Artwork

Contenu fourni par IndustrialSage. Tout le contenu du podcast, y compris les épisodes, les graphiques et les descriptions de podcast, est téléchargé et fourni directement par IndustrialSage ou son partenaire de plateforme de podcast. Si vous pensez que quelqu'un utilise votre œuvre protégée sans votre autorisation, vous pouvez suivre le processus décrit ici https://fr.player.fm/legal.
Player FM - Application Podcast
Mettez-vous hors ligne avec l'application Player FM !

InOrbit: Florian Pestoni

48:39
 
Partager
 

Manage episode 326455214 series 1508937
Contenu fourni par IndustrialSage. Tout le contenu du podcast, y compris les épisodes, les graphiques et les descriptions de podcast, est téléchargé et fourni directement par IndustrialSage ou son partenaire de plateforme de podcast. Si vous pensez que quelqu'un utilise votre œuvre protégée sans votre autorisation, vous pouvez suivre le processus décrit ici https://fr.player.fm/legal.

Florian Pestoni of InOrbit discusses the challenges of creating company culture, scaling robotics companies, and adopting industry automation.

ACCESS THE FREE PERSONA BUILDER hbspt.cta.load(192657, 'ee6f69de-cfd0-4b78-8310-8bdf983bdcc9', {});

Danny:

– Well hello and welcome to today’s IndustrialSage Executive Series. I am joined by Florian Pestoni who is the CEO and cofounder of InOrbit. Florian, thank you so much for joining me today on the Executive Series.

Florian:

– Yeah, Danny, thanks for having me.

Danny:

– I’m excited to get into your story. So for those who aren’t familiar with InOrbit, if you could give me a little high-level, who you guys are and what you do.

Florian:

– Yeah, we’re a data platform for robot operations. So essentially when you roll out autonomous robots or smart robots into the world, you still need to keep track of them. You still need to manage them. So we provide the infrastructure in the cloud to manage all these robots, everything from knowing what they’re doing to stepping in when necessary.

Danny:

– Alright, excellent. Well that sounds interesting. It sounds like it’s a very needed thing, especially as there is an exponential amount of robotics companies that are coming out, and there’s all kinds of unique and different applications. I suspect that there’s a nice challenge that poses that you guys are solving, and we’ll get into that here in just a little bit. But before we do that, this is the segment where we like to dive in and get to know our guests a little bit better. This is where we want to know a little bit more about Florian. So Florian, if you could just give me some color on your background. How did you get into the technology space? Take me way back.

Florian:

– Yeah, I’ll tell you Danny, it has not been a straight line.

Danny:

– It rarely is.

Florian:

– I think in the curves is where something interesting happens. Maybe a little bit about myself, I’m originally from Argentina, born and raised, studied there. Came to Silicon Valley essentially pursuing the best technology. Worked at research labs and then eventually got closer and closer. into the business. Did an MBA here at UC Berkeley. I went from big companies to start-up, back to big companies. Usually a lot of people will either pick one track or the other. But to me, it’s always been just different ways to have an impact. I’ve worked at some of the very well-known companies like Microsoft and Atlassian and the company now known as Meta as well as a lot of those start-ups that you probably never heard of with various degrees of success, as start-ups go. I think what’s been the common thread throughout all of that is really technology with a global impact. I had the chance to work on products, for example on video, on the internet when it was relatively new, when it was still being proven out. Being able to reach a billion people with my products was really fantastic. But then I’ve also worked in really new technologies like, for example image processing, cryptography, all of these foundational technologies that we’re now seeing in spaces like the one I’m in, in robotics.

So what’s interesting is some of the things that I did very early in my career without being able to foresee where they were going are things that I’m utilizing these days in this new space of robotics. Then in terms of robotics in particular, the way I like to say it is I blame my cofounder for getting me hooked on this. He’s had over 12 years of experience working in robotics. His name is Julian Cerruti. He worked with a research group called Willow Garage that was extremely influential. He started telling me about this recurring problem that robotics companies had, which is they work really hard to make a robot, but then once you go into the real world, you need to scale. Scaling is hard for every business, but it’s particularly hard for robotics. We built a lot of the foundational technologies to help them out.

Danny:

– Wow, that sounds—well definitely a very interesting story there. You told me how, you said you came from Argentina, and you went to UC Berkeley, you were looking at the MBA program there. Take me back before that. What’s the draw into the technology space? What pulled you into it?

Florian:

– It’s a good question. For me, it’s always been about what’s the thing that can have the most impact? I think we’ve seen the positive and sometimes negative impact of technology. I think that different professions will have impact maybe at more of a local scale. If you’re a physician, for example, you can have a huge impact on people, but you have to treat them one-by-one. Whereas through technology, you can reach a lot of people. Like I said, I had the fortune of working with products that had global distribution with over a billion users. So I think it’s really about that reach and that impact at scale. I could give you an example of a prior company that I started and shut down as these things go, which was in the digital health space. This one was really a labor of love for me, a very personal thing. My dad went through a lot of health issues, and it motivated me personally to go try to help people. I thought that through technology I was able to do that. This particular company didn’t pan out, but now you see a lot of digital health companies that are doing just that, especially when people are less able to go to the doctor. So technology does have that empowering effect, and in a very strange twist of fate, I now work on the health of robots instead of the health of people.

Danny:

– Yeah, that makes a lot of sense. That’s a good point there. Was that some EMR-type technology or some sort of digi—yeah.

Florian:

– It was a lot of digital tracking of—in the US, despite all the news that we get from Covid these days, cardiovascular disease is the number one killer.

Danny:

– Absolutely, yeah.

Florian:

– So I worked with two physicians who were specialists in treatment of cardiovascular disease, and a lot of it is about daily management, so blood pressure for example has a lot to do with food intake and so on. A lot of this is very lonely for people. You’re living with your condition every day. You don’t have a lot of support. This was meant to bring that extra level of support through data and help you stay on the path to maintain your health as opposed to treating the disease.

Danny:

– I think it’s always fascinating with technology as you were talking about just—you talked about having a global impact and being able to—it’s sort of the one-to-many effect versus maybe one-to-one. I think it was interesting, especially; you were talking about you were in the medical space, and that one-to-one treatment versus this technology being able to go out and open that up and touch tons of people. You were involved in a platform that had billions of users. That’s pretty exciting, and certainly would be able to touch a lot of people there. Throughout your career, who has had a big impact from an influential standpoint?

