A decade of quantum on the cloud

Welcome to the second season of " Coherence Times," where we bring coherence to the tangled world of quantum computing . This is your host , Ryan Mandelbound. Every two weeks I will bring you stories about how scientists, developers, and companies are turning quantum computing into a tangible reality. We will explore the latest research and developments, highlight new advances in the field, and see how users are trying to expand the uses of quantum computing to include practical, real-world applications. This year marks a milestone in the history of quantum computing. On May 4, 2016, IBM launched the IBM Quantum Experiment. For the first time, anyone can conduct their own experiments in quantum computing on a real quantum computer accessible via cloud computing. This event sparked a decade of exploration, and today there are hundreds of thousands of users who have performed trillions of operations on IBM's cloud quantum computers. Today I am pleased to welcome two guests who have been present almost from the beginning . The first one is Jerry Chow from IBM. Jerry worked at IBM for 15 years and was part of the team that put the first quantum computer into cloud computing. Today he holds the position of Chief Technology Officer for quantum supercomputing, and is an IBM Fellow working on the large-scale development of quantum computers . I am also pleased to share with you Travis Hempel, director of the Quantum Science Center at Okiage National Laboratory. Travis has been involved with these systems since their inception. His 2017 research paper with Keith Brett was among the first to use IBM's quantum experiment. Today, Okedge scientists are conducting stunning molecular simulations on IBM quantum systems. I thought it would be interesting to go back to the beginnings of quantum computing on cloud computing, while being aware of how much progress we have made since then . So, maybe we'll start with you, Travis. What do you think about taking us back to 2015? How did you feel while conducting a quantum computing experiment back then? Well, in 2015, the situation was very different from the experiences of many today. Access to quantum computers was extremely limited compared to the current situation. In fact, quantum computers were the preserve of laboratories that invested in building their own quantum computers. If you were lucky, you could collaborate with one of these labs and get a chance to try new things. But even those experiments at that time were fraught with risks , and focused mainly on developing the latest technologies in the field of coherence and accuracy, and even the invention of pixel units. Therefore, much of the work has focused on theory, modeling, and simulation. This is what I spent most of my time doing, trying to think about what it would look like to access a quantum computer, and how to run programs on these systems. I have spent a lot of time thinking about how to scale up quantum computers, and how we will eventually reach sizes capable of handling extremely difficult problems. But our understanding back then of what structures should look like , and what applications should be like, was far less mature than it is today. So I would say that at that time , in 2015, much of our effort was very exploratory, and we were not yet ready to move into applications and implement them; it was more like wishful thinking about the types of software we would like to write. Absolutely true. And Jerry, for your part, how did it feel to be responsible for owning a quantum computer and allowing others to use it? Have you worked together before ? And how did you make it accessible ? Yes. No, I mean, at that time, I think it was an area for physics laboratory experiments that we were conducting in these various laboratories. As Travis pointed out , it was really about conducting experiments on these devices. It was about who you knew and the type of academic collaborators you had, wasn't it? I mean, this goes back to my days before I joined IBM at Yale University, where we were working with theoretical scientists who wanted to conduct experiments on a device the size of a cube or two, right? Even at IBM, when we started developing these devices, and while we were improving them and participating in conferences to present our results, we were getting questions like: "Can we perform X, Y, and Z on this device?" But we were always in constant motion , weren't we? We were always trying to do something new. We didn't have enough time to sit down and plan together to develop a research plan, submit a proposal, and obtain funding to conduct this experiment. Therefore, we often had to refuse many requests, didn't we? We simply couldn't go in and try out their ideas. Well, that was a completely different time, wasn't it? The whole thing was in the hands of those who had actual control over the equipment in the quantum physics laboratories. completely . And then, Travis, there was no interface for these computers either, was there? So, if you wanted to conduct an experiment, you had to write everything down manually and tell the hardware development team exactly what you wanted to do, right? What was the situation like? I think the gap between theory and experiment was very deep at that time, partly due to the circumstances