00:00:05:11 - 00:00:34:00 Unknown Welcome to the podcast. 00:00:34:02 - 00:00:52:11 Unknown Awesome. Thank you, thank you. What a pleasure to be part of the podcast. 00:00:52:13 - 00:01:20:05 Unknown Of course. Well, Cesar Martinez, I'm a hardware engineering manager at the moment at TCI. I have 15 years of engineering experience, and if I count the amount of things that is assembled in my life, probably almost 40. Yeah. So I'm, you know, in the world of quantum computing at the moment. 00:01:20:06 - 00:01:40:00 Unknown And enjoying it. Enjoying it. 00:01:40:02 - 00:02:12:22 Unknown Yeah. So, you know, I was a little kid that was a tinker since a since beginning, you know, disassembling a lot of stuff. Very curious. I probably owned a lot of money to my parents that, you know, a lot of things that are broken. And I was not able to put back together. Then at some point when I went to school, I was into the allowed into physics and mathematics and so on. 00:02:12:23 - 00:02:42:23 Unknown And my tendency at the time was, let's say, why not physics? But then I saw like, okay, is there's not a lot of market in my home country for a physicist. So then the immediate step was engineering. And I have some family that was in the electrical engineering environment kind of with them I like it. And yeah, that's how I started in engineer basically. 00:02:43:01 - 00:03:18:02 Unknown Then, you know, my first work was in technology is by color doing, you know, implementations of 3G at a time. And, you know, probably a technology that is way bypassed right now. And then a couple experience in the public service, more in the it and so on. And then I decided to give the jump on, on, you know, I always got the like the bug in my mind that I wanted to do. 00:03:18:04 - 00:03:47:17 Unknown Well, my background in school is electronics and telecommunications. And I was about, you know, I want to do something marrying high frequency. And then I decided to came to do a master's degree. And once I finish that, a couple things, like I started in optical communications area and then drones to drones in competitions and 00:03:47:23 - 00:03:57:19 Unknown finally PCI. 00:03:57:21 - 00:04:04:04 Unknown Correct? Correct. 00:04:04:06 - 00:04:35:12 Unknown Well, you know, I've been learning about this journey to is something that it kind of is in the mind of everybody and a you don't fully understand it. But, from in the my research group back in, in college and what I've learned here in Chicago, it's a different method of doing computations. Right. And how is the different method is in two ways. 00:04:35:12 - 00:05:09:19 Unknown Basically one is how you represent information and the other one is how you approach to solve problems with it. And a big example can be on this, you know, analogous the first computer that you had in the in the world, you know, a pebble is how you represent a unit, right? And that's how you represent your first a point of information, a classical computer, you know, your cell phone, your laptop, you know, what it does is represents information, information of voltages, right. 00:05:09:21 - 00:05:51:06 Unknown Quantum computers represent information using the behavior of atomic or subatomic particles. Right. And then how you approach the process is basically how do you process how you solve problems with it is, you know, in your course is basically you moving a pebble from one corner to another one. For classical, it can be, you know, turning on switches, transistors, that is basically what is going on in your processors or the other X and in classical is and and those are very approach. 00:05:51:07 - 00:06:25:03 Unknown Right. You know, you have to follow along in what you're doing. It's very systematic happens in cadence right. But quantum computing is more like approaches in a more in a stochastic way is, you know, and is how you see those particles and how you later measure the state of energy that these particles get. So and this is a very common example around, you know, if you have to solve a maze, you're if you, you know, you need to walk each part. 00:06:25:03 - 00:07:08:12 Unknown And then if you get out at the end, you have to return back. And that's very a deterministic way to solve it. And that's how you normal computer, your classical computer will solve it, right. On the other hand you continue computer would float all the parts at the same time and find the solution very efficiently. So look for the minimum or maximum energy to solve a problem. 00:07:08:13 - 00:07:23:09 Unknown Correct. 00:07:23:11 - 00:07:45:02 Unknown That's correct. So that's that's the approach right. It's are various very probabilistic a way to do it. You know like you are always in the middle. And then you, you let this particle like, you know, you have cited and then you the particle deciding which in which a state follows in a lower or higher energy. And most of the times you don't have all in the same time. 00:07:45:02 - 00:08:12:14 Unknown If the according to the excitation that you provide and those particles can be different right to. And then we can start describing what kind of particles and what kind of technologies are there. You know, the one of the concepts that all the people is around when you talk about Google, Google, IBM's, is that server conducting a technology that is basically electrons called interact in a separate conductor in a superconductor. 