Florian:

– Yeah, let me answer this way. It’s an interesting question. People have influence in different ways. I had a great mentor really early in my career. I was doing research, and she was a world-renowned researcher. She took the time to get to know me and to understand what my motivations were and to coach me through that. I’ve always been greatly appreciative of that extra time, that extra effort that somebody puts beyond what their job description is. I also wanted to mention, there was another person who had an influence through inaction. I don’t think it was intentional on their side. I’ll tell you the story.

Danny:

– I’m intrigued, yeah.

Florian:

– Really early in my career, I was already very passionate, I would say, about technology and about having this global impact. I was working at a very traditional, slow-moving organization, and I kept pushing and pushing because I knew we could do more. We could have more impact for our customers. But people just wanted to maintain things as they were. And at one point–– again, this was probably my second manager ever— he comes to me, and he says, “You know, Florian, I love all your energy. You keep pushing and pushing, but sometimes when you push so hard, the rope breaks.” So he was basically telling me, look, just pipe down and play along. It was fantastic feedback because within a couple of weeks I was working somewhere else. That helped me move my career forward. He basically said, look, if you want to stay here, you just need to go slow. And I said, nope, not for me. So I think it helped me, and years later I remember that conversation. Again, it wasn’t a pleasant conversation, and it wasn’t intended as advice. Sometimes influence that pushes you in one direction or another can come from any side.

Danny:

– Yeah, absolutely. I love that; that is kind of a funny story. You were saying yeah, a little bit of an unintended, maybe potential consequence or action because of that. It certainly makes a lot of sense. It’s a very interesting difference between, you were working with gigantic, monolith companies. It’s interesting; even the companies that had start-up roots, they were lean, and they could pivot quickly. But then as you grow you start adding all these other layers to be able to essentially stop the innovation, or you mentioned to maintain, and you keep adding layers and layers and layers and layers and layers. I’ve always found that interesting. You’ve got the root, innovation, move, we can pivot; let’s go. And then it gets siloed, and then it’s just this giant ship you’re trying to move. It’s very difficult to make change there. That makes a lot of sense to me. I think it’s interesting. What was the phrase again? It was, sometimes if you push too hard…

Florian:

– I think it was something, if you’re pulling too hard to get everyone to come along, the rope breaks. The phrase that stuck in my mind was the rope breaks. Again, I think it was meant as a, just play along. I think the value that it brought me was clarity that that wasn’t a fit for me.

Danny:

– Absolutely.

Florian:

– Even today, InOrbit is still a relatively small organization. We’re about 20 people, and my cofounder and I spent so much time on culture. I know people talk about culture in grandiose terms. For us, it’s really helping people understand why we’re doing what we’re doing and really helping them bring their own way of contributing. I think—maybe this is too much of a rathole, Danny, but—

Danny:

– No, this is good. Keep going.

Florian:

– We hear a lot from companies that talk about culture fit. But I don’t think that’s what you should aim for because then you’re trying to get people to conform to a pre-existing mold. I think what you want is culture-add. As you bring someone new, are they contributing? Are they adding to the culture in the general direction that you want to take? I think, as you said, it’s really hard to do that at a greater scale. It’s a little funny that, with a company of 20 people, we’re already thinking about okay, how will this work when we’re 1000? But I believe that you lay the foundation for that culture really early in the life of a company. And then if you’ve done your job right, it’s self-reinforcing. You tend to attract people who want to contribute and add to that culture. If you get it wrong really early on, you did it with your hands. There’s the roots and then the tree. Maybe the tree will be skewed, and so I think it’s important to get it right. We spend a lot of effort on that.

Danny:

– That’s awesome. I would love to dig into that a little bit more if we can. I love the idea around culture, and it fits really well into your story there as far as how that transition goes. I love the concept of, you said a culture-add versus culture fit. What are some concrete things that you’re doing there to set the tone for the culture?

Florian:

– Yeah, maybe a little bit about the company. We’re a distributed team, so we have people all the way from northern reaches of Canada to South America. As a distributed team—and throughout the US also, from the East Coast to the West Coast. So that adds an extra layer of challenge when it comes to having people feel like they’re part of an organization that has a shared meaning and a shared objective. One of my—people don’t usually love meetings, but I do have a favorite meeting that we do every two weeks on Fridays. We call it the wins meeting. This is a meeting where we get together; there is no agenda, per se. People from across the company, we all get together. It’s not mandatory, so it’s not an official thing, but usually everyone is there. It’s really to share wins and lessons. Wins is, could be something that you did at work, you completed some development, and you want to demo to the rest of the team. Or somebody won a big deal, and they want to celebrate that. I think that’s pretty normal to do celebrations of things like that. On the other hand, we also do lessons. I think that’s really valuable, which is something that didn’t go according to plan. What did you learn? And sharing that with everyone else. This meeting kind of took a life of its own, and I’ll tell you a concrete story. We had an intern, and it was—one of the things we do is we rotate who is basically the emcee for the meeting, for the wins meeting. It’s a very participatory thing. We were having our wins meeting on May 4th, and for Star Wars geeks, that’s an important day. It’s May the Fourth day. We ended up; I said, well why don’t we do something with Star Wars? So this intern who happened to be hosting, she put up this amazing trivia game about Star Wars. Then we have some people at our company that are insane with trivia, like really, really good. They beat me every. Single. Time. But what’s funny is, this became kind of a tradition, so a lot of times we start the meeting with, tell us something about yourself, something embarrassing that you did or a trivia game. And it took a life of its own, and what I love about it is, it wasn’t like a top-down thing. We just created a space, and then it’s self-reinforcing. I don’t know. It’s a great way, like I said, my favorite meeting.

Danny:

– That’s awesome. That sounds like a great meeting. I love that concept of doing that and creating a cadence and letting it go beyond maybe what it was initially set up for and grow a little bit more organically. I think I see now a little bit more where you’re talking about the culture add aspect. I see that very clearly in that example. That’s fantastic.

Florian:

– So for example, we have an engineer on our team who does parasailing.

Danny:

– Very cool.

Florian:

– I’ll admit, I look forward to when he shares a video or something because it’s like vicariously living through his excitement. But you get to know people beyond just work. I think that’s really important for me.