described by Jerry, where I needed to invest in experimental development, and that was a full-time job. The idea of ​​being able to support programming, evaluation, and applications simultaneously , as you know, did not yet exist. So, in theory, developing algorithms and applications, you were wondering, you know, what would this future design look like? Regarding these computing platforms, there was much less talk about hybrid computing, as well as about integration. The focus was on the idea that quantum computers, as a computational model, would allow us to test new algorithmic principles and solve problems in innovative ways. And of course, a sudden and significant change occurred, which proved to be enormous. But at that time , this division was very clear, and the demands for progress that we were imposing on ourselves did not encourage this kind of cooperation. Jerry, can you tell me a little about the debate surrounding putting a quantum computer on the cloud and starting to provide access to it? Well, I think it was part of one of our sessions ... I didn't know that at the time, but it seems there are cycles of being able to suggest new ideas and projects. Thus, Jay and Mita, at the then IBM Research Spring Strategy Meeting, floated the idea of trying to make a quantum computer accessible via cloud computing using the appropriate application programming interface (API) . We wanted to obtain approval for the project plan and its funding. We thought about it from an amazing perspective, as we were constantly hearing from everyone that our devices were extremely stable, and looked very good in terms of coherence times. We knew it was stable, so we decided to find a way to make it available to people. And so the project began internally. Then how did it go from being just a project plan to something that was actually implemented? We then had to draw up a detailed plan of the project's layout. It is truly fascinating that initially we were aiming for something too big to be published. I think in the first version of the schematics I was trying to add a 13-bit or 17-bit processor, right? But, as you know, during this design process , and what could be built and what was possible, both in terms of hardware and software, we gradually narrowed down the scope of the project . Then we reached a stage where we were able to determine that this was a genuine MVP product that we could develop and deploy with confidence. And we started implementing it, didn't we ? We had a team that focused on the hardware side , another team that focused on software implementation and programming, as well as a user interface development team, responsible for how the product is presented to users. All these efforts had to be combined to complete and launch this project in a very short time . Tell me if I'm right, but if memory serves, there was an IBM conference in 2015 where someone went up on stage and asked : "If we had a cloud- based quantum computer , would you use it?" Can you tell me more about that? Yes. Yes. It was at our conference at the time, which was called "Think Q," and Professor Ike Truang from MIAT had gathered the attendees and asked them if they were interested in using such a device. We had certainly made great strides in its development by then, hadn't we ? So, it was really exciting to see this huge endorsement, with so many people raising their hands to express their desire to try, explore and use a capability like cloud quantum computing . Travis, would you have raised your hand then? Oh, definitely. I think I would have been one of the first to do that because this type of opportunity was very new, wasn't it? It was something we had never seen before. So, this was innovative and exciting . amazing. Now, let's go back to launch day. So, what were you thinking, Jerry? As you know, you are about to release this thing to the world. How did you feel ? Well, the launch day was May 4, 2016. I remember we turned it on very soon, around midnight, and we started to see some of these jobs coming in. That was around the same time as the release of Crest, wasn't it? So, articles about him also began to appear . We knew the platform was working, did n't we? Because we conducted a media tour beforehand to showcase the demo version, and to demonstrate everything. I worked there, you know, on all sorts of tasks that we had . But when I started to see the real work begin to flow , it was a little funny . Jay and I were monitoring the control unit, making sure that everything was working and that everything was going well. I think we stayed up for hours after midnight to make sure everything was working properly. But, in the few days following the launch, we started to see a lot of people using it, talking about it, and tweeting about it, and that was very exciting. Then Travis, when did you first realize that this thing had become available, and what was your first reaction? I think, as Jerry said, that was important news for the community. I don't think we all realized at the time how important this radical paradigm shift was . But within about a year, people began talking about using IBM's quantum computing systems for testing, evaluation, measurement, and demonstration. I mean, by everyday standards, these might seem like tiny chips to most people. But the truth is, it was