The AR Show: Paul Powers (Physna) Teaching Computers to Understand 3D and Innovating at the Speed of Imagination

5:03PM Oct 4, 2021

Speakers:

Jason McDowall

Paul Powers

Keywords:

people

technology

world

understand

search

model

fisma

part

company

augmented reality

create

spatial computing

thought

learn

build

software

opportunity

enterprise

mindset

geometric

Welcome to the AR show right dive deep into augmented reality with a focus on the technology, the use cases and the people behind them. I'm your host Jason McDowall. today's conversation is with Paul Powers, the founder and CEO of Physna. You may have caught Physna in the news recently, because they raised $76 million over the last year, including 56 million earlier this summer from Tiger Global Sequoia and others. Paul and his team have created a new way to search and compare 3d models in a way that is faster and more effective than what's come before. This ability to recognize and understand physical objects combined with things that are online database of 3d models, creates the potential for them to become at the Google search of AR. Here Paul describes how they are enabling computers to understand 3d in the same way they understand text.

What this is able to do something very similar, but it does it with the geometry. So if you were to take a vase, smash it up against the wall, when you had tons of shards of ceramic leftover from that, we could take one of those shards and match it to the vase, and vice versa. But we could also tell you, these shards can be built into a base if you put them together in this exact same way. And that was when we realized we were onto something truly unique, right? We kind of bridged the gap between the physical and the digital allowed the digital technology to truly understand physical objects.

In this conversation, Paul describes his unique educational experience and his path to becoming a tech entrepreneur. We talk through the technology and we discussed leadership, entrepreneurial mindset, a different way of thinking about goal setting, and aligning and motivating others. He also shares his perspective on the potential impact that glasses based AR can have on the human experience. As a reminder, you can find the show notes for this and other episodes at our website, the AR show calm, let's dive in.

My daughters are back at school now, after a nice summer break. And they were actually really excited to go back with this beautiful summer vacation. And I think they had enough time away that they're excited to see their friends, and really get back into it. What were what were some of your memories heading back to school when you were young?

was a pretty short trip because I was homeschooled. So I didn't have to go very far at all. Actually,

what was it that led to your homeschooling? Well,

we didn't live in the best neighborhood. So the local public schools were pretty bad and private school wasn't option for us. So it just kind of came down to that.

You grew up in what city? Cincinnati, Ohio, Cincinnati, Ohio. I don't think many folks associate Cincinnati, Ohio as having a bad neighborhood. But I guess like any other city, there are not so nice pockets.

It used to be known for that. But it's definitely gotten a lot better in recent years. But when I grew up, there were definitely some parts of town that were not they were a little rough.

Yeah, actually is a brief aside the summer we spent eight weeks on the road driving around the country. And we got to spend a couple of nights in Cincinnati. And we went down to Finley market area, which was beautiful, but clearly was was also a neighborhood and transition. I've one of the things that really stood out for me about Cincinnati since culturally was there are so many gorgeous murals on the side of buildings and walls. It just really stood out as a beautiful, beautiful addition to the city.

That's an observation that's really nice feature about our downtown area.

Yeah. Is you had now this opportunity to study at home. How did you focus those studies? or What did you focus on?

Well, not long after I started being homeschooled, my father was in a car accident and had to have a series of back surgeries he was laid up. And in the meanwhile my mother had to go back to work. So for a long period of time, and essentially from seven or eight, all the way till I was 16 or so I was primarily just self taught. And so that gave me the opportunity to focus on things that I thought were interesting, and ignore things that I didn't. If you're homeschooled, at least this is how it used to be you had to take a standardized test every year to prove that you were learning at least the basics, but they only really covered reading comprehension and math. So you didn't have to read certain English literature, you didn't have to know certain bits of history so much. And I got to kind of essentially just explore the areas that I thought were the most fun. And for me, that was science and technology. I was a bit of a history buff in a way but not for traditional, you know, the nutritional curriculum more just what fascinated me were these great men and women in history who stood out by figuring out how to do something fundamentally different like Einstein, Galileo, but yet people like that, or even from a business perspective, you know, Henry Ford, and they really transformed the way that the world works in a fundamental way and it really finished My early mentality I think one nice thing about being homeschooled is that you don't have you don't equate success with how you do compared to your peers, because there are none. So when you're looking historically, you know, when you have you're basing your own future and success off people you've read about, and well, if you're reading about them, they've obviously done something big. And I had my very young age really wanted to do something big. And because I loved science and technology, I wanted to do it in that field. So I really focused in on that area

as an eight year old or nine year old. How did you motivate yourself like the if your parents weren't writing your, you know, writing you throughout the day, or your teachers? And you knew that really, the bar was just math and reading comprehension? What was it that kind of kept you going back and reading those histories and exploring that science and technology,

curiosity, I was always in style, I'm a very curious person I like to I like to learn. And it was just fascinating to me. And I always thought of sciences, like magic in a way. It's, it's amazing. I remember having my first chemistry set, and just fascinated by it. And I remember, when we got a computer for the first time, and things that you were able to do with it, it was just game changing in so many ways. And I thought, Man, if I could do something that could have that sort of an impact on society, then I could be one of those rare people in history that, you know, do something that better the lives of future generations. And that really early on became like a mission for me. And it's probably due to a lack of outside stimuli, right? I didn't really have a lot of them, I had friends that I knew, but I didn't spend a lot of time with as many people my age, and I had a lot of time to myself, and we didn't have a lot of, you know, gaming systems alternative things to do. So what am I supposed to do during the day? If you're curious, if you have a lot of curiosity, and you have the opportunity to read and even watch documentaries, to some extent, to just to absorb information, it's surprisingly easy to do under no circumstances.