Danny:

– That’s a great takeaway. I’m writing some notes here. I think we may be implementing something like that similar here. Anyways, I’m selfishly very excited about that. Thank you. What a great thing, talking about culture. It felt like a great, certainly an absolutely great segue as you’re building out InOrbit and maybe some of that inspiration. What’s the story behind how you started it? I know you mentioned that you said you blamed your cofounder there a little bit, but what’s the story there? How did you start? When did you say, we’re going to make this jump, and we’re going to start this venture?

Florian:

– As with many companies, it started with just a conversation about interesting problems to solve. Like I mentioned, my cofounder had been in the robotics space for quite some time. I came from the cloud world, so it was completely foreign to me. Maybe like a lot of people who may be watching this had my preconceived notions from movies and so on about robots. As I started learning more about it, I learned a couple of things. One is, the old-school robots that worked in factories, for example if you go to an automaker’s factory floor, you’ll have these massive robots, million-dollar robots. They just do the same thing over and over. They’re programmed in a very precise fashion to do the same thing over and over. What got me really excited was, there’s now a new generation of robots enabled by really new technology where the robots are coming into the open world. So now you’ll see robots in warehouses, in farms, on construction sites, at retail stores, in airports. So really pretty much wherever you go, there is a robot. I live in Mountain View in California, and when the pandemic hit we were really worried about our close neighborhood store, our grocery store, mom-and-pop shop, we know the owners. We’re like, okay, how are they going to hold up? Well this being Silicon Valley, within two weeks they were doing delivery to the home via robots.

Danny:

– Really?

Florian:

– What I thought was interesting was people’s reaction to the robots because at first it was just like, oh, my God, what is this? People taking pictures. That lasted about two weeks. Then it became part of the (of course) robots. So I think it’s interesting how we as consumers ourselves, we respond to the technology, first maybe in awe, but then it becomes part of our landscape. I think that’s what we’re seeing what’s really exciting about what’s happening with robotics. Back to InOrbit, we said, okay, great; how do we help? We see the potential. We see the impact. How can we add to that? Before we even started the company, within I think it was 25 interviews with founders of robotics companies, with users, really to understand what their problems were, we landed on this idea that, when a robot goes from the lab to the real world, there’s a big transition.

A lot of times, and this is not probably very well-known, but a lot of times in those early days, robots had chaperones. They literally have an engineer following the robot around to make sure that it behaves as needed. Again, in the early days you probably had more engineers than you had robots. Then as you start to scale, when you want to go from 5 to 500, let’s say, a lot of things that worked in the early days stop working. That’s something that I’ve seen in the cloud, and that’s where my connection to this was–– which is, I’ve seen this work in the cloud where in order to operate the massive data centers that companies like Facebook or Microsoft operate, you couldn’t do it one server at a time. You have to have these scalable tools. So we set out to build those scalable tools for robotics, and we landed on this concept that we call rob-ops. If you’ve heard of dev-ops, which is tools and practices for operating software at scale, we’re bringing a lot of those ideas to the world of robotics.

Danny:

– That’s awesome. Can you say that again? I had not heard that, and I’m not ashamed to admit that, so what was…

Florian:

– So in the cloud space, there’s something called dev-ops.

Danny:

– Oh, dev-ops. Okay, I thought you said something else.

Florian:

– Yeah, like development and operations put together. So now we’re doing rob-ops which is robotics operations as a cohesive idea.

Danny:

– Perfect. That makes a lot of sense, and that’s great. Great analogy there, or connection on that. I’m curious, though; how did—you said you talked to several robotics companies just wanting to get an understanding. I think you said 20 or so leaders and end users. What was that moment in your mind? It sounds like you’ve done this before, so it’s not like—you’ve gone through this process, but did you have a particular moment where you said, okay, yeah, this is it. This is a big problem that I know we can solve, and it’s time to go. Let’s do that. What was that moment like?

Florian:

– Yeah, maybe to the point that you made about having done this before, there’s a method to the madness of starting a company. It’s generally called customer discovery and customer development. Customer development is a framework by Steve Blank. You go through the process, and that’s part of what guided us. In our case, maybe the moment of insight was precisely this idea that robots are, if you want to think about it this way, it’s like servers on wheels. It’s like a complex computer server with lots of sensors and moving around. So it’s in the—if you designed a data center for hundreds of thousands of hosts or computers, you need reliable power. You need great networking. You need ventilation to dissipate all that heat, so you design it, and you control as much as you can.

A fleet of robots, I like to call it, it’s a data center from hell. It’s the worst possible way of operating a distributed computing platform because a lot of times they’re in uncontrolled environments. They have networks that come in and out. They’re on battery, so they need to come back on their own to recharge if they’re a mobile robot. So the challenges are much, much higher. When we talked to a lot of these robotics companies, what we realized is a lot of them weren’t necessarily thinking about those problems of scaling. If you’re very passionate about robotics, maybe you have a PhD and you work as a roboticist, you really have to focus on building that robot. We felt we could augment and help them focus on that, and we’ll take care of the boring part, if you will, of running the robots at scale.

Danny:

– Yeah, that’s exciting. So maybe an oversimplification of what you do. I’m a little nervous about saying this, but it sounds like you’re helping to solve—and there’s a lot of different challenges, but I know there’s a big challenge of interoperability, certainly, in this space because there are so—it’s interesting to see the growth of robotics over the last several years. It was a fast trajectory of growth, and then you pile on the pandemic. The huge needs just grew exponentially, and they continue. I can’t remember; there was a stat that I’ve seen as far as how many new robotics companies are on the—you have players who’ve been in here for years and years and years, and then you’ve got constantly new start-ups.

It’s interesting because, like you mentioned, a lot of people think, you watch Terminator or all these other robotics and AI movies. Oh, wow, it’s going to take over the world and all these things. The reality of it is, is even though that, yes, you’ve got from an automation standpoint you mentioned automotive, and you’ve got these giant, maybe these FANUC robots that have been there, and they’re moving, and they’re—but they’re very set in place, or they’re caged. You’ve got all of these AMRs that are now, they’ve got more autonomy, and they’re going around all over the place. It’s totally opened up the playing field. But the problem being that, that’s great, but you can still only solve a finite amount of problems with this particular robot. You may need a whole fleet of 10 different types of robots to solve some of these different challenges. Is that part of what you’re solving there, with some of the software solutions?