the only thing that needed to be obtained. This also contributed to understanding the practical aspects of trying to program a quantum computer, and building a bridge over that gap we talked about earlier. I truly believe that one of the most important benefits of this whole experience was that shift in the intellectual paradigm, and the unification of the community’s efforts around the question: What does it mean to own and use a quantum computer? What early work did your team do back then, and how were you working to bridge that gap using these early 5-kilobyte cloud systems? At that time, the focus was largely on understanding the language of quantum computing and overcoming the obstacles to its adoption. Of course, people were aware of the theory of quantum computing, but the rules for programming a quantum computer, the tools needed for it, and believe it or not, error correction and data validation, all of these things had to be built up while we were in the process of using these systems. Much of our early research at Okedge focused on trying to translate concepts previously used in quantum computing theory into these practical applications. I believe one of the best problems solved in the quantum computing community is the energy of the hydrogen molecule, and I believe it is available. It could be implemented on a relatively small number of central processing units, but the tools needed to manage all that specialized knowledge, chemical knowledge, as well as the quantum computing interface that had to be built, were complex. And the people who built it were not in Okedge, at least, the same people who built the hardware system. And so we found ourselves in a vicious cycle of feedback . Perhaps one of our early experiences is understanding how to provide feedback to help the next generation of devices and interfaces progress. So I think that in the early days, a lot of attention and effort was directed towards this aspect. Simply put, we documented what we accomplished, explained the challenges we faced and how we overcame them, and then passed it on to future generations to make these explorations largely repeatable. That's definitely part of it . The other part was trying to prove the validity of our position all the time. As you know, once you get hold of this quantum computer, you will do something that will make all the time and effort spent so far worthwhile. These early demonstrations played this role. It has shown that quantum computers have scientific importance, not only because of their computing capabilities, but also because they are used as tools for other types of exploratory discovery processes . So I think that was on our minds as well. Despite our great enthusiasm, we know that the long-term success of this program, this quantum computing campaign, depends on how we take advantage of this opportunity. Jerry, as you know, you have this emerging research, the researchers who are facing these challenges , and so on. In those early days, how do you respond to the community that writes this research, and how do you begin to develop the program to interact with them? Yes. So, I mean, a big part of the beginning was about developing this community: working with them, listening to their feedback, figuring out what they were using, and in many cases, even connecting with people to work with if they had certain problems that weren't working well. Much of this contributed to our design of how to program these devices. Almost a year after launching the Quantum experiment, we unveiled KissKit, an open-source software toolkit for programming. And from here we were able to obtain another set of, as you know, to begin this development journey for developers , to program quantum computers. And so you began to see, as Travis mentioned, different layers of people who were not hardware specialists, but who gained access to different parts of the ecosystem, and saw where they could best get in. Thus, we moved from a basic physics experiment to presenting it in an elegant way, through an application programming interface ( API). Now you have begun to build this rich software package and software ecosystem that should ultimately be geared towards applications. We've broadened the scope of work to include a larger number of people who can now interact with this, haven't we? For us, a big part of it was seeing this openness, listening to what the community was looking for, and starting to incorporate that into both the software and, where possible, improvements to the underlying hardware and hardware developments, in terms of, for example, a larger number of cubes, etc. Was there any research, work, or anything that someone did on the system that made you feel that it had taken off, that people were doing what you hoped, and that it was pushing the field forward? I mean, I think you know that a lot of that has always been about testing the performance of devices, determining how well they perform, and showing some things about their stability and ease of use. Therefore, those experiences have always been special to me. There are other unique experiences as well. We started to use it to study some kinds of chemistry problems, like what Travis mentioned, even in those early days when people started to study the hydrogen molecule or something like that , right? To