So in many ways, you begin to compare your future self against the giant minds of an accomplished men, women of history. Yes. And as opposed to Jimmy down the street, you know, who can kick the soccer ball of a better or that better attention in class or from from peers? Precisely? How did that inform the path that you took from that early self directed learning to where you are today

defined it, I think it was a defining part of who I am and how I grew up. I later on went to, I was able to get into Harvard pretty young. I was one that first semester, but I was there for it. I went there at 16, I was studying astrophysics. And I decided to leave because I had an opportunity to go on an exchange in Switzerland. And I was becoming kind of disillusioned with the idea of becoming a direct or an individual contributor, so to speak, whether in science or technology. And I thought that the better way to leave an impact would be through business, I didn't really know exactly how to do that. Because I was very academically driven, I wanted to have an academic accomplishment. And I thought this year in Switzerland would be interesting, because obviously I hadn't traveled much before that. And it'll be a really good period for me to kind of reflect and think about exactly how I was going to get to that goal. And I learned the German language relatively quickly, because when you're that age, you can learn languages faster, if I tried learning a new language, now it will be a nightmare. But at the time, it was, you know, your brain is still developing. And so you can build those neural pathways a lot faster. And so I was able to learn language relatively quickly, comparatively speaking. And I applied to go to MIT and I got accepted into MIT. That was to go into physics. I was also considering studying computer science. And but it really at the end of the day, I ended up narrowing it down to either business or law that thought those two degrees, you know, there, if you're going to be a startup founder, the degree you have really does not matter. What matters is who you are, essentially, as a person, and then what you're able to endure and and to some extent, the people that you've met and the perspective that you have, and I thought living abroad would give me an interesting perspective. And if I were to study law in a different language, that would give me a different, that would be fairly unique thing and helped me meet a lot of interesting people. And that's the route that I took. So I ended up moving to Germany and lived there for about another nine years. And the 10 years of total include Switzerland, and ends up passing the German bar. And immediately right, I came back and started a company. It was never about the profession of law. I actually be quite honest with you. I hate the profession of law of being an attorney. It's never had any appeals to me whatsoever. But the experience of going through that did prepare me for being an entrepreneur and bring back full circle. Being able to Learn on your own, made it possible for me to do that. Because the strategy I had to take to pass the bar exam in a foreign language was that I had, I was learning a different pace, I had to look up tons of words that were that everybody else in the class, they were all German, they knew the words, and they were simple words to them, and on completely foreign to me. And it took longer in some senses. But I also knew that I could read fast with them. And then people typically talk, and I could absorb information very quickly that way. And so I, the Germany doesn't have a requirement, at least at this time, it didn't have a requirement for you to attend class, all they cared about was you passing the exam. So I went to almost none of the classes, I'm very, very few. And the rest of the time I just study, I just read books, that's how I ended up being able to go through that. And I think that that experience and the competence that you get, by being independent, realizing that if you set your mind to it, you really can do just about anything, you're not a victim of circumstance, you can power through just about anything you want to if you really are intent on accomplishing it. That was, I think, possible, because of my my upbringing. And it's an essence of a lucky coincidence that I, you know, had had to go through that as a child that made me far more independent as an adult.

Yes. I can only imagine having two daughters that are kind of the elementary, middle school years, how much self empowerment you had, being such an independent learner and such a curious, self motivated person was extraordinary. How did your parents deal with this, the precociousness of your youth and going to college at 16. And, and going off to Europe after only a semester,

they hated it, they thought it was stupid, but at the same time, they knew that I really wanted to do it, they didn't have enough money to really pay my way through this, I had to work at the same time. So I was also fairly financially independent. And so it wasn't, they weren't the kind of people to hold me back from doing something like that. They gave me a fair degree of autonomy growing up. And that's another very precious lesson that I learned, which is if you give people space to do what they love, they thrive. And I think that's an important leadership skill, right is the ability to know when to give people the right amount of autonomy. And typically, that's a lot more autonomy than less. So they dealt with it fine. I mean, they didn't, they never thought that it was a safe route to go. And they certainly didn't want me to be gone for, you know, years. But they knew that it's something that I had really set my mind to, and they accepted it. And I really appreciate that about that.

Yeah. How did that experience in Germany lead to FISMA?