Florian:

– Absolutely. If I can get really geeky here—

Danny:

– Please, get really geeky.

Florian:

– For a minute, we like to say we solve the four Os of robot operations. So the first O is observability. This is kind of like a fancy word for monitoring, but it’s really knowing through data what’s up. What is the robot doing? It has its own set of challenges as I describe because they’re all over the place. They’re on weak networks sometimes. So we solved that problem, and that becomes the foundation because it’s the data. So at our core, we’re a data platform.

But then from there, there’s an operational aspect which is, robots get into situations where they’re no longer autonomous. They’re stuck. This happens—I won’t disclose numbers, but it happens a lot more than everyone anticipates. I think we have this picture of a robot; oh, it must be perfect. Far from it, especially when you’re in crazy environments. You talk about a warehouse, for example, but we’ve seen robots in retail stores while there are shoppers around. So there’s going to be a kid with a shopping cart and ketchup stains on the floor. You can’t expect the robot to be perfect, and that’s where some of these—augmenting the robot by human operators is a way to do that. So that’s the second O, operations.

I think then you get into optimization. What you really want is to not have those problems happen as often. How can you solve it? We do things like tools for root cause analysis. Root cause analysis allows you to identify cases where things didn’t go as planned and derive insights from it. And just to give you a sense of scale, each robot may generate a terabyte of data per hour. The approach of, we’ll just throw everything to the cloud isn’t really practical, especially as you’re thinking about we now have fleets of thousands of robots operating in the world. So that takes us—optimization is how do I run my fleet more efficiently?

Then the last O is getting to what you were mentioning before. We call it orchestration. Again, that’s a very cloud term, but it’s really, how do I bring different components together in a harmonious way? Now you’re more like the conductor of the orchestra. That can include many different types of robots from different vendors. In a warehouse, for example, you may have—let’s say there’s a cargo dock, and a truck comes in. Well you need to unload it, so there are now autonomous forklifts that can do part of that job, or maybe all of it one day. Then you have pallets. You need a depalletizing robot that will break down the pallet into its components. Now you need to take it into the warehouse proper. How do you do that? Do you do it one by one? Traditionally a lot of these warehouses have conveyor belts. The conveyor belts are really efficient, but they’re really expensive and they’re really inflexible. So what an AMR, an autonomous mobile robot gives you is flexibility because it’s controlled through software.

So that’s where we’re able to bring all of these robots from different vendors and give someone a complete view. The term in the industry is a single pane of glass, one way to see all of your IT, or in this case OT, in one view. I think that’s where this is going, and you see really pioneering companies like FedEx really advocating for this ability to control different robots.

Danny:

– That’s super interesting. Thank you so much for getting, I think you said geeky with me. Getting into the technical weeds. I appreciate that. Actually one thing in particular that I had no idea, never heard this before: one terabyte of data per hour, one robot could generate. That’s a ton of data, especially if you’ve got a fleet of hundreds of these things. Transmitting that and getting that, storing it and then analyzing it seems—That’s crazy. That’s a lot of data.

Florian:

– A lot of times, Danny, all of those terabytes of data, you just need this one piece of data, the proverbial needle in the haystack. I think that’s what we do is, we try to turn all of that data into insights that people can take action on.

Danny:

– Right, absolutely. Well there’s a ton there. A couple other questions that I’ve got coming up then, how do you engage with your customers? Are your customers—obviously you talked about robotics companies. So I imagine, would your customers be robotics companies, end users, and integrators? Do you work with one, all—how does that work?

Florian:

– We work with different players in the ecosystem. At the end of the day, because of our orientation towards operations, we really care about maximizing the potential of every robot. As a robot vendor, of course you want a more efficient robot, a robot that fails less often or that requires fewer interventions. So we can provide a lot of data for them to improve on their robot. As an end-user, you’re really running the robots. In fact, I like to say, and this is a little shocking for companies in the robotics space, the end users, they don’t really care about the robots. They care about their process and what the robot can do to it. Their needs are going to be a little different. They’re all based on the same foundation of data, but what they’re trying to do is, they’re trying to drive this SLA higher and higher and higher.

What’s interesting is there’s this middle ground between those two cases that I mentioned which is this idea of robotics as a service. Increasingly you see companies that, instead of selling you a robot and now you do whatever you want with it, they’ll essentially sell you a service that is backed by their own robots. We’re starting to see—this is very new—third party companies that will say okay, I’m not going to make the robot. I’m not an end-user. I’m just going to run these robots at scale. So we can help them with our infrastructure to do that like I said at scale, across robots, across sites, and across customers.

Danny:

– Yeah, that sounds super interesting. I have heard, definitely over the last little bit or so, robotics as a service. I think it’s a super fascinating play. Certainly with all the labor issues that we’ve got going on now, certainly a lot of companies are turning towards automation in any way—there’s a huge shift in the digital transformation that is happening. It was happening before, but again because of the pandemic in all areas, you’re seeing the supply chain, manufacturing, all up and down, logistics, and beyond. End users, you were mentioning retail and grocery for example. I love the store that you talked about, how the response with the local grocery store in the neighborhood, you said being Silicon Valley, what do we do? It’s fascinating, but we’re continuing to see more of that. And we’ll see more. What does the future look like for you?

Florian:

– I like to think of this future where we have millions of robots around the world that are positively impacting billions of people. Each robot can have this multiplier effect, whether it’s a robot in a farm making—so one of the companies that we’re familiar with is a, basically a robot that goes in a farm, identifies weeds, and then literally zaps them with lasers. Instead of having to have herbicides and spray it, you can turn that to zero. Another customer of ours has been able to grow leafy greens with a 90% reduction in the use of water. Again, all of these are things that are not moving the needle a little bit. They’re moving it a lot. We’re seeing a lot of electrification also which a lot of times goes with autonomy together. We see that reducing, for example, the carbon emissions. These little robots that deliver around Mountain View, they’re replacing somebody driving around in their car, getting out of the car, dropping the groceries, and getting back in the car. I think these things have huge impact.