program a variable quantum analyzer into a simple five- or sixteen-bit quantum processor, we began to see that it was possible. Good. Understood. Solving inverse quantum equations is one of the early applications of these algorithms, used to solve chemistry optimization problems , thus demonstrating the capabilities of quantum computing, isn't it? Well, then Travis, you know, pointed out to Jerry about performance measurement and early chemical experiments, and I know that most of your work back then focused on that particular aspect. But what really surprised me was how forward-looking some of the early research was, focusing on things like: " How can we integrate this technology into high-performance computing?" Even I was genuinely surprised that your team was thinking from the beginning about how to use this technology as a real solution to science problems. Can you tell me more about how you thought back then? Yes, the Okeege National Laboratory has long been home to some of the world's fastest supercomputers, and this is generally due to our forward-thinking approach and our search for tipping points in this technology. Quantum computing has always been a top priority for us, because we know it has strong theoretical reasons to have a huge impact. But the question of how to balance the capabilities of quantum computing with the latest generation of traditional high-performance computing technologies remains a hot topic even today. In fact, this is what we continue to work on. At that time, it was much more unclear what quantum computing systems would look like and how they would be integrated. I think that at that time we were relying heavily on the GPU acceleration model for building supercomputers . I clearly remember that I was advocating for a quantum processing unit (QPU)-accelerated architecture, where I would replace the graphics processing unit (GPU) with a quantum processing unit (QPU) everywhere. Regardless of infrastructure, technology, compatibility, or even cost, right? This must be the best way. I would say that over the past ten years, things have shifted from this perspective to what I consider a more balanced perspective on how quantum computers and high-performance computing work together. But it was important for us to try it at that time because it was the latest technology. As you know, this was our understanding of computing and its available limits. I believe that access to this technology was a game- changer, because it clearly showed us what the additional costs are associated with putting things together. What tools are needed for integration? And perhaps unexpectedly, what language barriers do people face when trying to do this? We are still educating ourselves in the field of high- performance computing to make sure that we understand what it means to talk about quantum computing in this context. Now, I really hope to talk about this in detail . Well, maybe I could ask Jerry this question and ask him the same question, which is, you know, at that time - I mean, today of course this is starting to take shape in society - the importance of the quantum supercomputing model, and the need to build these things together. But were you thinking about that too at that time? Or how did this topic take shape as a dialogue? I mean, it has always been a direction we have been guided by, and Okedge has been a great user, in terms of his ability to take advantage of these types of devices, and to study how this type of hybrid architecture has evolved. Let's be realistic too, at that time we were working on these devices on a limited scale, devices we were running on simple experimental problems, weren't we ? So, for me , the main motivation was how to put the devices on a path of continuous improvement and push them to a level beyond what could be achieved by traditional methods. A large part of that was the continuous updating of our hardware, as well as following a roadmap that eventually enabled us to surpass 100 kilobytes for our processors. I believe this has brought about a qualitative shift in guiding society towards expanding the horizons of what can be achieved in the field of quantum circuits, at a more advanced level . So, when we talk about society, I think it's great to have this two- part dialogue, isn't it? As you know, Travis and Jerry, your teams are trying to shape the future, but you can't do that unless there are people who have access to these systems and know how to use them. Okay, let's start with you, Jerry. Initially, these systems were offered free of charge and openly to everyone. So why was it so important that people could have open access to it? Because this is the best way to develop this type of system, isn't it? The term " quantity" is clearly a barrier for people, isn't it ? We don't want people to be immediately put off by it. But the main idea was that we looked at it from the perspective of the entire Kiskit community , and how we could encourage more people to successfully run the first "Hello World" program on a quantum computer. And if they wanted to do that, we wanted to make sure they were able to do it, and that wasn't limited to a subscription fee , but was part of their education and understanding of quantum computing. And then going deeper, we have seen tremendous growth in the use of KisKit