 Thank you for coming, I appreciate everyone's interest today. I'm going to quickly highlight that we as you in the state of Arizona have launched every broad quantum initiative with funding from the state. So, one of our key partners and that initiative is Quantinium. And so we've been on a Quantinium headquarters a few times in the last few months in Colorado, we've been able to see their systems up close talk to their atomic physics engineers about the state of the system and what has to happen for them to advance quantum computing. Quantum is a really complex landscape. For many of us that are used to sort of evaluating emerging technology I think we recognize that there's been some difficulty in knowing where it's really at, because there's a lot of commentary maybe even in some cases hype about physical systems about other architecture. I can tell you quantum sensing is already reaching the market quantum networking is is on its way as well. Quantum computing is so much further along than I thought it was when I did sort of our deep dive to see the state of the technology. So, we feel a lot of urgency, and part of that urgency includes the need to develop workforce areas and you know, not just quantum computing but software engineering associated industrial engineering all kinds of stuff. So if you're a student, and you're interested in this at all. You could participate in our initiative absolutely. We're just getting off the ground we're about five months in things are developing quickly we're about to get into a lot of project activity. So, reach out to me my name is Sean Dudley. We've got this thing it's called the quantum collaborative initiative overall. But if you if you have questions if you're curious if I want to learn more please reach out. We're excited to have mark here. We found out mark was already on his way here through actually some student and other connections. So, we've sort of sponsored the event now through the initiative make sure you guys have some some food and coffee. But otherwise really excited to have mark join us Cambridge quantum and Honeywell kind of quantum forces combined to create quantity and quantity is not a new sort of spin out per se about 500 staff or something like that. Big company, you run the Cambridge side. Yep. So, without further ado, let's take a look at ticket here for mark. Thanks. It is great to be here. I've taken many trips to Arizona states over the past few years one of my best friends is Malik Peric who was in the physics department and I saw last night in the elevator he'll be talking about what if Victorians had discovered quantum computing. So I encourage you to go to that and please do give Malik a hard time for that topic. It's great to be here again and to be talking about continuum. How many of you have heard of continuum before today. Fair, fair number of you. So, so what Sean said is correct that. So quantum has about 20 years of quantum expertise, even though we're a company less than a year old. And the reason for that is because there was Cambridge quantum computing which developed quantum software. And that's where I had come from. We developed partnerships with all the major quantum hardware groups like Google and IBM and Microsoft and Honeywell, and the partnership with Honeywell went so well that almost exactly a year ago we merged so Honeywell cooperation spun out their quantum division. Cambridge quantum merged with them to form continuum. And so we're less than a year old. We're still privately held although we've made big plans to go public at some point probably. So in this presentation I want to tell you about our company about the hardware that we do and about the software that we do. And I want to emphasize ticket as Sean mentioned, so I'll be talking a lot about that. This assumes that you kind of have some basic quantum programming background. So if you don't that's okay I think this is being recorded and so then you can come back and watch this. So, so to tell you a little bit about our company, we are the largest full stack quantum computing company in the world, and John was correct that we have about 500 people now which is amazing because when I started at Cambridge quantum, just over five years ago we had about 30. And there were only maybe 10, maybe 10 quantum startups. So it's, it's wonderful to see how much the field and our company has grown. The focus is in building quantum applications and I'll be talking a lot more about that very soon. These applications are built on quantum algorithms and these could be in chemistry or machine learning or optimization or cybersecurity. And all of these algorithms interface with tickets and as I said I'll be talking much more about that in a moment. Ticket is the liaison between the algorithms and the hardware or simulator, and that could be our hardware, or it could be anyone else's ticket is completely universal. The way that we work with companies is that they come to us with some sort of problem. So they say we, we can't do this kind of problem with a normal computer, do you think quantum could help. And our scientists study the problem and identify the ones that we think that quantum could make a difference in in the next few years. So we're not interested in, in maybe 20 years from now we want to do kind of in the next three to five years, which we think is a realistic time frame. And if we do identify something that our scientists think that quantum could make progress in and it's interesting for this company. Then we start working on it. So we do a very thorough background and literature search. We write it from scratch. We run it on actual quantum hardware, either ours or someone else's. And then we assess is there a quantum advantage, and we can scale this and keep working on it and eventually lead to something that's commercially interesting, or not. And so we, we try to honestly assess whether we should continue or not. We've had dozens of projects in this manner in fact just yesterday we announced one more with Mitsui, the Japanese company. We have offices all