, and you had a computer powerful enough to actually map all the different potential paths of all these different, you know, millions of packages on daily basis.
Rob Hayes (00:07:19) - And think of it as like print out a custom ticket like Google Maps does when you're trying to drive your car from one destination to another. Uber. If you could do that, you could save fuel, you could save time. You could save, you know, human labor. And that would be real money saved for the economy and rolled out. It's just not practical to do that today.
Dave Tabor (00:07:36) - I'm going to ask you. You're not going to give me an answer. We're going to ask you anyway. Like in that example, if you could fantasize that quantum computing were solving this on a daily basis or almost a real time basis, do you think that we'd save 10% of the fuel, 10% of the man hours? Are you looking like 50% or 5%? In fact, 5% is so valuable that it's worth it.
Rob Hayes (00:07:58) - That's that's my answer is, if 5% is so valuable, which is 5% of the fuel costs for UPS on an annual basis. I don't know what that is, but it's a lot. Right?
Dave Tabor (00:08:07) - Yeah. Yeah.
Rob Hayes (00:08:08) - Now could we do better? I hope we can do better, 20%. 30%, we don't know, right, until we do it. But that's the nice thing about optimization problems. If you just come up with a more optimal solution and the value or the cost however you want to look at it, if the solution space is large enough, then people will implement it.
Dave Tabor (00:08:24) - That makes sense. And I've also thought about, like, I've seen other examples in drug discovery. Talk about that.
Rob Hayes (00:08:30) - Yeah. So in general there's like two kind of major classes of problems that quantum computers could solve. The first one you mentioned is an optimization problem and network optimization like the traveling salesperson problem is a good example of that. The other class of systems are quantum systems sort of natively. Right. So when we talk about quantum we're talking about atomic physics and subatomic physics. Right? And when you start talking about pharmaceuticals, they're molecules, they're, they are quantum, you know, the way molecules interact with each other, interact with different diseases and the environment and things like that. That is a quantum problem. And so using a quantum computer to save what's to solve what's natively a quantum problem makes a lot of sense. Now, the reality is that these are very complex things to go model to simulate multiple molecules and how the electrons interact with each other and things like that is actually takes a lot of compute power. And so this would be a marquee example of what you could do with a quantum computer is like cut the drug design, you know, like cycle down from, you know, months to maybe weeks or days, you know, potentially that would have a tremendous value on society and economic value for the pharmaceutical industry and so forth. But we need much larger systems and more capable systems to actually be able to do that. So today, people are sort of learning and using what we call toy models to figure out how to do that. So when the systems get big enough in the future, they'll be able to deliver that, that, you know, economic value that everybody wants.
Rob Hayes (00:09:57) - But that's still out there a little bit in time. But I think the value prop is clear. And I also think that maybe some of the optimization problems, for the reason you said, like if you could just save 5%, it's real value. Those are going to be a little bit more near-term and probably easier for us to implement, you know, in the next year to five year kind of horizon. And the molecule simulation and chemistry and pharmaceutical kind of use cases are probably more out in the latter part of that horizon and beyond.
Dave Tabor (00:10:24) - Got it. Now, even without quantum computing, cloud based, like ChatGPT, for example, gives me answers to complex prompts in seconds. So what's the difference between the complex problem solved with quantum computing and what kind of current computers can do now?
Rob Hayes (00:10:42) - I think there's probably different ways we can look at this, is it's a common question is how does AI and quantum computing relate? And I think there's different answers. On one hand, you could probably make an argument they're competing, right? They're competing for mindshare and just resources of, you know, developing and moving them forward and apply applications based on them. I think that's true. You could also look at and say maybe they're competing because maybe I will get so good. It'll be enhanced by AI in and of itself that it will close the gap between classical computing, you know, that's using AI and the promise of quantum computing. So maybe that pushes out the value proposition or diminishes the value proposition quantum in some cases. I don't know if that's true or not. Time will tell. But I also believe they're complimentary. And if you think about like an AI model, AI model, except that you generally an enormous set of data, like all the text on the Earth, or, you know, all the photos or a big sampling of photos on the internet or something like that, and it trains a model, and you're training a model to try to infer, what am I looking at? Am I looking at cats or dogs in photos? Am I hearing Chinese or English? And what are they saying? And what is the meaning of the saying and what is the appropriate response and those kinds of things.