and our community , especially with regard to some of the core capabilities we provide. That was extremely rewarding. Even today, we still offer free access to our latest devices, as they remain an excellent starting point for many users. So, Travis, you know, as you're forming this team at the Quantum Science Center , how do you encourage your team to start working on programming new quantum computers on the cloud computing model? How do you build a team around this topic and develop individual skills, etc.? correct. It is certainly a very important priority , not just for the Okedge Center, but I would say more broadly for the Office of Science in the Department of Energy. They recognized early on the value of access to quantum computers, precisely for the reasons Jerry just mentioned. You want people to be innovative. You want them to have the tools they need to test those ideas, and you want to create that community, that cycle of feedback and dialogue. Therefore , at Okedge, we have already started a quantum computing user program as part of the Leadership Computing Facility to enable people to access these computers. IBM was our first partner in this in terms of providing access to their hardware so that people could test the software. What distinguished the QUP program, as we call it, was our ability to target this user base around applications for scientific purposes. Chemistry, high-energy physics, materials science, and nuclear physics remain priorities. What was interesting at the time was that these communities were traditionally early adopters of high-performance computing technology , as they had a background in quantum mechanics and quantum physics. Therefore, the ideas of accessing these systems, overcoming language barriers, and enabling the development of tools were a ready-made set of people. Through the program, we have found that we have been able to develop a group of scientific end users capable of supporting and maintaining these ideas . I can say that they have grown along with the systems themselves, reaching 100 billion bits and more. They serve as the first line of defense in learning how to use this technology. This is a concept I've previously brought up in the podcast, but I really feel it after talking to many people: we're always talking about quantum utility, or quantum advantage, right ? Any benefit people derive from quantum computing. But it seems that since its emergence on the cloud, there has been a quantum benefit, because researchers have been able to do things they were not able to do before. I mean, and it sounds like you're saying the same thing, that these early adopters are very excited to get this technology so they can do the work that excites them. Do you feel this enthusiasm, and do you agree with this opinion? Yes. Therefore, I say that the discoveries made possible by access to this technology, the ideas, the development of methods, numerical analysis, and normative evaluation have all contributed to enriching the scientific community. As you know, nothing attracts crowds like crowds. This in turn propels people forward in a way that we were not enthusiastic about or focused on before we gained access to these systems on a large scale . I think there are subtle technical differences that explain why it is necessary to limit the benefit and advantage to matters with economic impact, etc. But I think your point is correct. The ability to run quantum computing programs in a tangible way is extremely important, and I think we are still gradually discovering its benefits. Understood. So, we're not announcing the advantage of quantum computing in this podcast episode right now, but it makes perfect sense that this excitement, acceleration, and refinement of research will happen once access to these systems becomes easier. As you know, it was clear how important it was to read research, and perhaps...as you know, IBM's job is to respond to these needs, to respond to this growing enthusiasm, more research, and so on. So, how do we jointly develop our systems in a way that respects the needs of these researchers? Yes, of course. I will address only one aspect , which is the societal aspect in terms of the value provided to society. We see that in more than 5000 research papers , don't we? Which were produced using these capabilities. But also, a large part of enabling this type of research comes from our response by putting forward better and better treatments. Therefore, a large part of this joint development lies in identifying needs in terms of readability improvements, or improvements in the accuracy of basic binary cubic gates. So, we always make sure to update our devices, right ? We added, for example, dynamic circuits, to be able to create more adaptable circuits, and to expand the range of circuit types that can be operated. So, we always listen to your feedback, don't we ? From a software perspective , the software was designed to be scalable, with a focus on both scalability and performance, right? This has been a major strategic shift for us with KisKit, especially when we launched version 1.0 a few years ago, hasn't it? We needed to make sure that it evolved from just a software development kit to showcase a 5KB processor for the first time, to ensure that it was improved to a level that would benefit