over the world. So our headquarters is in Broomfield near Denver. This is the numbers here are slightly outdated. We're hiring so quickly that HR hasn't sent me a newly updated one but this kind of gives you an estimate. So we have about 200 people in the Broomfield office and that's where our headquarters are, and that's where our hardware team is primarily based. We have a software team because they came from Cambridge Quantum that was of course a UK company and so they are mainly based in UK between London, Cambridge and Oxford. We have smaller offices elsewhere. So in, in Minnesota in DC in Paris and Munich and Tokyo, I happen to live in New York City we don't have an office there I just travel around a lot and I hope they don't open an office here because I like working from home. It's getting like crazy. So if you are interested in this, please go to a website and check out the openings. We have we have all the current positions open right now and we're always updating it. And if you don't see something that interests you and takes advantage of your focus area, please send us an application anyway and just indicate what kind of position you would like to have. And again, we're always updating this. And I also want to emphasize because I know there's a lot of students in the audience that we have internships, and this is exactly the right time that you should be applying because we make up, we make decisions around December for next summer. And, and this is a very important process for you to get to know us and us to get to know you, and many people have been hired through this route. So please do get in touch with us. If you're interested in the internships. So I first want to talk about the hardware that we do and as I explained this comes from the honeywell side of things. And so, so this has been in development for about 10 years. You might know that there's many different technologies that can be used to build quantum computers. And we happen to be using the iron trap technology. So there are a few others like superconducting and photonic and full diamond such and they all show promise. We have decided to focus on this one. And the results have absolutely been amazing. You're probably familiar with the Moore's law, which has shown over several decades that classical computers have doubled in power every 18 months, roughly. The quantum is on a much faster trajectory than this. So this is a chart showing the quantum volume over time and the quantum volume sort of an estimate of the power of a quantum computer. And this turquoise line here this shows the progress we've made it was about two weeks ago that that last dot of 8192 is up there, and that corresponds to 13 perfectly working qubits. And we have 20 physical qubits, but but this corresponds to effectively like 13 perfectly working qubits. And that's quite a bit more than even the closest competitor which is IBM of 512. So so we're very proud that this is the highest performing quantum computer in the world that we have built right now it looks like this. It's a round trap and it's linear, which means the qubits the ions, they simply move back and forth. We physically move them around is a turbium ions that we use in this to say system. So to explain it in a little more detail how this works. So the system loads and Adam and ionizes it. So now it's a qubit, and we trap it using the electrostatic potential there. And then when we want to do things. So we move it to what's called the gate zone. So we actually transport it to a new region, and then we zap it with lasers. So this manipulates the qubit. So you're changing the value of it. If you're, if you just need to operate on it by itself, you do this. But the really interesting thing happens when you entangle them. So this means that you're getting the value of this qubit to be correlated with the value of this other qubit. And with that you have to put them in the same place and zap them simultaneously. And now they're entangled. So they're in this quantum uncertainty state together. So to see how this works in practice. This is a very simple quantum circuit, which you might have seen. And so, so for those of you who haven't seen it before so each horizontal line there represents a qubit, and you read it from left to right. And so at time step zero, all three are going to get initialized. And then over here you can follow along and you can watch what's actually happening in the machine. So at time step zero, they all get initialized so they're in the zero configuration. And then at time step one, we applied the Hadamart transformation to the green and the red. And then in anticipation of what's going to happen next, we need to move the green qubit over, because we want to entangle the green and the red. And so we do that together. So that corresponds to the CNOT gate here. And then in anticipation of the next time step, we want we want to move the red one back over near the blue. And so then they get entangled because of this CNOT operation. And then finally we want to measure the green one. That's that operation there. And then be using so so so you see that a quantum circuit corresponds to a very sophisticated dance that you're doing moving these qubits around and manipulating them. This is an actual picture of what the qubits look like in our system we use a special camera. And it's very short distance 150 micrometers there. But we still can't take pictures of this. We have a roadmap for the next decade. So you see that right now we're only at the first step. So even even though we've built up all this technology we're still at a very preliminary stage. So it's linear. We still have the highest measured quantum volume of any machine in the world. We have 20 qubits with a very high fidelity rate. Very soon, in the next months, if not weeks, we will be unveiling the second generation, this race track formation. There's still one path but there's kind of two ways to get there. So it can move around like that. So there's a shorter route sometimes, and the qubits now need to navigate the turns. And the