Rob Hayes (00:11:55) - And when you're inferring you're almost by definition trying to come up with the most probable answer about what someone's looking for, right. And that's actually where quantum computing can come in enhanced because unlike, you know, binary classical computing where it's always sort of like integer math, it's a finite answer. Quantum computing is exploring along a large range of solutions and trying to come up with what's the most probable answer based on that computation. And so I think we can easily imagine a world where there's quantum computing as part of the workflow for an AI model to help the inference gain more accuracy or more precision.
Dave Tabor (00:12:34) - So if quantum computing is predicting the most probable answer, if you will, does that mean it's wrong sometimes?
Rob Hayes (00:12:42) - Well, I mean yes and no. Right? It's wrong sometimes in that there's a lot of noise in these systems today and they're error prone. I think that's a little different than the question you asked. But yeah, I mean, at the end of the day, you can almost think of quantum computing as analog computing where you're trying to, you know, find the signal through the noise, if you will.
Rob Hayes (00:13:00) - Yeah. It might not always give you the right answer, but there's other ways you can triangulate and try to verify. Did I get the right answer? You can run things multiple times. You can run them multiple ways. You can triangulate and make sure that I'm getting to the right answer. But yeah, you don't always get the right answer.
Dave Tabor (00:13:15) - I see, but no, what you just explained, though, Rob, is interesting because even though maybe in a particular instance, you get an answer that isn't as the ultimate precision on point, that quantum computing also lets you explore multiple answers and to your point, triangulate, figure out very quickly and through lots and lots of variables to figure out what should the right answer be in a way that traditional computing couldn't.
Rob Hayes (00:13:39) - Yeah, yeah, we're ten minutes in. You're already getting it.
Dave Tabor (00:13:43) - I'm getting what it does, but not how I'm going to give up on that part. But thank you for that compliment, Rob. That means a lot. There's also a dark side of quantum computing. I mean, people worry about things like security hacking and stuff, right?
Rob Hayes (00:13:58) - Yeah. I think the thing that probably gets the most attention, at least in the also say in the popular press, is this idea that quantum computers are going to, you know, break RSA encryption and all of our data will be out in the clear at some point in time. And it's actually kind of true, right? I mean, if you look at the way these algorithms were created, they were basically they took very, very large numbers and then they basically build encryption schemes around the prime factors of those numbers. And it's very hard to figure out what the prime factors are of a very large number. So unless you have the keys, it's almost impossible or would take an inordinate amount of compute time, you know, to break these codes that we use today for encryption. Now, this is a well known problem. The National Institute of Standards and Technologies has been working for a few years on what's called post-quantum encryption standards, so that we can upgrade all of our networks and computers and everything well ahead of the time when quantum computers will be able to do this prime factoring of large keys and make this a non-issue. So. Well, it gets a lot of attention. I actually believe it's going to be like Y2K, where by the time we get there, it's going to be a non-issue because it will have already been solved.
Dave Tabor (00:15:07) - Well, that's good to think. I mean, it is nice and almost refreshing that what you've just painted is a picture of us actually getting out ahead of a problem. That's kind of nice.
Rob Hayes (00:15:16) - Exactly.
Dave Tabor (00:15:17) - There does seem to be a race in the world of quantum computing for practical deployment. Right? So is it really a race and is there a lot at stake here? There must be.