00:08:12:16 - 00:08:41:06 Unknown Then the Kukai approach that is basically using light and photons in order to to do the computation. Then, for example, you have another one that is a semiconductor base that's basically using electrons to capture, you, you put it in a semiconductor, and according to the number of electrons that you, you, you put together in the space of area, you can determine what is the energy level of those. 00:08:41:08 - 00:09:00:13 Unknown You can use ions. That is basically, you know, you put an iron that's basically an atom that is a charged particle and even excited with a laser. And then you figure out in which stated for, so you can just take a normal atom and also excited with a laser or water source 00:09:00:15 - 00:09:23:07 Unknown and will do what this information. Right. 00:09:23:09 - 00:09:31:10 Unknown Yes. 00:09:31:12 - 00:10:20:14 Unknown Right. So it behind every quantum computer most of the times is a classical computer in some extent doing control doing looks right. So it's not that you're going to kill, you know, with a quantum computer, your, your normal classical computer, you know, it's complementing. It is boosting how in general you solve the problem. 00:10:20:16 - 00:10:53:13 Unknown So, you know, she is is basically in photonics and, and we are leveraging photonics in the entire level. So we can say that we are a Harvard company that is focusing in all the quantum spectrum. And that is, you know, a if we're doing quantum computing by itself, we are doing quantum intelligence, we are doing quantum cybersecurity, quantum and quantum communications when doing quantum remote sensing. 00:10:53:13 - 00:11:18:15 Unknown And, you know, to go a little bit more on that is, you know, you can't computers kind of what is it before in quantum intelligence. What we are focusing is basically AI modeling with another kind of quantum computer like different kind of computer, which for each one of us we have a product. We just launched a product focusing in AI. 00:11:18:17 - 00:12:00:13 Unknown And then for remote sensing, basically we are using lie to, you know, determine how the atoms in different materials or something like that reflect any kind of photons in some direction and all this stuff. And basically we can determine a lot of things. Right? Or, you know, the one of the big advantage of using optical and light in general is, is, is that the coherence of, of the, of the states of, of of of the of the bits qubits take, they maintain themselves for a long, long, long time in distance or, or in time basically. 00:12:00:19 - 00:12:15:11 Unknown And that allows you to prosecute communications. So, you know, cryptography and all that stuff can be boosted up for in sort of the purposes of, of, of, of having a secure, a more secure environment of 00:12:15:14 - 00:12:51:15 Unknown communication, basically. 00:12:51:17 - 00:13:18:23 Unknown Yes. And that goes back to the technology they are using to. And that's also one of the main reasons that we decided to hyper focus in photonics, because it allows us to, you know, be accessible and scalable and, and shrink it as much as possible and working room temperature, if you, for example, go back to the situation of superconducting Kuka, sorry, IBM's Google quantum computers. 00:13:18:23 - 00:13:45:16 Unknown Basically, you need to have a these devices in stereo Kelvin, which, you know, it's power hungry, very power hungry. Right. You can hear that Google and IBM are doing very big investments in nuclear energy. So they can phoned out AI and quantum computing the long term. And the advantage that allow us is that, you know, you can work in room temperature. 00:13:45:19 - 00:14:09:02 Unknown That means you don't have to use all that humongous amount of energy to, to put them in low temperature, and also that you can shrink as much as possible in photonic sizes. Right. And that's how she has as a whole in the long term. Right. Or more mission is like put quantum in the hands of a billion people. 00:14:09:04 - 00:14:23:06 Unknown And, you know, we're trying to make it accessible, accessible, like, you know, bring it to every data center and potentially put it in in your cell phone and scalable that, you know, it's it's a 00:14:23:08 - 00:15:00:01 Unknown continuous chain after another one, right. 00:15:00:03 - 00:15:27:23 Unknown So overall Hawaii right now is is is done is basically your high power hungry, also a GPU that is living in the center of somewhere around the world. And that is, you know, and it's not just one is a bunch of them. Like trying to figure out when you say thank you to get, you know, probably without you're burning a lot of power. 00:15:28:01 - 00:15:59:02 Unknown So overall, what we are looking to, to the QC approach is basically to boost and speed those kind of things without the constraint of energy, without the constraint of, the largest spaces, without the constraint of being a heavy object. So overall to targeting to air is like we are trying to complement it. We are trying to make it faster. 