Well, it's, it's interesting. So it's kind of both experiences that led to FISMA in a weird way. So the brief stint and astrophysics again, just just a semester. But you know, before that, I had spent years reading books on astronomy, on quantum physics, on string theory, the whole area of just fascinated me, everything to do with theoretical physics, or astrophysics, just really, really fascinated me. And so there was a certain mental model that I had for how you can sort of find relativity and dimensional settings, right, if you like, spatial reasoning, if you will. And the problem that I found came from law school where most technology, this plagiarism search tools, right, so if you turn on a project, they can find if you stole something from, you know, online, the same thing is true for images, and video and music. And it's really easy to pick up on, you know, trademark infringement, or somebody stealing your song or whatever. But all that's two dimensional data. And so when what I was finding was that, you know, the majority of the problems were, especially economically speaking, coming from patent infringement, people who were basically recreating things that you had designed and you had the intellectual property for, and there was no way to reliably search against that. And so I remember people having discussions along the lines of Well, we know that XYZ company, downloaded this model, probably from ABC website, whatever it was, but they couldn't find it. They couldn't prove it, because it wasn't saved, as you know, what, as XYZ model, right? It was to say to something else, and so I remember asking somebody wants a Why don't we use the the, the image recognition algorithms, why don't we use that tool that we have to identify that and they said, well, that we've tried doesn't work because it orientational throw it off. And obviously, it's not going to necessarily be an exact same orientation. And it's just this is not this also that works in the classification level. If I show you a picture of a table, it'll say it's a table it won't necessarily tell you it's this exact table of messy exact words. So that made me realize that there was a, there was a huge opportunity. And the more I looked into it, the more I realized that, even though there were some things called geometric search and shaped search, they were all based off of that same methodology, they were all based off of these 2d primitives, if you will, this, this notion that you have to search based off of the metadata, which is really, you're not social geometrically, you're searching based off of data that's entered by the person who designed the model, or you're searching off of a 2d representation of that model. But then you have the same problem, you don't get to the classification, or the category level, you can't match precisely, very precisely, even if you do match very precisely, the big problem that you still have is you can't match different types of relationships. So an example I can give you is, let's say that you had a model of Donald Duck, and a model of Mickey Mouse, and you swap their heads, right, so none of those search algorithms would work anymore. Because they would say, well, this isn't the same geometry, you need to be able to find any subset of data within any other subset of data to be able to find parts inside of parts. And the way to approach that was their math. And there were a lot of ideas for how to do it, it was kinda like inventing the light bulb, we spent millions of dollars and many years trying out countless theories on how to approach this problem. But reaching back into the some of the concepts from astrophysics and other areas that I was just obsessed with, earlier in my life, gave me some ideas for how to develop these algorithms differently. And that eventually led to the breakthrough, where we were able to create what is not fitness or physical DNA. And what we do is we are able to take any data from any format, or just about any format, any format that our ingestion client can handle. But you know, that doesn't really matter. It's any three dimensional information from a scan from a model from something to 3d print from whatever, we can normalize that and turn it into something that for your, for this for software is somewhat akin to text. So when you use Google and you do a search, you don't look up the entire URL, it's stupid, that's defeats the point of going to Google right, you're looking for maybe a very specific thing, but you're gonna look for it a little bit differently and online with a portion of the text. What this is able to do something very similar, but it does it with the geometry. So if you were to take a vase, smash it up against the wall, and you had tons of shards of ceramic leftover from that, we could take one of those shards and match it to the vase, and vice versa. But we could also tell you, these shards can be built into a base if you put them together in the exact same way. And that was when we realized we were onto something truly unique, right? We kind of bridged the gap between the physical and the digital allowed the digital technology to truly understand physical objects. And the amount of value in that is hard to even wrap your mind around. It's, it's so much more than that there's so much more information in the world than there is knowledge in the world. And the differences knowledge requires human intervention, no, some of you, if I were to describe a pen that I might have here that you can't see or that don't have over there physically with you, I would have to spend a long time with a lot of text trying to describe this pen in a very detailed way through text, right, if you had the pen, you would have all the information instantaneously. And you would never really with a video, even with any two dimensional data set, you wouldn't really have all the information about the pen that I happen. So we thought if we were able to, you know, essentially digitize that and turn that into every aspect of that geometry into a code snippet, then as far as software is concerned, we can make it searchable and not just searchable but a computational asset, right? So you're able to compute spatial relativity, this of how things are related to another geometric relativity. And so we can show you how any and all things, physical things are related how one screw that you might use for your motor, or your engineer car might also be something you can use on your lawn mower or your door and, you know, etc, etc. But there's not really limitation to what kind of objects those are because it's all based off of mathematic principles. But we did find, though, was that the use case that I had in mind originally right to helping people protect patents was the tip of the iceberg because of that, right? So there were hundreds of use cases. And these aren't things that we thought up these were use cases that people asked us about. They said can you do this, can you do that? And we eventually realized and it took us a while to realize this because we always thought that you know that one idea of If the next idea was better than less, they'll eventually did realize, hey, the right strategy here can't be that we build hundreds of products for each application. the right strategy has to be that we build a platform through which other people can develop applications that were previously impossible, building a previously impossible technology to solve a previously unsolvable problem, and giving everybody the opportunity to do that with this core technology. So then the question became, how do we get there? Well, starting a platform business is really difficult if you don't have awareness and momentum and evidence that there's an economic reason to work with this technology. So we decided on a two prong strategy or a three prong strategy, including the platform, and that is one to create an enterprise tool. So companies have the ability to search against their own database, if you're an engineer, you kind of like autofill. In a sentence, you can find parts that were designed in the past by you or anyone else, and you're not recreating things from scratch and supply chain, you know, alternative parts and where there are duplicate. So you can consolidate your orders and save massive amounts of money, and headache and remove your dependence on so many different vendors that you know, that are all actually just telling you the same thing. And then in the field, we have a tool that allows you to take a picture of of a part, and we can tell you not just what category it is, but because we go from 2d to 3d, and then from 3d to this DNA, if you will, that we break down, we're able to be very precise, and tell you this isn't just a whatever is this very specific, whatever. And by the way, this is how you can use it. This is where you can go these are alternatives. These are slightly different alternatives. And this is how they would work in that context. And when it comes to AR, what we realized was that, you know, there's a coming wave of the technical user experience, right? If you if you ask anybody, how do you think technology was going to look 1020 years from now? Everyone, I have a slightly different answer. But what you'll find is there's a common a common thread. And that is that everybody seems to describe a world in which technology is more immersive, right? It's something that you live in are not just bound to a rectangle in your pocket. And so for that to happen, technology has to understand the world that you live in, which is three dimensional, you have to have spatial geometric reasoning for that to happen. And software, even with sensors is woefully inadequate at understanding what's around you, they might be able to say Watch out as a wall, but they won't necessarily be able to tell you, hey, that shape over there is, you know, this very specific part. And the second thing is you have to be able to well augment that reality, you have to be able to insert things into that reality. And so what excites me about the augmented reality space is that we've been able to add that into the product. And so there's a consumer product as well, I mentioned the enterprise side, the consumer product is called thanks, th amgs. If you go to fangs calm, it's the world's largest search engine for for cab models. But it's also the only public geometric search engine for parts. There's a collaboration aspect to it as well that we have an alpha right now. And if you go to the site, you can sign up for beta, which is almost like a GitHub for 3d designs. But one of the things that we have built into it is the ability to take a model from any format doesn't have to be in augmented reality file, it can be literally any three dimensional format that that at least of the dozens that we accept, and you can without requiring an app on your mobile phone, you can open that in augmented reality in your environment. And so you can actually see if I were to buy this thing, how would it fit here? If I were to 3d print this, would it work? Right? What size when I have to make it etc. And so there's a lot of utility there. And that's something that's not otherwise available any meaningful way. And it's something that were brought to the enterprise, or we're bringing to the enterprise side now as well. And it was a long answer to your question, but

that's totally fine. I'll get kind of jump back on to the earlier part of the recognition that some of the insights that you had gained from studying astrophysics and your general interest there in the problem of trying to identify from a patent infringement perspective, that one part here and this patent is the same as that other part that other patent, what is the the sort of geometric understanding, you know, that ultimately the language is math, but it's an understanding of edges or services or points or is it something else entirely?