Rob Hayes (00:15:28) - I look at it as a race, just like any competitive race, electric vehicles or standard computing or, you know, competition and industry and, and government is is always a race and whoever can maintain the pace of innovation, the fastest and the most economical, generally, you know, wins their unfair share of the rewards of that, of that race. Right? So that's how I look at it.
Rob Hayes (00:15:54) - you know, I worked at Intel for 20 years. I worked at Lenovo. So I've been in the computing business for a long time, and we were reinventing our product line every single year. We would come out with a next generation of CPUs or, you know, servers or PCs every year. They would always be 20 to 50% better than they were the year before. It's about the same price. Right? So that's an economic value. And if any company got off the train on, you know, that evolution and that improvement cycle, then they would stick on competitive and you're relevant in the market. And so I think quantum will be the same. I don't think it's unusual in that we're all racing to get more qubits to get, you know, lower error rates to get faster speeds, like there's lots of performance, you know, kind of elements of building a quantum computer that are very similar to building any, any system. But yeah, you've got to keep up a pace of innovation in order to win that race.
Dave Tabor (00:16:51) - Yeah. Are all the quantum computing companies basically using the same theory and then racing to figure it out, or are there different approaches to what we're calling quantum computing?
Rob Hayes (00:17:03) - There's a lot of different approaches and that comes on different levels. In one level there's an approach called an analog quantum computer. Then there's one called a universal gate based quantum computer. So some companies are going after one approach somewhere after the other. The difference there is how you program them and what are the appropriate applications for each of those types of systems? We're going after universal gate based systems. And even within that, there's many different architectures and technologies under the hood that people are pursuing. The original technologies that companies were pursuing were superconductors. They're basically building chips and they're building qubits on these chips, and they're manipulating the electron, you know, states and energy levels and things like that within these chips to build qubits. There's ion traps, there's photonics, we're doing neutral atoms. There's, you know, there's probably 5 or 6 others that are probably not as mature but are in various stages of research and development at this point in time. And the winning horse isn't clear. There's advantages and disadvantages of each and there's reasons why each company might pursue one versus the other. And that's kind of where the state is at the moment.
Dave Tabor (00:18:09) - Wow. So I mean you must, this has to be incredibly expensive. I mean, every horse is incredibly expensive, right? I mean, the science, the people you have to hire. I mean, that's just brutal, right? What's that race look like?
Rob Hayes (00:18:23) - Yeah, there's a lot of capital that's been coming into this industry for the last, I would say probably five years in earliest. The research has been going on long before that, but there weren't as many players in it. Prior to five years ago, a lot of venture capital money, a lot of, you know, large company. There's large companies, you know, investing here Microsoft, Intel, IBM, Google and others. So there's a lot of money coming in from them. There's also a lot of government money coming in from the United States and internationally.
Rob Hayes (00:18:50) - My company has a DARPA program that we've been executing, too. We've got some NSF funded projects. So the government funding helps a lot. But it's, you know, it's similar to any deep tech early stage thing, whether it's space or semiconductors or, you know, energy and things like that. There's always government support, academic support and, you know, corporate and venture capital in order to get it going. Eventually there won't, you know, there's probably, I don't know, 20, 30, 40 companies around the world that are investing in this now. I don't think you'll have that many winners at the end of the day. I don't think it'll be the winner take all. But as any industry matures, you always kind of consolidate down to 3 to 5 major players. And yeah, I think that's where we're all sort of jockeying for position to make sure that we position ourselves to be one of those winners at the end of the day.
Dave Tabor (00:19:38) - Yeah. So how much funding have you received?
Rob Hayes (00:19:41) - My company has received about $80 million in venture capital funding. And then we've, you know, on top of that, we've gotten revenue from the government in the form of grants and other projects that we've executed with them. So in total, it's, you know, approaching $100 million. And we'll need, you know, more to keep going, you know, over time until we reach profitability, just like any deep tech venture capital company.