Travis and others who use these processors extensively beyond 100KB, right? Well, as you know, this kind of feedback from the ecosystem, which stimulates usage and drives new discoveries, is what we're all focused on at IBM. What excites us all is precisely this scientific aspect. I mean, after ten years, we've reached this point where we're pushing the boundaries of possibility with these cloud-based quantum computers, and we're actually entering an era where we've declared the viability of quantum computers: they're capable of doing things that traditional classical simulations can't . We are approaching the point where quantum computers appear to be publishing research that shows circuits testing the limits of classical computing capabilities. Okay, Travis, how does it feel to look back and see that we've come to this point, as a member of a team that operates like this ? It's very easy to look back and see the path that brought us here. Well, that's the benefit of looking back. If we go back to 2015 and 2016, and ask: What will things be like in ten years? I don't think I could have predicted it accurately, because the value of this collaborative development , the feedback, the ecosystem, and everything we were focusing on here, was aimed at pushing the field in a direction that was difficult to predict. We are now at a stage where quantum computing is moving from being just an experiment to an aid in other experiments, isn't that right? As a tool in itself , it will likely require a great deal of work to refine and perfect it , to get it to a stage where it can be widely adopted. But we are already seeing it happen , and that feels great. Because you know how difficult it is to adopt these new ideas, bring these people together, and develop these tools, and despite everything I've suffered over the past decade, I say it's all worth it, considering what we're seeing today. Before we talk a little about the future and integration here, I would like to ask if you could tell us about the results of Okedge’s recently published research on neutron scattering, which roughly defines what I just explained, using quantum mechanics not just as an experiment in itself, but as an aid to it. Can you give us a brief overview of what happened and why it was so exciting? Absolutely. I would like to mention that this work was sponsored by the Quantum Science Center. One of the National Quantum Information Science Research Centers of the Department of Energy, in partnership with IBM and Purdue University at Los Alamos, as well as several other institutions, aimed to verify the possibility of using quantum computers themselves for this purpose, i.e., creating a scientific simulation. I do not mean here to evaluate the performance of the quantum computer , as we have been doing that throughout the past decade by using scientific problems to evaluate the efficiency of the quantum computer's performance. What's new in these latest results? This is called quantum simulation evaluation using neutron scattering experiments. We are now actually running a simulation of one-dimensional material models on a quantum computer. This was at IBM's headquarters in Boston, if memory serves. Then, by comparing that simulation, which relies primarily on time dynamics to calculate what is called the dynamic structure factor, with the neutron scattering experiment carried out at Okage's home neutron source , one of the brightest neutron sources in the world, which directly measures the dynamic structure factor of scattering, we find that the simulation and experimental results match perfectly, which strengthens our confidence in both the quantum computer— something we had anticipated —and the model that the quantum computer represents , and the dynamic structure factor that it calculates, which is in fact similar to the model in the experimentally tested system. Yes, as Richard Feynman stated, nature is not classical; it needs something quantum to simulate something else quantum . That's what we're already seeing, isn't it? We have experiments that conduct in- depth quantum research into the internal structure of the molecule, and we have a system designed to simulate the quantum behavior of the molecule. So, it seems like it is actually happening . This is an example of it happening. It's truly amazing . correct. completely . correct. I would say that the added benefit is that we know how quantum computers will evolve in the future. We are confident that as it becomes more complex, as it gains access to more complex materials , and as it looks at more precise energy accuracy , all of these things will become within our grasp . These are things that are currently beyond the reach of our traditional methods, so this is very exciting. I mean, Jerry, you know this is a challenge for you, right? How do you feel when you look at that research? Now, our mission is to expand, is n't it? I mean, that's the basic idea, we have plenty of time ahead of us. Now, it really comes down to how we deal with these real problems, specifically in areas where we know there is no chance of even looking at them using conventional approximation solutions or, as far as we know, only conventional computing methods. This, in my opinion, is precisely why there is a public and transparent roadmap outlining our destination, both