00:15:59:02 - 00:16:28:15 Unknown We're trying to make it better or in good sense, you know, your LMS and all the stuff are basically basically in a reservoir model. Right? And we have implementations of that in quantum based too. So for example, we have this product called it the narrow wave. That is we just launched it like I think two weeks ago. That is basically a quantum computer that is targeted for AI. 00:16:28:18 - 00:17:01:08 Unknown And, you know, we are trying to call it with the current approach that is all around for limbs and agents and all this stuff. But what we're looking also is to speed up without the burden of energy. 00:17:01:10 - 00:17:07:23 Unknown That's that's correct. 00:17:08:01 - 00:17:31:15 Unknown Right. But, you know, this is is becoming more mainstream also because, you know, now you have a. Well and I think it was a situation in which AI kind of put a scope of roles because, you know, everybody was talking about AI and all the stuff, and the questions came around like, oh, what is next? No. The next thing is quantum. 00:17:31:17 - 00:17:54:15 Unknown So, you know, we got a lot of us over us, and we have seen that by the investments that we have received, by the approach of people that is trying to figure out and doesn't want to miss out. Right. In general. But I will say, you know, yes, I really is a very limited resource in some extent. 00:17:54:17 - 00:18:24:20 Unknown And I will say also that it's not that we are not that that far away from having a quantum computer doing in the background things. So for example, we have we have put some of our quantum computers open to people and allow people to play around with them. And that's, that's the goal is also, you know, bring it out to, to the common user or whatever it is. 00:18:24:22 - 00:18:41:19 Unknown The thinker that wants to figure out how it works, wants to learn, that wants to be in the in the first step of of pushing this technology to. Because the more applications that we know that we can apply this, this, this, this is the better for us to design 00:18:41:23 - 00:19:22:01 Unknown it. Right. 00:19:22:03 - 00:19:56:02 Unknown Yes. And this comes from this famous physicist, Richard Feynman, that we're very well known. Nobel Prize. At some point he proposed proposed to do computation with quantum mechanics. Right. Well really happens in your in your, in your, in your classical computer is basically that you make up abstraction, a representation of a natural system. Let's say you want to compute how the wind blows from one direction to another direction, based in data that you collected. 00:19:56:04 - 00:20:19:22 Unknown And you know, there's a lot of variables that is going on because, you know, you can don't predict that a car is going to pass by there or a plane is is going in the same direction and things like that. So the best, the best way to do it is, you know, represent nature by nature. And that's that's the principle of this. 00:20:19:22 - 00:20:30:07 Unknown Right? So we are trying to represent physical phenomena that is all around us with another physical phenomena that is 00:20:30:09 - 00:20:52:23 Unknown very correlated to it. 00:20:53:01 - 00:21:21:01 Unknown So overall, you know, this is a scar that every engineer starts growing up as soon as I get out of the school. Right. It's it's the it is mostly, you know, first of all, you need to try to figure out the if your product is, is capable to get to wherever you want. So to put it in perspective, you know, you're going to try to develop IoT device. 00:21:21:02 - 00:21:45:23 Unknown Right? And you say, is this for this specific application, one of the first steps that you need to do is to try to validate if the application is open to your solution. And the best way that way to do it is just go straight forward and ask, right. And some people can tell you. Yes, yes, it's a good approach. 00:21:46:03 - 00:22:14:11 Unknown And then you go build it and you realize that it's not probably the most optimal solution. So. Right. So the, the the faster that you make that step I think is the easiest to learn. What is your next steps. So basically that experience, you know, one of the common falls is that you first go to the signage and then you go to show it off. 00:22:14:11 - 00:22:41:11 Unknown And then basically you were now the same amount of money and some amount of time. And you know, time is money and money is time. And basically you're under the underwater. So that's I think that's one of the principal things on that. So initially, you know, validate very, very easy your idea very, very fast even even without having anything built. 00:22:41:12 - 00:23:08:11 Unknown Right. And then this fake it, kind of fake it. Yeah. You know, you can put a painting, you know, you can do an emulator of that, you know, something that takes you more than ten hours go as I show up, say, this is what I'm intending to do with clearing. What are your process and your objectives and figure out if someone wants to go in the path that you are showing up. 00:23:08:12 - 00:23:44:15 Unknown And then once you collect that feedback is put together a plan. So put together a plan of, of what you want to do. Because once you get started getting a rejection or, or something like that, it's good to put it on paper, even to know how far away or from your estimations in general. Right. So, you know, it's good to put requirements in a, in a, in a document from those requirements, kind of start drawing all the specifications, thinking of how you're going to test the device when you start building, how you're going to be integrating it. 00:23:44:15 - 00:24:11:09 Unknown Because most of the times, integration and testing is where most things get delayed and that nobody takes into account, nobody takes into account. So, you know, sometimes go, people go and make very crazy lines and say, yes, I can build it in, in, in two weeks, okay. You can build it into it. But integrating and test is going to take you 2 or 3 months. 