reason this is so important is because the next generation will be the grid, or it looks like a city grid, and has to move around with these junctions and such. And this is all building up to integrated optics. So that's really where it gets scaled when you're going to have many, many qubits doing things together. So this is clearly much more sophisticated than where we are now our engineers think that this is a realistic roadmap. So this is where we're headed within 10 years. This is just a little more technical detail about things with with today's machine so as I said it has a volume of 8,192 20 physical qubits, the coherence time is three seconds. That's an amazingly long time for these qubits to retain their special quantum properties. Yeah, this just again shows the progress in the quantum volume over just the past few years and we believe that we can continue with this. We believe that we will have commercial applications in as soon as three to five years, which is really amazing because it takes time to develop the software. So the companies that are starting to do things now will see a benefit in just a few years. So the next thing I want to talk about is ticket. Right now, as I mentioned, we have many different technologies to build quantum computers. There's the ion trap like we use but there's many others. And there's all these different algorithms that we write and the algorithms have to be written in a quantum language and there's many different languages right now. So it's kind of a confusing state of affairs. There's lots of machines and technologies and there's lots of languages. So what do you do to make programming easiest. Well, several years ago, because we as Cambridge Quantum of software company we had this challenge. So we developed tickets, and at the time this was purely for selfish reasons because we wanted to focus on the software and not have to worry about the hardware generations that were coming out. And so this made it easy for us to just focus on the software and then take it would automatically run our software on on whatever hardware we wanted in the most efficient manner possible. And this proved so successful that about a year ago we made it open source. So not only is it free and anyone can use it, but every line of code is now published. And so you can inspect and see exactly how we did things. And if you have some idea, you can contribute. We've had almost 600,000 downloads of ticket. And so it's, it's now proving to be one of the most popular quantum software packages online. And by far the most universal. So these other software development kits like kids kid and circ and such. You can use them, but you're kind of a limited in which platforms they support, because kids kit developed by IBM was really intended just for the IBM system and they've opened it up a little bit more, but ticket from the word go was designed to be universal. So ahead and keep your software written in kids kit, you just add one line saying use ticket to compile this on such and such a machine, and take it will automatically do the translation and figure out the optimal path to do this. So there's two ways that ticket does this and I want to go into a little more detail for those of you who are quantum software developers. It tries to optimize just your circuit itself. This has nothing to do with the hardware. It's just this code that you've written. So it goes locally through your code so it looks at the individual gate combinations there, and it looks for ways to simplify it. So your, your code will mathematically be identical. It won't touch that at all, but it will find a simpler version simpler way to express that. And to see how this would work. So this is a typical circuit. This is a little bit more sophisticated than the simple three qubit one I showed you before. So this is kind of a more typical circuit with many qubits and many operations. So take it will go through this and it will look for all these combinations of operations, and it will look for patterns that it can identify as ones that can be simplified. The simplest one is up on the upper left hand corner here. So that's to see not operations there in a row, and mathematically, you might know that that's actually identical to nothing, because they cancel each other out so mathematically this means it just leaves the qubits untouched. So if you ever sees this combination, it just removes those two operations, those two gates. And this means your program runs a little bit faster, and it's a little bit less likely to have an error introduced, because you don't actually have to do anything now. And these other replacements are less obvious. But if you're curious, you can go ahead and check this article here I've given the preprint. This explains the mathematical theory behind us so our scientists spent a lot of time finding patterns that can be simplified. And if it does that, then it takes into account what hardware or simulator you want to run your code on. And there are several things that it takes into account. So the first thing is, how are the qubits connected, because in ion trap, every qubit can move to every other qubit but in in superconducting, for example, the qubits are physically locked in. And so you have to take that into account. And the second thing is translation, because the gates which are native to that hardware aren't necessarily the gates that you would use in your machine. And then finally, some machines like ours can have advanced features like mid-circuit measurements. And so take it, we'll take that into account as needed. So to see some examples of this, these are three superconducting platforms. So by IBM and Google and Rigetti. And even though they use the same basic technology, you see that the way that the qubits are connected is very different. And what's more, the gates which are native to those platforms are very different. And again, these aren't necessarily the gates that you would use in your code. So take it, we'll automatically do all this translation for you. And to see how essential this