But to take it a little bit further, where we see the future is the collaboration between robots and people. I think we’re only starting to see the shape of that. I like to talk about this what we call a roboteer, like a puppeteer, but for robots. Maybe your job isn’t to go deliver food or isn’t to unload a truck. You’re monitoring. You’re the boss of a group of robots that are doing those things. Just like we said, the robots may not work fully autonomously, fully independently all the time. There’s always going to be corner cases or areas where somebody needs to intervene. But now maybe you can do–– instead of sitting in a forklift and driving it around–– maybe you can supervise 5 forklifts or 10 forklifts. You don’t even need to be there. You could be doing it through our platform or through other people’s platform. You could be in an air-conditioned room sitting in an ergonomic chair and responding to events as they happen. I think that vision of collaboration between people and robots and obviously the cloud and AI that come into that, I think that’s a great, in my mind a very optimistic vision of the future.

Danny:

– That’s really interesting. I like that, and it’s interesting. People always harp on, “Robotics is taking people’s job away,” and the answer we keep hearing is—I love this; I haven’t heard this, roboteer. It makes a lot of sense. You’re seeing that’s a—I do certainly believe that is where the future is going, too, when you look at a lot of these jobs as well. Obviously we’ve got major issues right now for a whole host of reasons, but the trend for years though before this has been, a lot of people, they don’t want to work. I don’t want to work in a warehouse, or I don’t want to work in manufacturing or logistics. There’s been a struggle trying to attract, so if there’s a way to say listen, we’re already seeing a decline in that. If we can bring in a solution over here from a technology standpoint to be able to assist people in those jobs, it’s creating an increased net positive effect.

Florian:

– Yeah, and you say I don’t want to work. I think it’s—these are tough jobs. I had the opportunity to tour some facilities, and without automation your job may be to literally walk 15 miles a day within the warehouse pushing a cart back and forth. Let me tell you, that’s not good on your knees. It may be super healthy because you’re walking a lot, but it’s not necessarily the best environment where you want to do that. What we’ve seen is, we’ve seen this tremendous increase in productivity while also improving the health of the workers where you can have, instead of somebody pushing a cart, the cart knows where to go. When you think about a robot, some people may have this idea of a humanoid robot. Our robots have very boring shapes. They resemble something that has existed all along, but now they add autonomy to it. So if the cart can go where it needs to go, now you’re focused on picking the right products and putting them on the cart, not on walking back and forth. So that has the dual effect of better, less wear and tear on the body and also greater productivity. Some companies will see 300% productivity increases by introducing an AMR. I think it can very much be a win-win.

Danny:

– Certainly, absolutely. Alright, my last question is a very difficult question. Since you brought up Star Wars earlier and you’re working in the robotics space, I’m curious. What’s the difference between a robot and a droid?

Florian:

– Ah, I think it’s just a terminology thing. Everyone knows R2-D2.

Danny:

– Who’s that? Sorry; no, I’m just playing.

Florian:

– So that is actually called an astromech droid. This is me showing off my Star Wars nerdiness. They do have different models. Not all of them are R2-D2. You do have, in the Star Wars universe, different types of droids that do different things. They already had this orchestration problem in Star Wars, and maybe they’re using InOrbit—you know, InOrbit, star—maybe there’s something there.

Danny:

– Maybe, and it’s a great parlay from the question of the future. So there we go. Listen Florian, I really have appreciated our conversation. Thank you for obliging. I loved your answer. Thank you so much on that, and for just sharing all of your insights about the industry, what you guys are doing, sharing your story and how you got to where you are today. It sounds like you guys have a very bright future, no pun intended with the Star Wars analogy. But I really do think that you guys are solving some big challenges and creating some unique opportunities in a space that is exploding. I’m going to be very excited to follow you guys. So for those companies who would like to learn more about you, I’ve got the website here, or the URL, inorbit.ai. I want to make sure that’s correct. We’ll have that in the show notes.

Florian:

– You got it.

Danny:

– And Florian, again, just thank you so much for your time with us today.

Florian:

– Danny, it’s been a lot of fun. Thank you so much.

Danny:

– I have enjoyed it too. Alright. Alright, well that wraps up today’s IndustrialSage Executive Series with Florian Pestoni who is the CEO and cofounder of InOrbit. You can go check them out at inorbit.ai. I loved today’s episode. Great story; a lot of great insights here.

So if you like today’s episode, you need to go to IndustrialSage.com if you’re not there right now, and you can subscribe because you’re missing out on great content like this. Not only do we have the Executive Series. We also have news, and we’ve got a lot of other great content coming down the pike. You don’t want to miss it; I guarantee it. So subscribe; don’t miss out. And I’ll be back next week on another episode of IndustrialSage.

Thanks for reading. Don’t forget to subscribe to our weekly newsletter to get every new episode, blog article, and content offer sent directly to your inbox. You can also subscribe wherever you download podcasts so you can listen on the go!

Sponsored by Optimum Productions

  continue reading

241 episodes

Artwork
iconPartager
 
Manage episode 326455214 series 1508937
Contenu fourni par IndustrialSage. Tout le contenu du podcast, y compris les épisodes, les graphiques et les descriptions de podcast, est téléchargé et fourni directement par IndustrialSage ou son partenaire de plateforme de podcast. Si vous pensez que quelqu'un utilise votre œuvre protégée sans votre autorisation, vous pouvez suivre le processus décrit ici https://fr.player.fm/legal.

Florian Pestoni of InOrbit discusses the challenges of creating company culture, scaling robotics companies, and adopting industry automation.

ACCESS THE FREE PERSONA BUILDER hbspt.cta.load(192657, 'ee6f69de-cfd0-4b78-8310-8bdf983bdcc9', {});

Danny:

– Well hello and welcome to today’s IndustrialSage Executive Series. I am joined by Florian Pestoni who is the CEO and cofounder of InOrbit. Florian, thank you so much for joining me today on the Executive Series.

Florian:

– Yeah, Danny, thanks for having me.

Danny:

– I’m excited to get into your story. So for those who aren’t familiar with InOrbit, if you could give me a little high-level, who you guys are and what you do.

Florian:

– Yeah, we’re a data platform for robot operations. So essentially when you roll out autonomous robots or smart robots into the world, you still need to keep track of them. You still need to manage them. So we provide the infrastructure in the cloud to manage all these robots, everything from knowing what they’re doing to stepping in when necessary.