It's all of the above to an extent and I mean, all the above is known. The way that we do it is we've actually patented and or nationally, but it's not just one thing. It's a series of algorithms. And we essentially use a series of these sort of like discrete embeddings with that are Are these very specific types of geometric patterns essentially, that we then analyze, and they create a numeric signature or mathematic signature for shapes and sizes that are extremely unique. And so if there's a duplicate, there's no doubt if we say two things 100% identical they are, you can't, you can't get a false positive, I mean, not in the main algorithm, there's there are other, you know, you could do it a scan, if you have the settings on the absolute bottom level, then yeah, maybe you can get 100% of two things are very similar, but not for regular matching. It's very, very precise. And so it's a very unique way how we break down the data that we get. So we don't just take the data. And as it is, we take it through a process where we, we analyze that data, we normalize it, and we don't really care how it comes in, if it's tessellated, whatever we, it doesn't matter how you've sent us the model. This is how we're going to analyze what you've sent us. And we're gonna create a mathematic profile for that and every aspect of it, so you can show you how it relates to anything else geometrically.

This is a very literal association to the concept of DNA, that every piece has a unique code that identifies that uniquely, yes. And then once you got that computers know numbers, they can compare one number to another. Exactly what are the implications? So if somebody were to run a search, for example, using some alternative method, assuming that they had some ability to teach the computer how to understand 3d, versus the method that using what are the kind of the practical implications of being able to successfully run a search?

Well, there there are a lot I mean, so the practical implications are very, very similar to those being able to run A to D search, right? So if you're an engineer, people don't store part numbers, and the most obvious way that you would think, right, they, they store them, and very different types of systems, and sometimes no system at all. And then it takes so long to find these models, because you're not searching for the medium with the medium, the great thing about searching for tax with tax is that you're searching for tax with tax rates. But when you start searching for geometry with tax, it's difficult, it's kind of like, if you were to look up cat, they know that the word cat and Google Images, you would find cats, if you wanted to find your cat, very specifically your cat, it would take you a really long time. And no matter how descriptive you got, it would be really, really difficult to find that gap. Because there's just so many cats. And what if you didn't save the file as cat? What if you save it to something else, so the best way to search for an image would be through an image. And the best way to look for geometry is their geometry. And so that's really, the crux of what this does is it allows you to search in that native medium, we understand the world and its natural format without requiring a human translator. And the beautiful thing about geometry, unlike really anything else is that, or I should say math in general, is that it's a universal language. I mean, we see all the time companies or users on things in particular, right? Finding results for their part, or something that would probably go into or that would go into their part in the community, from designers or from other sources that are, you know, in a totally different country, and it's labeled as something totally different language. And they never would have found it the traditional search. So the implications are every bit as big as text search, it changes everything. And if you think about AR specifically, you know, this is an incredibly important thing. So if you have smart glasses, right, and let's just imagine we're a couple years in the future, and we have, let's say it's Apple glasses, and it would be very unfortunate if it let's say that the glasses only had LIDAR on them, you know, or even if they did have a camera, that camera, like we just talked about traditional type, 2d image search is very imprecise. But if you don't have light, or you don't even have that, so I mean, it would be very difficult to identify for your smart glasses to be all that smart, because they could just tell you about the spatial surrounding. They couldn't tell you what any of that stuff was. But imagine if you could copy paste things. What if you, this may be a stupid example. But I think it resonates with a lot of people if you were to see a boat on a lake, and you can't get to the boat because it's the lake. But you could scan that right, essentially, through depth THROUGH THROUGH YOUR LIDAR or whatever. And you could, we could match that to the original vote. You could say, okay, cool, we found the model. And now that because we can make it presentable and augmented reality, you can now put that boat on the shore and look at the 3d model of that boat. As if you were out there on the lake. For AR to really take off. You need these core technologies, you need to be able to do the kinds of things with the physical world that you can do with packs, otherwise it's AR is going to be more of a I don't want to downplay it, but it's not going Is there we're not gonna have the level of utility from an immersive technical environment that we currently have from our two dimensional software environment. And that's what we're the bridge we're trying to get. Yeah. Or the gap we're trying to bridge.

Yeah. So I think that one of the Northern word immersive right now I'm having I'm having personal issues with the word metaverse and immersive as relates to augmented reality smart glasses. But aside from that, the notion that augmented reality, this vision that we all have for smarter glasses that will bring on a face that have some understanding the real world, they have primarily utility because of their understanding of the context in which we are currently in. And that context comes from, of course, some perception, understanding of the 3d space that we're in, and all of the objects within that 3d space, because that's the thing we need help interacting with, generally, right, we could have some, the other system could be also smart ish, in, you know, some sort of enterprise IoT setting in which it's telling us, hey, this is what I am. And I've logged this in the database that this beacon is coming from this particular object. And that could be one level of early understanding. But really, the goal is to use the sensors on the camera to perceive and then understand. And that's exactly what you're talking about here. One of the things that I think is, is really extraordinary about the structure in which you are trying to articulate this, the thing itself, the things themselves, is not just that each item has a unique code that you can identify your exact cat amongst all potential images of cats in the world, or three miles across the world. So you have a level of understanding of similarity, that this is 95% or 98%, or 20%, similar to the thing that you are interested in. And that that relative to the whole understanding, I think, is also really interesting and powerful.