is, this is a typical machine. This is the IBM Melbourne machine. And you'll notice that it has at most four qubit couplings. So at most one of these qubits is talking to three others. But the circuit on the right there, this has some five qubit couplings, the two qubits on the bottom are coupled not only to each other but the top three. So naively, we might think there's no way to run that circuit on that machine. That's actually not true. We can do it. We can do something called a swap. So a swap is where you swap the information between two qubits. So physically, of course, the qubits have to stay in place. But this means that the information of the qubits move around. And this is a perfectly legitimate operation. The way that you do it is by introducing those three CNOT gates there as shown. But of course, adding more gates is what we don't want to do. So we only want to do this swapping unless it's absolutely necessary to get your code to run. So make it knows this and it will figure out the way to do this. So it will find that the most efficient way to do this routing the swapping such that your code runs. Obviously we use ticket for our own internal development, but many other groups have started using as well. So these are some articles written by Google and IBM and even CERN. In almost every case, ticket will allow the code to run faster and use less operations. And there were even some benchmarks done earlier this year where they can, they can take it to Qiskit and CERK on several different platforms. And in every single case, they found that ticket outperformed the others, which is amazing because it outperformed using Qiskit on IBM's machine. And they did better than IBM on their own machine even. So we are constantly updating it. As I mentioned about a year ago, we made it open source. So every line of code is now published. We're always adding new features. And there's some big things coming up, especially in that we're moving it to the cloud. So right now you download it from GitHub, but very soon we'll be having it so that you can use a cloud version which makes it much easier to collaborate. It makes it easier to use cloud platforms like AWS or Azure. This is the GitHub repo if you want to take a picture or something like that, or you can just Google Ticket and GitHub. So it's right here. There's lots of documentation and examples. And actually we recently hired another evangelist. In case you didn't notice, that's what my job title is. Another evangelist focusing specifically on Ticket. And so she's going to come up with some examples in workshops and notebooks. So there's going to be a lot more material available soon. You simply use the usual PIP install for PyTicket plus whatever platform you think you might be using like Qiskit or SIRC, where would have you, and then you're ready to go. So it's completely straightforward. So I next want to talk about some of the software that we have developed. And so these have all been projects that we've worked on with companies where they've hired us to do this. When a company does this, sometimes they want us to promote it and we can talk about it publicly. And sometimes they don't. So obviously these are the ones that we can talk about publicly. There's plenty more projects that we've done that we haven't talked about publicly. So first of all, this was a quantum machine learning project with JSR Life Sciences, also from Bio. It has been known for some time that there's a correlation between biomarkers and genes and how effective different oncology drugs are. And simple correlations can be done on classical computers, but anything more sophisticated than that really doesn't scale very well. And so there's really no hope for doing that. So they came to us and asked if we thought that quantum machine learning could help with this. And we believe it can. So in this two year project, we've been looking at correlations between these biomarkers and the effectiveness of different cancer drugs. There is a lot of interest from financial groups in quantum computing. And this is because of course, all of these, these trading algorithms, these are all just machine learning they look at past data and try to anticipate what a stock or other commodity is going to do. And the mathematical basis for this is something called Monte Carlo simulation, which is a way of modeling the apparent uncertainty of something like a stock market. And we found out that there's a quantum Monte Carlo algorithm, and it is quadratically faster than the classical version. So that means that you get the same level of accuracy running 1000 quantum simulations as you would with a million classical simulations. And so a lot of financial groups are very excited about this and we worked with several to do this and it was actually one of our scientists, Stephen Herbert, who, who proved that indeed in the real world, you do get this quadratic improvement. And so we actually have a huge team focused on this. It's my personal opinion that this will be one of the first commercial applications. And the reason for this is because when you're doing this kind of financial modeling, you don't need to be correct 100% of the time, you actually just need to be more correct than the other people trying to do this. And so even a slight improvement like 0.1% more accurate modeling is financially very, very valuable. And so I think that given this that any slight improvement would be worthwhile. And the fact that these financial teams are spending a lot of money to do this like Goldman Sachs and JP Morgan Chase, they both have fantastic quantum teams. And so as I actually think that this will be one of the first to have a commercial application so I'm very happy that we have a major focus in this area. So optimization is another big application of quantum. And so we've worked on this project with Nipon Steel to optimize their job shop scheduling. So when they have this whole assembly line process and they're trying to actually manufacture the steel. We found a slightly more efficient way to do this, using quantum