Dave Tabor (00:20:03) - Yeah, deep tech is a good term for it. And, my question is this, investors who collectively, now Microsoft and Google they probably understand but venture capitalists, do they understand, like the questions I was asking you that I couldn't understand the answers to did they understand the answers?
Rob Hayes (00:20:21) - That's an interesting question. I would say it's a mix. Some absolutely not. Some try and some do. We've been fortunate that the investors that we have in our company, they all had done prior research and decided, number one, they wanted to invest in quantum computing. Number two, they knew what they were looking for in a quantum computing company. Number three, they were just looking for the right fit. So we didn't have to do a lot of education. It was more like a matchmaking exercise for us. I suspect that's probably true for many of our competitors right now. But as we grow up, you know, collectively as a group we move out of what they call the venture phase and into growth and, you know, eventually, you know, profitability and scale phases. The nature of the investors will change. They'll become more institutional investors and traditional banks and things like that that don't have, you know, the technical staff and chops to know exactly what we're doing. And they're going to look more like, well, you know, what is your revenue growth rate? What is your value? What is your profitability? You know, line of sight and all those more traditional financial metrics.
Dave Tabor (00:21:23) - Sure. So what does a quantum computer look like right now and how far are you away, do you think, from actually selling one?
Rob Hayes (00:21:31) - Well, physically, when most people think of a quantum computer, if you Google it, you're going to get some people referred to as the golden chandelier. It's actually a beautiful thing. IBM's done some really nice work in putting out artwork and photographs of it. It's something that kind of hangs down from, I call it the ceiling, but from above there's lots of gold wires connecting up the chips for the qubits. And then they put that into a dilution refrigerator, which, it looks like kind of a barrel or something that surrounds it eventually. If you came into our lab, you wouldn't see that. It wouldn't look anything like that. It looks completely different. Our technology is basically a vacuum chamber, so you'd see something about the size of a grapefruit. That's a stainless steel sphere. It's got some very clear windows into it, and we shine lasers into that sphere, and we trap our atoms in free space using what's called optical tweezers. So we just shine a bunch of spots of light into this vacuum chamber. There's a gas of atoms in there of a certain element, alkaline earth elements. And those atoms get individually attracted to the spots of light, where they're kind of the focal point is most intense.
Rob Hayes (00:22:41) - And then we can manipulate their quantum states, in our case, nuclear spin, qubits and their electron energy states with different colors of light or different pulses of light. This is all stuff that's been studied for the last couple decades. There's been Nobel Prizes won on the fundamental building blocks here. And we're, you know, one of the first companies to kind of commercialize it as a quantum computing platform. So what you see in our lab is that sphere, that vacuum chamber with a bunch of optical devices and lasers. it's very colorful, lots of different colors flashing. And, it's quite small, actually.
Dave Tabor (00:23:14) - And how much more powerful is this little metal sphere? Your grapefruit with windows and colors. How much more powerful is that then, you know, a typical desktop computer or even a supercomputer.
Rob Hayes (00:23:30) - Well, it's hard to compare them because they all do different things. The current system we have today is a prototype system. It's 100 qubits, you know, that's on the order of the size of any of the largest ones that are out there.
Rob Hayes (00:23:41) - There are some systems that have come out since that are a few hundred qubits. We're working on larger scale systems today. Physically the same size, we just add more qubits. One of the interesting things about our technology is that we can scale up the number of qubits in a single vacuum cell, up to millions of qubits over time. But we won't do that tomorrow. But we’re, we have a roadmap to get there. And many of the other technologies require modularity, where they can only fit so many qubits into one module. And then they have to connect multiple modules together. And that requires some technical breakthroughs that are still being worked on in order to bridge over between modules with these interconnects, usually using photons. So that's kind of one of the advantages we have, when we get hundreds of thousands or millions of qubits, we will transition into what we call the fault tolerant era for quantum computing. This is when utility scale quantum computing is going to really come into its heyday, and we'll be able to address, you know, some of these use cases that we were talking about earlier at commercial scale, which I think people are really excited about. When we talk about supercomputers, you should think about these large scale quantum computers as being something akin to a supercomputer, but in reality, supercomputers or classical HPC clusters, and quantum computers will sit together and they'll work together on workflows. There'll be some things in a workflow you're going to want to do on a classical cluster. There's going to be some things you’re going to want to do in a quantum system, and the software will stitch all that together into a workflow that makes sense for the researcher or the developer.