It's a huge part of what makes Bhisma unique, right? Because it's not just saying this is that, of course we can do that. But we can also tell you what's different, we can show you comparisons of this versus that. And what's more interesting to me is that you can take we can do not just one part by one part level, but we have a tool called component finder, this is not thanks. But it's in our enterprise product. Where, let's say that you upload a model, 3d model that has all kinds of components inside of it, you know, some kind of maybe it's a bicycle, maybe it's a, maybe it's an electric motor, or a traditional motor of some sort, or whatever, we can identify not just what that is what it matches, but even if that's just this one single file, there's no bill of material, there's no file hierarchy at all, there's just one single file, we can tell you all the parts that fit inside of that, and show you exactly where they go. So it actually highlights, here's where everything would go, even though there's no file connection between these things whatsoever. And they might be in totally different file formats. And so that starts to solve a lot of problems. When you add augmented reality, especially when we get the smart glasses. That type of functionality can become eventually fully automated. And it can change how we interact with our physical world in a very proactive way without us having to intervene. And that's what excites me about the platform, because you can go on things right now and use it and you can go, you know, go to our enterprise software and use it. And they serve a pretty wide variety of use cases that as we've talked about, but there are so many more things that you can do with this technology, that we as a company, we can't build out all those tools. And so what excites me about this is that we have this, because we have this core technology, and we are getting a lot of traction on things. And we are seeing a lot of adoption, written interest and momentum on the enterprise side, you now have this, this perfect storm, if you will. And we've we've already started having inquiries about this from purely tech companies, can we use an API to do to use this software uses this core capability that you have to do something completely different. And it really excites me, because it can fundamentally change how we think about both software and the world that we live in. And they too can become a lot more synonymous,

as you've worked through this understanding of the initial problem space, which was around patent search. And the solution to that problem space. And this, this growing awareness and appreciation for how broad how broadly applicable, the solution is to all sorts of related problems. You recognize that there is a Google search like opportunity, moving from, you know, searching the textual world, to now be able to search the 3d world. And, you know, as you noted, as we move towards the spatial computing and this notion of, of understanding our real world, whether that's for product design, or development as you're focused on with enterprise or just generally interact with the real world from a consumer perspective, it's a massive opportunity. Suddenly it doesn't happen that often this kind of this new way of teaching some assistive device, a computer, how to understand something, or any sort of electronic device, how to understand it, and how to help us as humans, whether it's communicating better or understanding and then helping us make better decisions. So giant opportunity in front of you how you kind of began to break it down. But how do you? How do you wrap your arms around this giant opportunity, while at the same time managing these massive tech companies? which I'm sure are, yeah, banging on your door, in various different ways, physically. overtures to your teammates, you know, trying to steal them, or whatever it happens to be, how do you manage all that?

That's a good question. Momentum managers that to be honest with you, we are moving and growing so fast, that we don't really have an issue, we have very, very, very little turnover at the company. I mean, I don't, I don't have no statistics off top my head. So I don't want to want to cite anything correct. But far, far less than you would expect for a tech company. And that's true for throughout the company. But even among engineers who this day and age can be notorious sometimes for turning over very quickly. We don't have that here. And that's because everybody knows that what they're working on, is really, really cool. It's not just that it's cool. It's, it's powerful. It's powerful. Because it's not just an app, it's not just a tool, it is a tool. But it's also a tool that lets you build other tools. And it's something that and when you build those other tools, these aren't, this is a new category of tool. This is a category of tool that does things that were simply impossible before and the use cases are just about as broad as you can imagine, because we live in three dimensional world. So anything physical, is potentially a use case. So how do we wrap our minds around it? Well, we're very focused on our products, right? The Enterprise products, if there's an enterprise and then thanks to consumer product, and building a bridge between the two, so that if you're an enterprise customer, you can search against supplier databases. And what's great for the suppliers is now you have people who never heard of you before, but because they are connecting these technologies, they know, hey, this company that I've never heard of has these parts that I've been looking for, and I have haven't been able to manufacture them or my current vendor that I use is too expensive or whatever, it overcomes a lot of different barriers. So so we're working on connecting those two products, but at the same time, they're both in service of that one, broad vision, which is empowering the next generation of software, and breaking down the barrier between the physical and the digital for a specific reason. And that is that when you do that, you enable an era of technology that allows humans to be more. And I think that that's one of the most valuable things you can do, right? I mean, you just look to democratize elements of 3d, it's notoriously difficult to design something in 3d, and it's even more difficult to actually manufacture it. But the combination of is nothing actually allows you to, you know, in theory, go through that process extraordinarily quickly. And starts to really democratize the creativity and the physical world and allows you to innovate at the speed of your imagination. But I think that anything that can unleash unleash more human potential is a valuable tool and the connection for me, as you know, growing up as a kid, I wanted to develop something that moves humanity forward, I wanted to be a contributor, I wanted to create something that was a major shift, and how we interact with the world around us, but specifically through technology. And physics, is turned out to be unbelievably more powerful than I ever thought it would be when we were starting it and it's given me To me, it's, I would say, my purpose in life, I really do feel that strongly about it.

That's powerful, is you can reflect on this specific experience and the momentum that you've generated. I mentioned in the intro, that you have a lot of momentum and momentum on the financial side as well, right, you raised 20 million late last year to raise another 56 million earlier this year. Earlier this summer, you announced a lot of financial momentum from some really amazing professional investor types as well. That firewall here for FISMA is really moving. But this experience is built on the foundation, not only of your earlier studies, but also your earlier experiences, building and starting companies. And so I think you started your first one at 14, is that right? 16 at 16. And what was the focus that first company is sold

off just as a tutoring company. I was tutoring people at 14 but i don't i wouldn't really accept that. You could call it a company. But I would I would say that it became more of a true company close to the age of 16. And then eventually, we're sort of studying, it became more of a true company in the sense that I was hiring other people to do the work and then over time it transformed into a little bit more Have an enterprise focus, the focus was on providing training to enterprises and specific areas. And then we became more and more of a product type company, because we started to work on this platform that allowed you to do it virtually and remotely. And this is, of course, that would be nothing new today in the age of zoom. But at the time, it was something that wasn't very, very common. And so that started that that was the focus of that company. And, you know, there were several other companies besides that, that I started, that all eventually became product focus, except for one that was product focus. And then, you know, it became a service company just for financial necessity, because the product partner that we had, they went out of business, and that was our sole source of revenue. It was only a few months old, but we had 20 employees, and we ended up having turned to a service company, yes, anyway, but physically, from day one has always been a product company. And this is certainly the first time I've gotten the company to the new kind of tech company with to have so much potential and so much momentum, with such a high ceiling, if any. So every day is a learning experience for me, but the way around that is surrounding yourself with great people who are experienced in areas that you're not, I think that, you know, part of my job as CEO is to hire people who are better than me in every way, shape, and form. And they really just keep my mind focused on the big picture of the vision and making sure that all the wheels are turning and that people's experience and their enthusiasm comes to life and creates a powerful organization. And that's what we've seen happen. And so I'm I'm very excited about our future here.