computing, and a similar project was done with Deutsche Mon, where we looked at at a simplified version of their rail schedule, and found a more efficient way to do this. Real world problems that we're interested in solving using quantum. You might know that chemistry was actually one of the first applications of quantum computing. So it was actually about 40 years ago that Richard Feynman first suggested quantum computing in the context of improving chemistry simulations. And so we've had about 20 projects with companies, calculated in different chemical quantities using quantum computing. And a few months ago, we released in quantum. So just like in quantity sorry in silicone in vitro and things like that so it's now in quantum. This is a enterprise great application which does chemistry calculations using quantum computing, but you don't need to be a quantum computing expert to use it. You just need to be a chemist, but you don't need to be a quantum computing expert. So we've consolidated all of these 20 projects that we've done into this one application. So you can just enter your commands, what you want to calculate and it sends it to our machine and send you the answer back. It's around 20 electrons that we we start. We think that we can start to do and that this is actually starting to get interesting for commercially valuable projects. So some of the projects that we did work on were things like co2 extraction using metal organic frameworks with total a second project with nipon steel on modeling the iron crystals, and then a project with Roche on potential drug for Alzheimer's and how this interacts with our biology. And so so these are all the articles here if you're kind of curious about the basis for this cyber security is. Notice this is the one thing that people have kind of correlated with quantum computing they know it has something to do with hacking, and there is a good reason for that. And there's been a lot of attention recently on what are called post quantum encryption algorithms. So these are new types of encryption formulas, based on mathematics that we don't think quantum computers can hack. I emphasize that we don't think because we actually don't know. There's this ongoing contest by NIST, which you might have heard of where they invited people to submit what so called post quantum encryption algorithms. And a few months ago they unveiled four winners, but then within a few days two of them were hacked. So we clearly have a lot of work to do there. So there's a lot of attention being given to these post quantum encryption algorithms. And what has been less publicized are keys. So keys are the string of numbers that you use to do the encryption or the decryptions they feed into the formula to do this. And most people don't think too carefully about where keys come from. They sort of just generated on a laptop or some other source, and they get some numbers which look random enough. And then they just say okay that seems to be fine. But it's absolutely essential of course that your keys are secret you don't want anyone to know what the keys are otherwise they don't need to do any hacking, they can just undo the encryption directly. And so, getting your keys from a classical computer is actually no good, because classical computers can't do anything random, they can only follow instructions. They actually have a pattern to them, and this has been exploited. We've known for about 100 years that quantum physics can produce truly random numbers. And there are so called quantum random number generators on the market, and might even know some of the names of them. The problem with them is that if you just have a string of numbers if you just have the output, you have no idea where those numbers came from. So that might be okay but you have no way of checking that those keys are secure. But what we've done is produce a way of proof of producing provably secure keys. And that might sound too good to be true, but the difference is, we've produced a quantum circuit, which produces entangled qubits. And there's then a mathematical test that we can do to make sure that that entanglement is there. And that actually is exactly what the Nobel Prize was awarded for a few weeks ago. So you might have heard of this. The three guys that got it it was actually for precisely this belt test, doing this statistical check for entanglement. And so as long as the test passes. That means that it is quantum physics producing the randomness and no one has tampered or he's dropped or done anything funny to it, in which case the output is secure, you can go ahead and use those keys just fine because no one has done anything weird to it. And if the test ever fails that means something weird is going on, and in which case the system shuts down alert you and says so. So we've produced this, and less than a year ago we unveiled quantum origin. So we've actually already commercially deployed this and this is the world's first commercial quantum application. So we ran the circuit on our machine, we stored all the keys and we verify the entire time that the test was passed. And then people with with cybersecurity systems they can just make API calls, and they can retrieve these. So, so this is already commercially viable, and we're already doing this right now. So, so I've talked a lot about the commercial activities that we've done at continue, but I finally want to say a few things about some of the fundamental research that we do, because we're still very much committed to this quantum is such a new industry, there's a lot of things that we don't know. And so we actually have an entire team based in Oxford researching more fundamental aspects that the technical academic term is called compositional intelligence. The more practical application is natural language processing, and you might be familiar with this from things like Alexa and Siri, where computers. So they figured out the words that we're saying and then they figure out the meaning behind it how the words are composing sentences and such. And so this is maybe