to unify our efforts internally regarding the actual technologies that need to be developed, and to serve as an indicator to the community of what is coming, and how we can begin to optimally plan for the problems we want to study at these different stages, such as the next Nighthawk release increasing the number of gateways it supports, or when we get, for example, to our fault-proof systems that we plan for later in this decade. As you know, I think everyone has seen the roadmap, but where are we headed? And how will we continue to achieve our vision for quantum computing in the cloud over the next two years? Okay, so where are we now? Today we have amazing devices, with capacities exceeding 100 cubic units, namely the Heron and Nighthawk devices. These devices operate electrical circuits with a scale of up to thousands of gates, even more than 5000 gates, and by using error reduction techniques, we obtain accurate results. We are currently conducting several experiments using these devices. For example, the neutron scattering experiment has been successful , in addition to some recent research that has pushed the boundaries of what can be done with these types of devices. Therefore, we will continue this phase where we will make improvements to these devices available to our customers in the cloud, and we will increase the number of gateway operations. So, we will reach more than 7000 gates. We already have over 10,000 gates, and over 15,000 gates in the next few years. But at the same time, we are already focusing on a quantum leap in what we can deliver near the end of this decade, namely a fully error-proof quantum computer . This quantum computer will be error-resistant on the scale of 200 logic units in our Starling architecture. This computer uses a new error-correction code that we developed, which is extremely efficient. We aim to be able to operate circuits with up to 100 million gates, right? Therefore, we will open up new horizons for algorithms and applications that scientists have theoretically proven over the years. Well, however, the connection also to the type of exploration and heritage feature offerings that we are trying to see and actually enable through the short-term roadmap is amazing. And then I mean, Travis, where does your business go from here? I mean, when you hear that you might be able to achieve an error-proof quantum computer in the next three years, how does that affect you? What is the roadmap for actual research? Yes, this does indeed affect the future of research in many positive ways. On the one hand, it takes us to a fault-resistant model where I can start recruiting users who know more about the details of the algorithm than they know about the details of the quantum device. Therefore, the separation of interests that Jerry talked about earlier becomes more apparent, and the benefit of this lies in expanding the base of users who are interested and willing to adopt this technology. I think adoption will eventually become an important factor in all of this. But in addition, you can start to define these boundaries, these problems that we just mentioned we cannot solve by traditional methods at the moment. Therefore , we know that we can surpass certain points, and this in itself stimulates investments in other types of devices and experiments. I will give you a great example of this with the scattering of neutrons in Okedge. We are currently building a new generation of neutron scattering facility called Target Station II, which is specifically designed to provide us with access to low-energy neutrons that can explore finer details in these materials. To complete this, we need plans related to quantum computers and their applications that can analyze the signals and fingerprints resulting from these experiments. You can imagine that with many other types of devices , such as microscopes, cameras , and others. Quantum computing and the existence of these maps that point to this, as I say, the quantum advantage in this context. Good. This points to a promising future for quantum supremacy in scientific discoveries. It is very impressive, and I think it is technically exciting. It also stimulates the scientific community, increases its adoption, and frankly, expands its integration into our scientific curricula. I think this is a great outcome for the field. As for you, as the person in charge of managing users and witnessing these changes, I would like to know how you build a team that is not limited to experimental or theoretical physics, but includes chemists and researchers who use quantum mechanics in their research. How do you know, for example, what a listener should do if they want to work in the field of quantum computing? And how does he join a team like this? That's an excellent question, and we do have some experience in this area on the side of high-performance computing , which as a society is going through generations of new architectures and hardware acceleration, which can sometimes be revolutionary . I think it's not as chaotic as we're currently seeing, partly due to tools, computational models, and the like. But at the Center for Quantum Science , the Quantum Computing Users Program, and other parts of the National Laboratories, we are essentially building these teams from scratch to focus on