Danny:

– Alright, excellent. Well that sounds interesting. It sounds like it’s a very needed thing, especially as there is an exponential amount of robotics companies that are coming out, and there’s all kinds of unique and different applications. I suspect that there’s a nice challenge that poses that you guys are solving, and we’ll get into that here in just a little bit. But before we do that, this is the segment where we like to dive in and get to know our guests a little bit better. This is where we want to know a little bit more about Florian. So Florian, if you could just give me some color on your background. How did you get into the technology space? Take me way back.

Florian:

– Yeah, I’ll tell you Danny, it has not been a straight line.

Danny:

– It rarely is.

Florian:

– I think in the curves is where something interesting happens. Maybe a little bit about myself, I’m originally from Argentina, born and raised, studied there. Came to Silicon Valley essentially pursuing the best technology. Worked at research labs and then eventually got closer and closer. into the business. Did an MBA here at UC Berkeley. I went from big companies to start-up, back to big companies. Usually a lot of people will either pick one track or the other. But to me, it’s always been just different ways to have an impact. I’ve worked at some of the very well-known companies like Microsoft and Atlassian and the company now known as Meta as well as a lot of those start-ups that you probably never heard of with various degrees of success, as start-ups go. I think what’s been the common thread throughout all of that is really technology with a global impact. I had the chance to work on products, for example on video, on the internet when it was relatively new, when it was still being proven out. Being able to reach a billion people with my products was really fantastic. But then I’ve also worked in really new technologies like, for example image processing, cryptography, all of these foundational technologies that we’re now seeing in spaces like the one I’m in, in robotics.

So what’s interesting is some of the things that I did very early in my career without being able to foresee where they were going are things that I’m utilizing these days in this new space of robotics. Then in terms of robotics in particular, the way I like to say it is I blame my cofounder for getting me hooked on this. He’s had over 12 years of experience working in robotics. His name is Julian Cerruti. He worked with a research group called Willow Garage that was extremely influential. He started telling me about this recurring problem that robotics companies had, which is they work really hard to make a robot, but then once you go into the real world, you need to scale. Scaling is hard for every business, but it’s particularly hard for robotics. We built a lot of the foundational technologies to help them out.

Danny:

– Wow, that sounds—well definitely a very interesting story there. You told me how, you said you came from Argentina, and you went to UC Berkeley, you were looking at the MBA program there. Take me back before that. What’s the draw into the technology space? What pulled you into it?

Florian:

– It’s a good question. For me, it’s always been about what’s the thing that can have the most impact? I think we’ve seen the positive and sometimes negative impact of technology. I think that different professions will have impact maybe at more of a local scale. If you’re a physician, for example, you can have a huge impact on people, but you have to treat them one-by-one. Whereas through technology, you can reach a lot of people. Like I said, I had the fortune of working with products that had global distribution with over a billion users. So I think it’s really about that reach and that impact at scale. I could give you an example of a prior company that I started and shut down as these things go, which was in the digital health space. This one was really a labor of love for me, a very personal thing. My dad went through a lot of health issues, and it motivated me personally to go try to help people. I thought that through technology I was able to do that. This particular company didn’t pan out, but now you see a lot of digital health companies that are doing just that, especially when people are less able to go to the doctor. So technology does have that empowering effect, and in a very strange twist of fate, I now work on the health of robots instead of the health of people.

Danny:

– Yeah, that makes a lot of sense. That’s a good point there. Was that some EMR-type technology or some sort of digi—yeah.

Florian:

– It was a lot of digital tracking of—in the US, despite all the news that we get from Covid these days, cardiovascular disease is the number one killer.

Danny:

– Absolutely, yeah.

Florian:

– So I worked with two physicians who were specialists in treatment of cardiovascular disease, and a lot of it is about daily management, so blood pressure for example has a lot to do with food intake and so on. A lot of this is very lonely for people. You’re living with your condition every day. You don’t have a lot of support. This was meant to bring that extra level of support through data and help you stay on the path to maintain your health as opposed to treating the disease.

Danny:

– I think it’s always fascinating with technology as you were talking about just—you talked about having a global impact and being able to—it’s sort of the one-to-many effect versus maybe one-to-one. I think it was interesting, especially; you were talking about you were in the medical space, and that one-to-one treatment versus this technology being able to go out and open that up and touch tons of people. You were involved in a platform that had billions of users. That’s pretty exciting, and certainly would be able to touch a lot of people there. Throughout your career, who has had a big impact from an influential standpoint?

Florian:

– Yeah, let me answer this way. It’s an interesting question. People have influence in different ways. I had a great mentor really early in my career. I was doing research, and she was a world-renowned researcher. She took the time to get to know me and to understand what my motivations were and to coach me through that. I’ve always been greatly appreciative of that extra time, that extra effort that somebody puts beyond what their job description is. I also wanted to mention, there was another person who had an influence through inaction. I don’t think it was intentional on their side. I’ll tell you the story.

Danny:

– I’m intrigued, yeah.

Florian:

– Really early in my career, I was already very passionate, I would say, about technology and about having this global impact. I was working at a very traditional, slow-moving organization, and I kept pushing and pushing because I knew we could do more. We could have more impact for our customers. But people just wanted to maintain things as they were. And at one point–– again, this was probably my second manager ever— he comes to me, and he says, “You know, Florian, I love all your energy. You keep pushing and pushing, but sometimes when you push so hard, the rope breaks.” So he was basically telling me, look, just pipe down and play along. It was fantastic feedback because within a couple of weeks I was working somewhere else. That helped me move my career forward. He basically said, look, if you want to stay here, you just need to go slow. And I said, nope, not for me. So I think it helped me, and years later I remember that conversation. Again, it wasn’t a pleasant conversation, and it wasn’t intended as advice. Sometimes influence that pushes you in one direction or another can come from any side.

Danny:

– Yeah, absolutely. I love that; that is kind of a funny story. You were saying yeah, a little bit of an unintended, maybe potential consequence or action because of that. It certainly makes a lot of sense. It’s a very interesting difference between, you were working with gigantic, monolith companies. It’s interesting; even the companies that had start-up roots, they were lean, and they could pivot quickly. But then as you grow you start adding all these other layers to be able to essentially stop the innovation, or you mentioned to maintain, and you keep adding layers and layers and layers and layers and layers. I’ve always found that interesting. You’ve got the root, innovation, move, we can pivot; let’s go. And then it gets siloed, and then it’s just this giant ship you’re trying to move. It’s very difficult to make change there. That makes a lot of sense to me. I think it’s interesting. What was the phrase again? It was, sometimes if you push too hard…

Florian:

– I think it was something, if you’re pulling too hard to get everyone to come along, the rope breaks. The phrase that stuck in my mind was the rope breaks. Again, I think it was meant as a, just play along. I think the value that it brought me was clarity that that wasn’t a fit for me.