a typical sentence. John walks in the park, but there's this whole mathematical structure behind this that the computer figures out. And it was this Oxford team that works for us that actually figured out that that can be represented by this circuit. And now a quantum way of understanding this and getting quantum computers to understand this type of meaning. And they've actually been going further than this. They actually believe that this is the first step in getting quantum computers to actually think sort of a universal picture of meaning. But our lead scientist, Bob Cook, he's actually written this book called Quantum Theory and Pictures, where he's developed a pictorial way of expressing. So instead of using complicated equations, which are difficult and confusing, he's developed a pictorial way of doing this aimed at high school students. And they're actually going to do a test this year, they're going to teach this book to high school students and then give them the same tests that they do to master students at Oxford, and see who scores higher. And of course you know the results of that. But, but this is what our entire team is based on. And then a few months ago, we were very excited to announce this collaboration we did with the Simons Foundation and their Institute in New York. It had been theoretically suggested for some time that there would be this new phase of matter based on these sort of quasi crystals or time crystals you might have seen in the news. So it actually it was a sequence related to the Fibonacci sequence, we coded this in the laser pulses and this actually shows up in quantum error correction. So they, they applied their theory to our machine, and they found that it was indeed true. And so this was a very exciting results again we did this with a foundation, it's just fundamental research there's no immediate commercial application, but, but we were still very excited about this. So that's it for my presentation about our company but of course I'm happy to answer any questions that you might have. Sure, sure. So, so to answer your question it's a really good question. So first, let's go back. Forget about quantum for a moment. How do we know that that the encryption algorithms that we use right now for normal computers how do we know that that they can't be hacked. So, so the most common one right now is called RSA aimed for the authors of it, and it has to do with the fact that it's easy to multiply numbers together. So I'm sure you've all figured out, you know, you learned in elementary school, given two numbers, you know how to multiply them. But you might have noticed at some point in your life that if someone gives you a big number, it's really tough to find the two prime numbers that needed to be multiplied to do that right stuff to factor numbers. But no one in history has ever figured out a fast way to do that. And so you basically have to guess and check. And computers aren't any different. So computers have difficulty doing the factoring as well. So several years ago, RSA developed this way of, it's sort of a one way door, not automatically speaking to do the encryption and using this multiplication. Who's to say that someone didn't figure out a really clever way of hacking that and just didn't tell anyone. So I actually can't prove that no one can. It's extraordinarily unlikely because we don't know of any examples where someone directly under the mathematics. When we hear about hacking, it's usually because someone installed the software wrong, or someone wrote their password on a sticky pad, or someone did something silly it's human or it's not mathematically that they directly undid it. So so we actually can't prove that normal encryption isn't defeatable. We just have good reason to think it is just based on the decades that we've used it. However, it was realized about 25 years ago that quantum computers could efficiently do that factoring. And back then. This was academically interesting, but there were no quantum computers and no one saw this as a serious threat, but the past few years people have taken this as a serious threat. That's why they've developed new mathematics which we don't think quantum computers can defeat the answer to your question is, we don't know, we're still figuring it out. And as I kind of made a reference to a few months ago there were four so called winners. And then within just a few days two of them were defeated people figured out shortcuts, how to do this so we clearly have a lot to learn about this. But even if we don't have mathematical proofs, we still can have good reason for thinking something is secure though. Yes, so. So first, I think it's important to distinguish the data sets from the process used to analyze the data set so so given the same data set so that we're comparing apples to apples. Quantum machine learning should definitely improve some versions of analysis because mathematically quantum machine learning is completely different from normal machine learning. I've sometimes seen in headlines that they sort of mistakenly just slap on the word quantum to imply that it's faster or cooler or something like that. But mathematically it's completely different. And in particular it can pick up on patterns that would have been missed by normal machine learning models. And so this is why we're very excited that quantum machine learning might be able to find patterns and data that that would have been missed otherwise. Yeah, it's correct. So what happens is that these qubits. So, so the reason that quantum computing works is because it uses two special quantum properties superposition, which means that it can be in two states at once, or more than one state at once. And entanglement, which means the value of this qubit is correlated with the value of this qubit. So those those quantum properties that are very fragile. And if there's any disturbance from the outside, then the quantum properties go away. And it becomes classical, it means it reverts to just like a boring normal bit, like a one or zero exclusively right. And so