00:24:11:09 - 00:24:13:18 Unknown So 00:24:13:20 - 00:24:35:15 Unknown or. Yes. 00:24:35:17 - 00:25:06:23 Unknown Yes. 00:25:07:01 - 00:25:34:16 Unknown Yes an engineering is a muscle right. So you know you can think that once you're out of the school, you basically are a baby that doesn't have a lot of restraint. You need to start making things and figuring out how long things take and how how difficult are to do something that, for example, I always encourage my team is like, I'm not the kind of people that put them at deadline just to get things done. 00:25:34:18 - 00:25:57:13 Unknown I try to make them make an estimation of how long it takes, and then try to figure out how close they were to the estimation. Most of the times you underestimate the things. Most of the times you'll just make the things, and that's how you start building your muscle, right? So that allows you to like, you know, bear more precise and all this stuff. 00:25:57:16 - 00:26:09:02 Unknown And then you can kind of start envisioning what are the issues that you're going to find in one part, what is the issue they're going to find in the other part and what you're going to try to join them. But what is the other issue that's going to come 00:26:09:08 - 00:26:51:13 Unknown along? 00:26:51:15 - 00:28:02:16 Unknown Also, budget also watches you multiply by by two because. 00:28:02:18 - 00:28:31:10 Unknown Yeah, that's correct. And most of the times, you know, when you start like kind of planning things like you kind of do it in the big level and you start always from the big level to the low level, but you always meet the tiny, mean, minuscule steps that are the ones that really take more long time. So, you know, you can be very, you know, trying to control a lot of manipulate what you're looking to look like. 00:28:31:11 - 00:28:45:17 Unknown But the tiny details that you are not putting in your plan are the ones that usually are going to take you longer and longer. And that's where the that the that time is going. Right? So that's where you're not kill a lot of time also 00:28:45:19 - 00:28:55:02 Unknown in the book. 00:28:55:04 - 00:29:03:10 Unknown So one is for example within the one is testing for sure. 00:29:03:12 - 00:29:07:01 Unknown Yeah. Right. So you know 00:29:07:03 - 00:29:41:00 Unknown you can build a feature. You can beat something in in very short time. Now when you need to put it with the software, when you need to match it with an optical device, when you need to match it with a mechanical box or something like that. That's, that's that's the real deal, right? You know, as I was teaching some young guys in my company right now, it's like, you know, we do a lot of now in the times of God, you get a 3D model or something like that. 00:29:41:02 - 00:30:13:10 Unknown I started at some point doing paper dolls, and basically you printed a big in a big in a big format paper, and then you started trying to mock up things to see how they fit or how they interact. Right? So that, that is, that is, is is is critical, you know, testing documentation, validation of your assumptions. I think that's that's something that is also kills a lot of time because yes, you can design it in one way. 00:30:13:10 - 00:30:52:06 Unknown You can see the sign in the other one. And your assumptions need to be very clear since the beginning. 00:30:52:08 - 00:31:05:00 Unknown Correct? 00:31:05:02 - 00:31:33:19 Unknown Yeah. You know, sometimes you change a little thing and things break. And if you do not comment it properly, that that also puts you in the, in the back. Right? It kills you a lot of time. So I think overall, you know, documentation is very important just even how your build is because, you know, human brain is extraordinary but at the same time forgets a lot. 00:31:33:20 - 00:32:03:00 Unknown So while you're right, so your attention goes very narrow in developing that feature, and you were thinking in something specific next day because you were thinking on launch that, that, that that principle was gone. And if you didn't documented, you can go review cause you can go review where it goes. It goes from point A to point B, something in the middle that you miss it out because you saw them launch is 00:32:03:02 - 00:33:41:03 Unknown gone. 00:33:41:05 - 00:34:07:07 Unknown Yeah. So you know, most of the times also that it comes because you kind of your brain is still ahead in the discipline that you are trying to feed yourself. Right. And that's something also, you know that if you know when you're going to school, you are very narrow in, okay, I'm an electrical engineer or a mechanical engineer or even I'm an accountant. 