Dave Tabor (00:25:14) - Yeah, that makes sense. Now, when do you think we'll see these other than in theoretical or test states?
Rob Hayes (00:25:21) - Well, the systems exist today. There are quantum computing platforms available today. You can go, you know, on the public cloud and get access to systems, not ours yet. But once from a competitor, they're relatively small scale. They're more like research and development platforms than they are like, you know, commercially viable, you know, application platforms.
Rob Hayes (00:25:40) - But we're already starting to see, HPC supercomputing centers around the world install quantum platforms and do that stitching together at the software level so that researchers can figure out how to build those workflows. We have a project that we recently announced with the National Renewable Energy Laboratory in Colorado and their Flatirons campus, where we're doing that, where we're stitching our quantum computer in with their what they call Aries platform, which is an HPC cluster they have where they're simulating the US energy grid. So these projects are already happening, but they really are research and development right now. They're not they're not really at the level where they're kind of like production quality. So we're working towards that, though, of course.
Dave Tabor (00:26:21) - When do you think that'll be.
Rob Hayes (00:26:23) - Oh, I think we're within the five year horizon. Yeah.
Dave Tabor (00:26:26) - Wow.
Rob Hayes (00:26:26) - Moving out of, like, this research and development phase where we're mostly working with scientists and, you know, academic-type folks and more working with enterprise practitioners and, you know, government application researchers.
Dave Tabor (00:26:41) - So are you going to like, sell yours to, to like, cloud based providers like Azure or, you know, Amazon or whatever? Or are you designing to sell a, you know, Atom Quantum computer?
Rob Hayes (00:26:54) - Yeah, I think our company is geared up to do both. I think the primary route to market for all compute services, you say is, is through the cloud. It's easier for people to gain access and pay as they go than it is to buy a big infrastructure, but if someone wants to buy a system, we'd be happy to sell them a copy of one. So I think our going to market strategy is both. But I would say cloud first.
Dave Tabor (00:27:16) - Why do you say a copy of one?
Rob Hayes (00:27:18) - Oh, a copy. Like, if I build one in my cloud, I can build a copy of it and sell it to someone to put into their infrastructure as well.
Dave Tabor (00:27:24) - Got it. All right. Understood. I didn't know if there was something, some other mysterious use of that term that I wasn't aware.
Rob Hayes (00:27:30) - No, I'm not going to see you in a second or third or fourth month yet.
Dave Tabor (00:27:33) - Got it. All right. Last question to kind of wrap things up here. If you put on, Robby, you put on this futurist hat. What are a couple of specific examples of maybe how the world will look different in, say, 25 years based on quantum computer?
Rob Hayes (00:27:52) - Yeah, I think I'm just going back to the use cases we talked about earlier, just using the two examples. Right? So I think it's really within, the realm of possibility that we, you know, transportation logistics, supply chain, you know, these kinds of companies that are running complex networks, they're in real time adapting their networks to get whatever it is they're delivering, whether it's bits and bytes or electricity or goods from point A to point B in a much more efficient way, using much less energy, much less time and human resources in order to get that done. I think that's going to deliver tremendous value, economic value to industry and society. And then on our health care with like pharmaceuticals, that's a huge issue is, you know, being able to cut down the amount of time it takes to go from, identifying a problem, when we want to go address a disease or a or whatever, or we want to create a vaccine or something like that to actually delivering that. If we can cut that down from, you know, years and years and billions of dollars to something more like weeks and months and something much less than the capital outlay, then we will be able to tackle many more diseases much more quickly. And it will have a benefit to society as well.