the trick is to get the qubits to last to be coherent as long as possible so that we can do our operations. Exactly. That's it. That's exactly right. Yeah. So we absolutely are, to me, we are trying to extend it. So that's one of the major goals of quantum computing is that we try to extend the coherence time of the qubits, so that we can do we can fit in more calculations in the in the time allowed. We tried to make the calculations as fast as possible, so that again we can do more. So yes, we absolutely do try to do it. Different technologies have different. They allow longer coherence time than other. So one of the reasons that we use ion trap technology is because it has a very long parents time, like three seconds is actually a really long time on the atomic scale. One of the other technologies superconducting. One of the disadvantages of that is that it has a pretty short coherence time in the order of milliseconds. The trade off is that the superconducting machines have a very fast operation time scale so the operations can be done in nanoseconds, whereas with ion trap, it takes about milliseconds to do the operation so it's actually a funny coincidence. The ratio of those the number of operations you can do is about the same on the order of hundreds or, or a thousand right now. So it's just a coincidence as far as we can tell. Not to my knowledge, to the best of my knowledge we are all in on ion trap technology, but but you're absolutely right. We don't know it's still very early. It's like one mile into a marathon. And so who's to say that in a year or something, someone discovers that there's some critical flaw we didn't know about an ion trap, and some other technology is much more promising so I can't rule that out that's not impossible. So the reason of that, if that were the case then of course we would just have to change direction. So I wasn't around back then, but I'm told that the reason that that Honeywell decided to choose the ion trap approaches because about 10 years ago when they first explored this. They looked at all the different technologies and they identified ion trap as, as not only promising, but it had all the expertise that honey well already had. So, we didn't have to go outside we already had people who could do all those things. And so I think that's why they went with that direction, and it's proved very successful as I showed with that chart right now we have the highest performing machine. But yeah, if something unforeseeable occurred then I guess we would have to change direction. Yes, so we'd, we don't have a machine learning platform per se. So, so our, our team has developed machine learning programs, they're proprietary, so we don't, we don't share them. So ticket would allow you to use anyone's machine learning software or anything else like that. It actually has a penny lane plugin, since you mentioned that specifically. You're exactly right yeah. So there is an initialization process it had, it's quick, but you have to do it. That's right. Yeah. In the back. Yeah, so we don't do anything with quantum sensing. I know many other groups and back I think there's a group here that does quantum sensors. Yeah, a continuum we're entirely focused on quantum computing. Just for those of you who aren't familiar with it there's actually kind of a joke that a bad qubit is a good sensor. Of course, with qubits you're trying to make them as isolated as possible you're trying to get them to not be aware and sensitive to outside noise, but with a sensor it's the opposite you do want them to be sensitive to outside noise. And so they're, they're closely related technologies but for kind of opposite reasons. But yeah we don't do that. Yeah, it's still very low level and people have commented on exactly what you just did yeah so right now. I showed you those, those circuit diagrams. That actually is how you program it so we actually use those pictures and we actually do those things. So you are programming every qubit individually and you need to think in quantum physics. At some point someone will probably develop something more user friendly. And in fact, some of you might know this, the software developed by MIT called scratch for children to learn programming because you just fit the blocks together and something like that. So I heard a rumor that someone might try to do that because it's open source for quantum, just fit the blocks together and that's how you would program so I would love to see that. Yeah, programming a quantum reader so counterintuitive that that it would be nice to have something more user friendly. Yes, in fact you actually hit the nail on the head so the reason that quantum is so different. So, if you if you get nothing else out of today, I just want to impart that you can't take classical software and just run it on a quantum computer, you have to completely start from scratch. And the reason is exactly what you just mentioned. It's because classical computer programs are just a very complicated flow charts. Right, it's a bunch of if then statements where the computer follows commands, yes or no data that ended up with a quantum computer though things are in superposition so it's true and false at the same time so which fork does it follow. It actually has to kind of follow both is a funny thing. And so that's why we program it with those qubits. So the qubits get all tangled up in superposition and entanglement, and then only at the very end do you do the measurements, and you actually get to see the answers but until then, you don't get to decide and see what all the quantum physics is doing. And so, so yeah so it's completely unlike classical, where you can trace the decision making. It also makes it much tougher to debug. Exactly yeah yeah so it's just a very complicated thing that the qubits are getting all mixed up. Yeah. Yes. And I think, in principle, yes, because you could just kind of. It would be very inefficient, because you could just, you could manipulate the qubits such that they only act like classical bits, you could, if you didn't do anything