Vision center, cheerleader motivator, what have you kind of discovered as you've gone through your own, or you've helped others go through the process of establishing a successful company, what, what do you think are some of those most important attributes for success as a founder, perhaps maybe on the other side, the biggest risk,

I'd say, the biggest risk, and the biggest prerequisite are almost same thing. And that is mindset. It's really easy to learn certain skills, right? It means that you can read a book and have the skills, you can work somewhere for six months, a year, whatever, and pick up a lot of skills. It's hard to have the right mindset. And some people have a much easier time with than others. But nobody has an if they say they do, they're like, nobody, nobody 24 hours a day has the perfect and entrepreneurial mindset. But training how you think, is crucial. If you're going to survive in business, especially if you're going into a startup because it's brutal, the risk is extremely high. But when you realize that you're in control of that risk, and you're real, and you're so passionate about what you do that it doesn't matter, because the biggest risk to you would be to not take that risk, right? Because that'd be that'd be the definition of failure. At that point, you're headed in the right direction. The reason why is that, if you have the right mindset, if you have the right drive, you will easily pick up the skills skills are easy, mindsets, not. And, you know, one thing that I've learned is on the skill side, though, you know, outside of mindset is that it's incredibly important to understand people and what motivates them and, and to be that cheerleader, and to give people that autonomy. And that's something that, you know, it's not always easy to do, especially when you're growing the company very, very rapidly, you have to realize, like, Hey, I have to delegate things very, very quickly as I build up this team, and I have to train them, but then I have to trust them and let them go and run with it. And that can be tricky. It's easier said than done. Because especially if you've worked on something for a few years, and you're very, if you're a bit of a perfectionist, and obsessed with details, which I think I am in many ways that's that, you know, letting go of an era can be difficult, but it's crucial, because if you don't your company can't grow.

mindset. How do you advise? How do you advise young founders or any founder, any age to to develop? What are the skills? Or what are the techniques that they use to develop their mindset?

That's a great question. So I mean, there's many different ways to develop the right mindset. But there are a lot of books on the topic. You know, one book that I'm reading, right now I recently finished, is from Fred Kaufman, he's, I've had an opportunity to meet Fred, he's a fantastic, personally, a genius when it comes to the area of leadership. And he wrote this book, I forget the full name of the book off the top of my head, but it's 2018 release and it's called transformative leadership, or trying to destroy a trend a transcendent leader or something along those lines. And the book is all about getting other people to share goal alignment with you. And the first thing you have to do if you're an entrepreneur is have a very, very clear goal. The worst thing I see a lot of people do, and I think this is why most startups don't succeed or even get off the ground at all is because they people in our society are trained to think in the wrong way. They think, Okay, well, I want to go to college so I can study this thing. So I can do I can become this thing later. And then after that, maybe I can do this other thing. And then after that, maybe I can do this other thing, but they start, they're starting at what's the next available option? what they should do is the opposite. They should say, Where do I want to ultimately be? And then work their way backwards and figure out what's the best way there. And so a lot of people ask me, Paul, why the hell did you study German law if you knew you're going to be a startup founder in the United States? And the answer was it honestly, isn't that hard to figure out why that would make sense. If you work your way backwards, if you work your way, that traditional way, it makes no sense whatsoever. That's not a common path. But if you think about the advantage of something like that could give you at least for me, it was a path that made just as much sense as any other but actually more than several ways. And that's why I chose that path, not because I wanted to become a lawyer. And a lot of people don't understand that they say, like, you must have had a change of heart. No, I didn't care that that road, if you will, traditionally goes to this town, I knew that it was a really good way to get to this other place. And we apply that when we go to the grocery store, right? When you're when you go somewhere, you don't think I'm going to turn on the prettiest Street, oh, there's a stop sign, I got some going back home, right, you don't do that. But in our lives, we do. So if you can find that goal, and plan that way, then you're out, you're headed for success. And then the next skill. And I would say that's a big part of mindset. But that's also a skill. And the next skill. And there's a lot of mindset involved with this skill as well, is inspiring others, to see the same goal, everybody's headed in the same direction and wants the same thing, and you have good people, you will succeed. It's when you don't have that, right that you can't succeed. Everybody in the company wants something different, the company will not thrive, it's impossible. I don't like even the word impossible, it's safe to say it's incredibly unlikely that you're definitely not optimized for success, because everybody's trying to get something else out of it. And that's why to me one of the biggest times if this is, you know, absolutely no politics, politics are self serving. And you know, we're as a company, we're a collective with a with a vision we have, the place you work should give you a sense of meaning. And I don't think of working at Fitness as a job or a career, I think of it as a calling. And we look for that when we hire people that they they want to do something that will have meaning that will impact people's lives that will change how people do things. And not many people feel that strongly about that. But it's super important, because when you get that goal in the company, I'll succeed. So I'd say those are the first two things that are very important to learn as a young founder, but they're also very important to keep in mind and continue to improve upon as you grow.

What has been your own growth through the last five years of FISMA? Have you seen yourself change over these these last few years?