these scientific problems. The value of their expertise lies in their ability to now make tools available to the scientific community, such as open-source software, libraries, and datasets, all of which make it easier for others who wish to participate. If I had an application library that I developed specifically for quantum computing, and I published its interface in an application library so that a computational chemist could use it, I might not be the one to discover the next breakthrough in chemistry, but they would be able to use that tool or library to perform simulations. This of course applies to many different fields, and goes beyond scientific research to include industry and other applied fields. Therefore, I believe that part of the challenge going forward lies in supporting and sustaining this infrastructure that we have built, with the expectation that people will increasingly need it to encourage its adoption. We are nearing the end , but Jerry, I want to hear from you about how, as someone who cares about these systems, you support and help build this growing community . Yes, I mean that this is a big part of our current focus on quantum supercomputing. We are seeing a strong trend towards what we call " high-performance computing specialists". There are many people at Okeidj and other parts of the U.S. Department of Energy who embrace a supercomputing mindset as their area of ​​expertise. How can we ensure the integration of quantum computing into this ecosystem? Well , a big part of our approach with Keyskin has been to develop the software package in ways that make it easy and intuitive for these users to use. In addition to providing appropriate coordination and resource management tools. We want to speak the language of high-performance computing users, and this will enable them to access it as quickly as possible. I see this as similar to what happened with graphics processing units (GPUs), which were initially marketed for vector graphics and gaming. But who contributed to its development to a higher level? Users of scientific computing, such as those working in national laboratories. Therefore, we want to see this same enthusiasm driving the wheel of science using these devices, and we are working with them to achieve that. I would like to ask one last question , which may be more interesting. Well, I usually like to ask what excites you most about the future, but on the occasion of the tenth anniversary, I would like to hear your look back at the past ten years. Let's start with you, Jerry. What is the biggest surprise? What is the biggest shock you have faced now after everything we have been through? What is the biggest surprise? I mean, everything was a surprise to me, wasn't it? I thought we would make a big splash by posting it online and attracting people's attention and enthusiasm. I never imagined it would change almost an entire industry. I assure you there is an entire industry dedicated to quantum computing, isn't there? There are startups, software, software integrations, and an entire ecosystem of economic entities that all rely on quantum today. Seeing that , you know, it's certainly gratifying to be in such a prominent position as leaders in this field at IBM, but also to have such great relationships with National Laboratories and with Travis at Okedge is truly rewarding. amazing. Okay, Travis, I'll ask you the same question . As you know, looking back, I've been here almost from the beginning, managing the users. So , what was the thing that surprised you the most and that you didn't expect at all at that time? I had a conversation with my wife . That was over 10 years ago when we were first talking about quantum computing, and she asked me, "Well, what if it does n't work?" And I think she was healthyally skeptical on her part, and I said, " Okay, we're not going to give up." I think the past ten years have proven that; getting to where we are today has been a huge challenge. I believe that the enthusiasm and excitement that society enjoys is not just a passing and temporary phenomenon, but something real, something contagious, and something we discover is constantly growing stronger. So I think this is the biggest surprise for me. It's easy to look at this as just a job, isn't it? As a professional scientist, I can say that there is something very inspiring to the human spirit that pervades the quantum computing community right now. This makes me incredibly happy . amazing! Thank you all very much for this conversation. It was truly enjoyable, and we thank you very much for your experience, and we wish you another ten years and more! Thank you very much. Thank you very much. Absolutely. Thank you all. This concludes this episode of the "Coherence Times" podcast. If you enjoyed the conversation, don't forget to subscribe to whichever platform you use to listen to podcasts. Share your opinions in the comments section, and share the episode with anyone interested in quantum science. You can listen to us on Spotify, Apple Podcasts and YouTube via the IBM Research Channel. For more episodes, resources, and in- depth analysis, please visit our website ibm.com/thinks/mpodcasts. This is Ryan Mandelbomb. Thank you for listening, and remember that the future of quantum technology is not limited to what you are building now.