with superposition and entanglement, and you limited yourself to very simple commands. Yes, but no one would do that because you know right now we have 20 qubits so a 20 bit computer isn't very worthwhile. But you could have also asked the reverse question, could you use a classical computer to simulate a quantum computer. And the answer is yes, up to a point. So you can actually use a classical computer to simulate you could use a laptop to simulate up to about 30 qubits. And you could use a supercomputer to simulate up to about 40 qubits, and that that might seem weird to you because 40 doesn't seem that much more than 30. The reason is that why is it going from a laptop to a supercomputer. But remember that every qubit doubles the number of configurations. So 10 qubits actually corresponds to a thousand fold increase in how many configurations. And so, if a supercomputer is about 1000 laptops, that's, that's why so so right now we do a lot of stuff on simulators because it's cheaper and easier than running on normal computer. But of course, at some point, but the whole reason we're doing this is that we want to we want to run it on normal on actual quantum computers. Yeah, they're, yeah, it's because of the fidelity. So we have much higher fidelity, our qubits are better and cleaner. Yeah, they do what we want them to do more often. And so that's actually responsible. It is a little bizarre that they have so many more qubits, but their volume is much lower. It's also funny because they came up with the quantum volume. And so, but anyway, that's just how the definition works out. And so it's much more important to have good qubits than the number, as you can see. Are there any other. Yeah. Yeah, sure. So, so the coherence time is how long the qubit will last. Right. So you need to do all your operations within that time. It is absolutely true that different operations take different amounts of time. So the C not gate that I mentioned so that takes some amount of time. Different operations will take other amounts of time. So there is a budget you have, you will, of how you spend that coherence time and how many operations you want to do. Yeah, that's maybe you can only do 50 of another. Yeah, will again some operations take longer than others. Yeah. It's just the amount of time yet. There's also an error rate. I didn't focus as much on that but different operations also have an error rate. If you do not. So you notice that some of the operations, no circuit diagrams, some of the operations acted just on one qubit by themselves and other operations involved to. So that's where they get entangled right. The error rate is much lower if you're only acting on one qubit, because of course, if you have to it's just messier, there's more things to go wrong. And so the error rate is much higher for the two qubit operations. Yeah, the which. Yeah, we don't do three at a time. So, so, so I know what you're thinking though, because sometimes the circuit diagrams it looks like you're doing a three qubit operation. There are clever ways to rewrite that as a series of two qubit operations. And so that's what we're doing yet. Three would just be a nightmare. The engineers tell me so. So, so yeah, we rewrite it as a series of two qubit operations to achieve that result. We initialize it and then we, we, we see how many operations we can do and around three seconds, things tend to go wrong. I don't, I admit I don't know precisely how they measure it. It obviously isn't exactly three and every, there's some sort of distribution I don't know exactly but but that's the number that that we've been told to to say yeah. That is that is a very good question and if I knew the answer to that, I would be very very famous yeah. So how does quantum physics actually work. I truly don't know I don't think anyone really knows. But yeah we've learned, we know what the math is basically linear algebra, but, but yeah we know we can initialize the qubits, we know that we can mix them up and do these operations, such that they're in this quantum state and then when we measure it. We're not passing it. So, even though it might be in a mix, a very complicated mix of states, when you measure a qubit you're only going to get zero or one. Those are the only two answers that you'll ever measure. But, but if you, as you learn to do quantum physics, you'll, you'll see that the intermediate steps are much richer than that. One or two more question. Yeah, yeah. Yes. Yes. Exactly. You can do that. Yeah, so, so with, so within three seconds you have to do the operations and then measure it. So you can get some answers and then yes, you can reinitialize everything and start over and run another program if you wanted, but, but yeah you have to do all the operations within that window. And then I saw one. Yes. Okay. Yeah. Yeah, the output is binary. So the output, the output you'll get are just zeros and ones. It's all that the intermediate things and the frustrate the amazing thing and the frustrating thing about it is that you don't ever get to peek inside and see what the values are. So in quantum physics, we have this thing called a wave function. And, and so, so we can model it. And we have 100 years of experiments which show us that this is the right way to describe reality. But, but you don't ever get to peek inside, unless you're running a simulator, in which case you can kind of cheat and the computer can kind of pretend this is what the way function is. But yeah, unfortunately nature doesn't ever let you actually look and figure out what the value is you, you can measure something and you'll get an answer. And that's it. It's a, it's a yes or no, even though you have all these complicated steps in between, every qubit, when you do the measurement operation, you'll just get zero or one, just like a normal binary type of operation. Good. Thank you so so much and again here's my email if you'd ever like to send me more and as Sean mentioned we just signed a collaboration agreement and so I hope that I'll be coming back much more often. So thank you all.