Not dramatically. I mean, even if I look back at just the past few months, I think I've grown dramatically. I mean, I think that one of the dumbest things you can think anybody can think is, I know it all, I have figured it all out. A lot of people want to mentor people way too soon, I keep telling people, I'll have people ask me to mentor them, which I think is funny, I'm like, I'm happy to give you my advice. But you should never take everything I say as the law because there are many other ways of thinking there are many other ways of getting places and I'm still learning myself, I'm definitely a student. And the sort of ways I've improved, you know, even just over the past few months, I've become a lot more open to feedback. You know, I think in the past I there were times when you know, some if you someone says something you don't like you can it's a natural reaction sometimes to get defensive or feel defensive. But when you realize that, that reaction is just preventing you from hearing more feedback, and that feedback has an opportunity to grow, it's actually kind of a silly thing to do. So that's an area where I've seen a lot of where I've improved a lot, realizing the importance of being close to your team, and really thinking of them as really getting to know them as individuals. And building trust is super important, especially in the age of COVID, where it's a lot harder to be much more deliberate about that, than you're used to, you know, you can't just be on zoom calls, you have to well, you know, in whatever way is safe, ideally get together in person, but at a minimum, you should be talking to somebody versus just typing to somebody you can't have completely slack based culture with no exceptions to that or I think that it does build a lot of it doesn't create that that bond. So those are some areas where I've learned to improve for sure. There's, I'm sure many, many others. The way that I speak sometimes I have a tendency to ramble what you've heard, so I can't say that I fixed that. It's something I've worked on the speed at which I speak is something that I've worked on. Now there's a lot that you learn if you listen. And the most important thing to start with, if you want to start improving yourself is the skill of listening. It's like the ability to read, write, and if you can read, you can teach yourself anything. And if you listen, you will learn almost inevitably,

it's a fantastic perspective. Let's wrap up with a few lightning round questions here. What commonly held belief about spatial computing? Do you disagree with?

I think I disagree with how far people think it has already come? in the traditional sense, I think a lot of people are very quick to use the term spatial computing without really asking, what data are you actually working with? And what are you actually figuring out, just figuring out distance, and there's a surface over there. I'm not saying it's not spatial computing. But it's a little bit of an exaggeration to say that that's all that's really intelligent technology, that's been around for a long time. So spatial computing, the way that we think of it if there's not, and the way that I see it evolving is that it becomes as powerful as traditional computing actually become more powerful, because we have more data and that significantly more data and the three dimensional real physical world than we do in our electronic two dimensional one. So I think the way people perceive it, and people are too ready to call something, whether it's facial computing, or, you know, 3d analytics, or whatever, that they're too quick to use these terms, when really what they're using are just things like convolution neural networks that are purely based off of two dimensional data,

you have some element of neural networks as part of your solution, you have some deep learning element to it, how are you using those.

So the difference is that we don't rely on convolutional neural networks to match right now, when it comes to 2d to 3d, of course, that's part of the equation because you have to train. But it's not the sole thing that we rely on the learning part, what's unique about physics is how we view deep learning, right? So it's not about matching. A lot of people think that oh, well, you learn how to match things. No, actually, the matching is pretty solid. What we're learning from is when we match parts, right? So if you can think about it like this, if you had a global database, everybody's DNA, and you could see how everybody in the world was related. Think about how the kinds of things you could predict with machine learning, right, the kinds of correlations you'll be able to predict. And see, I'm not suggesting that anyone ever do that, for humans. But when you think about that, in terms of objects, right, the ability to see how all things relate, and any, any and all changes in deltas. And if we can then correlating that to information we have about the outcomes, it allows us weak physics actually able to very, very effectively predict things like how much apart might cost to manufacture or to produce or even just to buy, what material or materials are likely involved, the process of how to create that thing, how something might perform, how it might work well with something else, another way of naming something another, other parts that might be relevant, maybe not directly in a geometric way. But we know that they're related or relevant because of the sense of the written data, right? That that we have the additional context are given. So the geometry is how we know how all things are physically related. But then there's all this data about that, right? The description, the materials, the feedback, right, this thing broke after just a week, whatever. It allows us to start to make really interesting predictions, and almost a prescriptive type of I wouldn't go that far just yet. But you know, we're definitely headed in that direction, prescriptive types of predictions for our prescriptive, if you will, because we have so much more data and so much pure data to work with, it's actually astounding how quickly you can make predictions in terms of how many assets you need. So if you were to send over, you know, 20 models to us, the amount of predictions we can make, far exceed the amount of types of predictions you can make with 2d, there's just so much more you can do with that. But even if you limited it to the types of predictions you would be able to do potentially with 2d. In most cases, it's orders of magnitude faster, and more accurate because there's just that much more data to work with. And so it makes machine learning orders of magnitude faster, but also deeper, you can predict things that weren't possible before. And that's why I think there's a big difference between machine learning for the sake of guessing what something is, which we already do extremely well. And machine learning to understand because of these relationships. What might be true about these other things are something totally new that you're working on. can we predict What'll happen next? can we predict what might go wrong? I mean, predict what you might be trying to do, right, or a better way to do something. And that starts to really democratize 3d in a major way.

That's super fascinating. That's a whole nother level of abstraction, where you're basically asking the system, a set of questions about how to build something, or is initially whether the thing that's being proposed to be built will meet the requirements that are proposed, but derived from that. And ultimately, then how do you build it? Right? This piece should be made out of stainless steel and nylon, or whatever. Exactly. Amazing. Besides the one you're building, what tool or service Do you wish existed in the augmented reality market,

there are hundreds of tools I wish existed in the augmented reality market, it's a very, very young market, and we don't have enough convenient hardware yet, we don't really have smart glasses that are light enough, portable enough, socially acceptable enough, and powerful enough to be used in a wide enough scale. So I'm eagerly awaiting for the hardware to arrive. But there are many applications that I worry are not, not only are they not produced, and not even capable of being produced yet, because of some of the limitations on true geometric or spatial reasoning. And so the answer is that the list is too long to go through. But many, many, many things do not exist that I think should or will eventually. And that's why I'm really excited about the platform. And the API's, because this gives other people the ability to actually build that. And it's extraordinary to see what people think of when you give them this tool.