Dave Tabor (00:29:03) - That's kind of interesting. You know, you think about a pandemic or something. And instead of taking if scientists could work through the theoretical solutions in minutes versus, you know, spending months in a lab, right. Is that kind of what we're going to see?
Rob Hayes (00:29:16) - Yeah. I mean, I think that would be the panacea is like if we could go and simulate many, many more molecules against many, many more diseases and things like that in a much shorter period of time. You can imagine a library of opportunities that we can just pull off the shelf and go and create drugs, and we'll start to do clinical trials, and all those things will take time and investment. But if we can cut that discovery and design cycle ahead of clinical trials down significantly, then we can address much more.
Rob Hayes (00:29:42) - And you also have to remember, like when we did the vaccine, we don't, you know, we, collectively not that I did it, but for Covid that was very quick. But that was a priority call. Right? There had already been some prior research. There were some things on the shelf, but, you know, the industry collectively gave up on some other focus areas so they could focus on what was really important. But these other things suffer. But if you could imagine if we could do this much more fast, more broadly and cheaper, maybe that opportunity cost gets lower and we can actually solve more problems simultaneously.
Dave Tabor (00:30:13) - Yeah. That's cool. So what do you think is the hardest part of getting from where you are to that panacea?
Rob Hayes (00:30:19) - I think we need to continue to evolve quantum information science at all levels of the stack simultaneously. So, you know, we at Atom Computing will continue to do our job to make the hardware bigger, faster, better and scale it up. But we need to have partnerships from government and industry on how to program these things. What algorithms will work best? How do you map those to the hardware? How do you get the results? All those kinds of things. So it's really going to be a collaboration application software and hardware, us and partners all working together in order to make that happen.
Dave Tabor (00:30:55) - Is your hardware going to be, I don't know, if open sources is the right term? Maybe it is, but I mean, will anybody be able to figure out how to build applications to run on your hardware?
Rob Hayes (00:31:07) - Yeah. And I think actually open source is probably a good way of thinking about that. The hardware itself will be proprietary because it's our, you know, we're building it and how do we do it? And everyone else will be building it in a slightly different way. But the industry's already sort of normalized on some common programming APIs and standards. IBM was out the gate first in creating Qiskit, which is a very popular programming language and SDK.
Rob Hayes (00:31:28) - And so we support that and as do many of our competitors. So if you write a program for their platform or somebody else's, you could port it to ours quite easily. There's also open source APIs like Open Chasm and a few others that allow us to port between hardware platforms with very little effort. And I think that's important because nobody wants to get locked into any vendor. They want to have seamless or frictionless movement, be able to try all platforms and see what performs best at the, you know, highest priority and the lowest cost for them. And so we support that.
Dave Tabor (00:31:59) - I think let's end on that note. I, I'm going to challenge listeners, if you can find some videos that actually help you understand the atomic level functioning of quantum computing, share those with me and I'll share them with listeners in exchange. In the meantime, I think, Rob, you've done a really good job of helping us understand. At least where this is going, the applications that we can look forward to seeing.
Dave Tabor (00:32:22) - So on that note, I'm your host, Dave Tabor, today on the Stubbornly Young Podcast. You've been listening to my conversation with Rob Hayes of Atom Computing. Rob, glad you could join me.
Speaker 4 (00:32:34) - Yeah. My pleasure. Thanks for having me.
Dave Tabor (00:32:37) - Listeners, this has been episode ten of the Stubbornly Young Podcast for those in their 50s, 60s and beyond, remaining engaged in the world and relevant to the younger people in their lives. Please do me a favor. Help the podcast spread by submitting a review and by sharing with your Stubbornly Young friends. Catch you next time on Stubbornly Young.
[END OF TRANSCRIPT]