00:34:07:08 - 00:34:27:17 Unknown Right. And my word is being an accountant. And I know my numbers are in this direction, not in the other one. And, you know, when you go to industry and you need to start interacting with the other disciplines, is that when you start realize what are the pain points that they have or the papers that you have from them. 00:34:27:19 - 00:34:51:20 Unknown And, and that's something, for example, that also I boost in the, in the, in the people in my team. Like, you know, I try to push them off to overlap other disciplines, understand what is going on in the other side. So you are not just narrowing what you are, but also can understand from the point of view what the other people is expecting from you. 00:34:52:01 - 00:35:00:22 Unknown So this becomes a very a situation that, you know, I understand that you're doing this, but this is my, my paper 00:35:01:00 - 00:37:38:03 Unknown and that's how progress is done. Right? 00:37:38:05 - 00:38:05:05 Unknown That is more visible nowadays. You know, that we are in the era that everything is becoming an iPhone. Right? And, you know, engineers are, but in designing for users but are very good in designing for engineers. So, you know, kind of is something that has been neglected for a long, long time. But now that everything is trying to become an iPhone, easy to handle, as soon as you get all the balls, you know, you know what is going to happen. 00:38:05:06 - 00:38:25:08 Unknown You know, you see a light out there, you push it out, push the button, and it's very intuitive to follow along. And and nowadays it's kind of coming more, more strict and directive to the initial faces too, because that defines a lot of how the product hits to your, your potential 00:38:25:14 - 00:39:33:18 Unknown customers in the beginning. 00:39:33:20 - 00:39:54:03 Unknown Yeah. So yeah. And you know, it falls again from, from from the, the situation that validating your idea is very required. And you know now you these are these new axis that is basically how are you making your user with your product. So that needs to be now in mind of 00:39:54:07 - 00:40:19:01 Unknown of it. 00:40:19:03 - 00:41:00:14 Unknown So and I'm going to try to put a blanket statement in all of them. You know as I mentioned before was validation of your idea because that that can open you, you know, to a very great opportunity or total fiasco. Right? Even so, to get great bonding or be a starving in a corner trying to collect pennies, the other ones is, you know, for sure is like a try to go at least have a happy part and a sad part, right? 00:41:00:18 - 00:41:30:12 Unknown You know, planning is something that ideally is ideally every company is trying to do in some extent, but the level on where you're planning is makes a huge difference. Nobody has a crystal ball. And, you know, I think planning is in the fact that, you know, you just need to see it yourself with your teammates and say, okay, this is what we're going to do. 00:41:30:13 - 00:42:07:18 Unknown We're going to deliver on this and this. This is the steps of how the cadence of things are going to come along, because that allows you to kind of put in synchronization, run in synchronization, absolutely everything. And if something doesn't come as respected, you can take the risk. It very easy. Right. In that and then you know, the moment that it goes for that you have something that is, is kind of validated, that you have built it, that in some extent it's getting some traction. 00:42:07:20 - 00:43:49:08 Unknown Try to to stepping out from the design phase to manufacturing is always a challenging thing. And that's probably, again, another Pareto analysis. Right. Because yes, The Shining, it could take you, you know, let's say x amount of time. It basically, you know, you say I built it in 2,020%, but really the 80% of work in manufacturing, how to make it very easy to manufacture is also an issue that is also neglected to and most of the time very out of the of the of the planet. 00:43:49:10 - 00:44:16:18 Unknown Yeah. So I've met funders, you know that you know when they are just starting like you know, they go very straightforward even to the, you know, try to automate absolutely everything. Probably there are a lot of things that doesn't need to go in that high level. Right. And that's what you mentioned. Like, you know, I can also feature that is sending an email to my users every night and boosting up. 00:44:16:20 - 00:44:47:03 Unknown And your users can be like just swiping it out of the of their phone. And they're in boxes and they don't care. Right. As far as the the real deal that you're offering to them is accomplished. They, they they, they would be happy. So that, that that for sure is something that, you know, the balance the focus. It's, it's, it's something that you need to keep in your planning and novel react. 00:44:47:03 - 00:45:08:08 Unknown But also don't leave it too loose. 00:45:08:10 - 00:45:34:14 Unknown Yeah. That's correct. And also, you know, something that I say and that is is it most of the times you don't learn from the right answers. You learn from the wrong answers. Like the more you are hitting out the wrong answers, the more the more you're going to be closer to a more possible path. 00:45:34:16 - 00:45:56:23 Unknown Correct? 00:45:57:01 - 00:46:00:12 Unknown The a pleasure. Thank you so much, Andrea.