Yeah. If you can sit down and have coffee with your 25 year old self, what advice would you share with 25 year old Paul,

that I had some things in the building on the mindset category figured out? that are important? There are definitely a lot of things I've learned since then, I think if there was one thing that I would have focused on more at that age, is taking time to really, really learn some of the soft skills about dealing with others and inspiring others and communicating effectively, and how to how to give autonomy with with accountability and in the appropriate way. Those are skills that you know, are easily learned. But it's a very expensive thing if you've learned the hard way. And I've definitely had to learn some of those the hard way. And so if I could, I would tell myself to save myself from having to learn those the hard

way. Moving others is a concept that I think sometimes people think of it more as the literal cheerleader, or the the coach that stands in front of the team and gives this amazing rah rah speech. A lot of people look at themselves and say, I that's that's not quite who I am, it's not my personality. What do you think are kind of the key elements to be able to effectively motivate others,

I think you have to be authentic. I don't think that you can motivate people by putting on an act, maybe you can, temporarily, but if there's not an old expression, or you know, sort of accepted wisdom, when it comes to sales, that if you don't believe in your product, nobody else will. There's a lot of people who say if you don't believe in yourself, nobody else will. And the same is true for your vision, if you don't believe in it, nobody else will. And you have to believe in it. with every fiber of your being, you have to know that it has to be something that is clear as they do. And that's why having a clear goal is so critical. Not just saying I want to do something in this direction, or you have to know precisely what outcome you want a dream. But imagine that, and then build this beautiful picture in your mind of what it looks like in the future that you're at once you've accomplished that. And if you can do that for yourself, you can do that for others. And if you're truly inspired by what you're doing, inspiring others comes relatively naturally. Because it's palpable, you can feel it when somebody is excited about something it's not a it's not a mask, it's not a facade. And so when it comes to how to motivate others, the first and most important thing is that you yourself are very motivated. And the second thing is just really taking the time to meet with them and and share with them this guy's we're not here just doing this, we're doing something much bigger. Here's the why behind the what and the how people aren't motivated or they shouldn't be, you know, the right people are not motivated to go work somewhere just because of the salary or just because of a title. And if they are, they're fundamentally thinking about life the wrong way because you won't get to live at once in a lot of that's going to be spent at work. So if you think about it as just work, and I think that you're in the wrong field and doing the wrong thing. People want meaning in life, they won't be able to derive meaning out of what they do. And so it's your responsibility as a leader to give that to them to say this is why we're here everyone. This is The, this is a future state of the world that we're trying to create. And this is the role that you're going to play, you are impacting other people and the world in a very, very major way. And that gets people to go the extra mile without being asked to go the extra mile, and that you're going to create a startup that's crucial. Any closing thoughts you'd like to share? I would say that for those people who are in the AR community, specifically, I think that they're in an amazing area, because we are seeing now all signs pointing towards a new type of revolution of technology that hasn't happened in probably any of their lifetimes. And certainly in the last time that there was a major breakthrough in technology, a truly major breakthrough and technology and digital sense, was when binary code was used to create computers, instead of talking about 1940 zero, right? That's a long time ago, 80 ish years ago. Before that, it was about 100 years before that, that we created that we had the first of these leaps, which was Morse code, we had unary code, and all of a sudden, you can communicate with people all over the world, assuming you have a wire connected, and they instantly received the message. And the progress that we as a society and we as humanity have been able to make, from going hand in hand deliver a letter by horseback to having telecommunication through a telegraph in Morse code. And then from just having Morse code to having everything that we have now, because of binary, it's these are quantum leaps, and what we are able to do as humans in our lifetime, and what we're seeing now is that there is the same opportunity, and that last final dimension, that you have the dimension of depth. And that leap will be just as significant as a leap from horseback to the Telegraph, or from the Telegraph, to the current age of software.

That pig, this understanding of the 3d world plus our ability as humans to directly or intimately interact with, through that technology with that technology, through a pair of wearables, or whatever the set of technical components is, is the next breakthrough, the human centric computing in, in computers ability to understand the physical world, the combination,

yes, the integration of your perception of the physical world, and your ability to augment and change the reality that you live in, in a productive or entertaining way. It's the ultimate level of freedom. And it's, it would be silly for me to even forecast every element of that because no one was ever able to do that. But with great breakthroughs, but I would say that this is really on par with just about any of the great breakthroughs in his those, the last two types of code that came out, change the world forever, I would say this is essentially trinary code that we're working on. And it's not just us, it's the obviously there's the hardware companies that are creating the right infrastructure for it. And they're the developers that are they use the core technology to create the applications people want beyond the couple that, you know, we've made for specific use cases. So I think the future is unbelievably amazing, it's, it's unreal, it's gonna be unrecognizable, and any sci fi film right now is going to get it wrong, just like Back to the Future totally got wrong, what the world will look like in 2015.

That's a wonderful perspective. We know that we're at the precipice of something amazing, we're riding that wave together, people are listening, you and your team and all the rest. Where can people go to learn more about you and the work that you and your team are doing there if

you can go to FISMA comm pH y s, n A, if you want to try out a very light version of the software, some of the geometric matching capabilities, not all of them, but some of them are present on Thanx th en GS COMM And that's 100% free and you can also use augmented reality for any model, which is cool. If you want to reach out to me directly. You can look me up on LinkedIn, Max Paul powers if you look up Paul powers in business you will find and I'm more than happy to talk to anybody who is interested in this technology and getting involved with it. Paul, thanks

very much for this conversation. Thank you.

Before you go, I'm going to tell you about the next episode. In it I speak with Guido Groet, the Chief Strategy Officer of Luxexcel, a company that's able to 3d print optical quality lenses for smart eyewear. Luxexcel has been in the news recently for their partnerships with Loomis and wave optics, where together they're able to encase the waveguides with 3d printed prescription lenses. We get into the meat of the technology and where it goes from here. We don't is a great storyteller. I think you'll really enjoy this conversation. Please follow or subscribe to the podcast you don't miss this or other great episodes. Until next time.