Danny:

– Absolutely.

Florian:

– Even today, InOrbit is still a relatively small organization. We’re about 20 people, and my cofounder and I spent so much time on culture. I know people talk about culture in grandiose terms. For us, it’s really helping people understand why we’re doing what we’re doing and really helping them bring their own way of contributing. I think—maybe this is too much of a rathole, Danny, but—

Danny:

– No, this is good. Keep going.

Florian:

– We hear a lot from companies that talk about culture fit. But I don’t think that’s what you should aim for because then you’re trying to get people to conform to a pre-existing mold. I think what you want is culture-add. As you bring someone new, are they contributing? Are they adding to the culture in the general direction that you want to take? I think, as you said, it’s really hard to do that at a greater scale. It’s a little funny that, with a company of 20 people, we’re already thinking about okay, how will this work when we’re 1000? But I believe that you lay the foundation for that culture really early in the life of a company. And then if you’ve done your job right, it’s self-reinforcing. You tend to attract people who want to contribute and add to that culture. If you get it wrong really early on, you did it with your hands. There’s the roots and then the tree. Maybe the tree will be skewed, and so I think it’s important to get it right. We spend a lot of effort on that.

Danny:

– That’s awesome. I would love to dig into that a little bit more if we can. I love the idea around culture, and it fits really well into your story there as far as how that transition goes. I love the concept of, you said a culture-add versus culture fit. What are some concrete things that you’re doing there to set the tone for the culture?

Florian:

– Yeah, maybe a little bit about the company. We’re a distributed team, so we have people all the way from northern reaches of Canada to South America. As a distributed team—and throughout the US also, from the East Coast to the West Coast. So that adds an extra layer of challenge when it comes to having people feel like they’re part of an organization that has a shared meaning and a shared objective. One of my—people don’t usually love meetings, but I do have a favorite meeting that we do every two weeks on Fridays. We call it the wins meeting. This is a meeting where we get together; there is no agenda, per se. People from across the company, we all get together. It’s not mandatory, so it’s not an official thing, but usually everyone is there. It’s really to share wins and lessons. Wins is, could be something that you did at work, you completed some development, and you want to demo to the rest of the team. Or somebody won a big deal, and they want to celebrate that. I think that’s pretty normal to do celebrations of things like that. On the other hand, we also do lessons. I think that’s really valuable, which is something that didn’t go according to plan. What did you learn? And sharing that with everyone else. This meeting kind of took a life of its own, and I’ll tell you a concrete story. We had an intern, and it was—one of the things we do is we rotate who is basically the emcee for the meeting, for the wins meeting. It’s a very participatory thing. We were having our wins meeting on May 4th, and for Star Wars geeks, that’s an important day. It’s May the Fourth day. We ended up; I said, well why don’t we do something with Star Wars? So this intern who happened to be hosting, she put up this amazing trivia game about Star Wars. Then we have some people at our company that are insane with trivia, like really, really good. They beat me every. Single. Time. But what’s funny is, this became kind of a tradition, so a lot of times we start the meeting with, tell us something about yourself, something embarrassing that you did or a trivia game. And it took a life of its own, and what I love about it is, it wasn’t like a top-down thing. We just created a space, and then it’s self-reinforcing. I don’t know. It’s a great way, like I said, my favorite meeting.

Danny:

– That’s awesome. That sounds like a great meeting. I love that concept of doing that and creating a cadence and letting it go beyond maybe what it was initially set up for and grow a little bit more organically. I think I see now a little bit more where you’re talking about the culture add aspect. I see that very clearly in that example. That’s fantastic.

Florian:

– So for example, we have an engineer on our team who does parasailing.

Danny:

– Very cool.

Florian:

– I’ll admit, I look forward to when he shares a video or something because it’s like vicariously living through his excitement. But you get to know people beyond just work. I think that’s really important for me.

Danny:

– That’s a great takeaway. I’m writing some notes here. I think we may be implementing something like that similar here. Anyways, I’m selfishly very excited about that. Thank you. What a great thing, talking about culture. It felt like a great, certainly an absolutely great segue as you’re building out InOrbit and maybe some of that inspiration. What’s the story behind how you started it? I know you mentioned that you said you blamed your cofounder there a little bit, but what’s the story there? How did you start? When did you say, we’re going to make this jump, and we’re going to start this venture?

Florian:

– As with many companies, it started with just a conversation about interesting problems to solve. Like I mentioned, my cofounder had been in the robotics space for quite some time. I came from the cloud world, so it was completely foreign to me. Maybe like a lot of people who may be watching this had my preconceived notions from movies and so on about robots. As I started learning more about it, I learned a couple of things. One is, the old-school robots that worked in factories, for example if you go to an automaker’s factory floor, you’ll have these massive robots, million-dollar robots. They just do the same thing over and over. They’re programmed in a very precise fashion to do the same thing over and over. What got me really excited was, there’s now a new generation of robots enabled by really new technology where the robots are coming into the open world. So now you’ll see robots in warehouses, in farms, on construction sites, at retail stores, in airports. So really pretty much wherever you go, there is a robot. I live in Mountain View in California, and when the pandemic hit we were really worried about our close neighborhood store, our grocery store, mom-and-pop shop, we know the owners. We’re like, okay, how are they going to hold up? Well this being Silicon Valley, within two weeks they were doing delivery to the home via robots.

Danny:

– Really?

Florian:

– What I thought was interesting was people’s reaction to the robots because at first it was just like, oh, my God, what is this? People taking pictures. That lasted about two weeks. Then it became part of the (of course) robots. So I think it’s interesting how we as consumers ourselves, we respond to the technology, first maybe in awe, but then it becomes part of our landscape. I think that’s what we’re seeing what’s really exciting about what’s happening with robotics. Back to InOrbit, we said, okay, great; how do we help? We see the potential. We see the impact. How can we add to that? Before we even started the company, within I think it was 25 interviews with founders of robotics companies, with users, really to understand what their problems were, we landed on this idea that, when a robot goes from the lab to the real world, there’s a big transition.

A lot of times, and this is not probably very well-known, but a lot of times in those early days, robots had chaperones. They literally have an engineer following the robot around to make sure that it behaves as needed. Again, in the early days you probably had more engineers than you had robots. Then as you start to scale, when you want to go from 5 to 500, let’s say, a lot of things that worked in the early days stop working. That’s something that I’ve seen in the cloud, and that’s where my connection to this was–– which is, I’ve seen this work in the cloud where in order to operate the massive data centers that companies like Facebook or Microsoft operate, you couldn’t do it one server at a time. You have to have these scalable tools. So we set out to build those scalable tools for robotics, and we landed on this concept that we call rob-ops. If you’ve heard of dev-ops, which is tools and practices for operating software at scale, we’re bringing a lot of those ideas to the world of robotics.

Danny:

– That’s awesome. Can you say that again? I had not heard that, and I’m not ashamed to admit that, so what was…

Florian:

– So in the cloud space, there’s something called dev-ops.

Danny:

– Oh, dev-ops. Okay, I thought you said something else.

Florian:

– Yeah, like development and operations put together. So now we’re doing rob-ops which is robotics operations as a cohesive idea.

Danny:

– Perfect. That makes a lot of sense, and that’s great. Great analogy there, or connection on that. I’m curious, though; how did—you said you talked to several robotics companies just wanting to get an understanding. I think you said 20 or so leaders and end users. What was that moment in your mind? It sounds like you’ve done this before, so it’s not like—you’ve gone through this process, but did you have a particular moment where you said, okay, yeah, this is it. This is a big problem that I know we can solve, and it’s time to go. Let’s do that. What was that moment like?

Florian:

– Yeah, maybe to the point that you made about having done this before, there’s a method to the madness of starting a company. It’s generally called customer discovery and customer development. Customer development is a framework by Steve Blank. You go through the process, and that’s part of what guided us. In our case, maybe the moment of insight was precisely this idea that robots are, if you want to think about it this way, it’s like servers on wheels. It’s like a complex computer server with lots of sensors and moving around. So it’s in the—if you designed a data center for hundreds of thousands of hosts or computers, you need reliable power. You need great networking. You need ventilation to dissipate all that heat, so you design it, and you control as much as you can.

A fleet of robots, I like to call it, it’s a data center from hell. It’s the worst possible way of operating a distributed computing platform because a lot of times they’re in uncontrolled environments. They have networks that come in and out. They’re on battery, so they need to come back on their own to recharge if they’re a mobile robot. So the challenges are much, much higher. When we talked to a lot of these robotics companies, what we realized is a lot of them weren’t necessarily thinking about those problems of scaling. If you’re very passionate about robotics, maybe you have a PhD and you work as a roboticist, you really have to focus on building that robot. We felt we could augment and help them focus on that, and we’ll take care of the boring part, if you will, of running the robots at scale.

Danny:

– Yeah, that’s exciting. So maybe an oversimplification of what you do. I’m a little nervous about saying this, but it sounds like you’re helping to solve—and there’s a lot of different challenges, but I know there’s a big challenge of interoperability, certainly, in this space because there are so—it’s interesting to see the growth of robotics over the last several years. It was a fast trajectory of growth, and then you pile on the pandemic. The huge needs just grew exponentially, and they continue. I can’t remember; there was a stat that I’ve seen as far as how many new robotics companies are on the—you have players who’ve been in here for years and years and years, and then you’ve got constantly new start-ups.

It’s interesting because, like you mentioned, a lot of people think, you watch Terminator or all these other robotics and AI movies. Oh, wow, it’s going to take over the world and all these things. The reality of it is, is even though that, yes, you’ve got from an automation standpoint you mentioned automotive, and you’ve got these giant, maybe these FANUC robots that have been there, and they’re moving, and they’re—but they’re very set in place, or they’re caged. You’ve got all of these AMRs that are now, they’ve got more autonomy, and they’re going around all over the place. It’s totally opened up the playing field. But the problem being that, that’s great, but you can still only solve a finite amount of problems with this particular robot. You may need a whole fleet of 10 different types of robots to solve some of these different challenges. Is that part of what you’re solving there, with some of the software solutions?

Florian:

– Absolutely. If I can get really geeky here—

Danny:

– Please, get really geeky.

Florian:

– For a minute, we like to say we solve the four Os of robot operations. So the first O is observability. This is kind of like a fancy word for monitoring, but it’s really knowing through data what’s up. What is the robot doing? It has its own set of challenges as I describe because they’re all over the place. They’re on weak networks sometimes. So we solved that problem, and that becomes the foundation because it’s the data. So at our core, we’re a data platform.

But then from there, there’s an operational aspect which is, robots get into situations where they’re no longer autonomous. They’re stuck. This happens—I won’t disclose numbers, but it happens a lot more than everyone anticipates. I think we have this picture of a robot; oh, it must be perfect. Far from it, especially when you’re in crazy environments. You talk about a warehouse, for example, but we’ve seen robots in retail stores while there are shoppers around. So there’s going to be a kid with a shopping cart and ketchup stains on the floor. You can’t expect the robot to be perfect, and that’s where some of these—augmenting the robot by human operators is a way to do that. So that’s the second O, operations.

I think then you get into optimization. What you really want is to not have those problems happen as often. How can you solve it? We do things like tools for root cause analysis. Root cause analysis allows you to identify cases where things didn’t go as planned and derive insights from it. And just to give you a sense of scale, each robot may generate a terabyte of data per hour. The approach of, we’ll just throw everything to the cloud isn’t really practical, especially as you’re thinking about we now have fleets of thousands of robots operating in the world. So that takes us—optimization is how do I run my fleet more efficiently?

Then the last O is getting to what you were mentioning before. We call it orchestration. Again, that’s a very cloud term, but it’s really, how do I bring different components together in a harmonious way? Now you’re more like the conductor of the orchestra. That can include many different types of robots from different vendors. In a warehouse, for example, you may have—let’s say there’s a cargo dock, and a truck comes in. Well you need to unload it, so there are now autonomous forklifts that can do part of that job, or maybe all of it one day. Then you have pallets. You need a depalletizing robot that will break down the pallet into its components. Now you need to take it into the warehouse proper. How do you do that? Do you do it one by one? Traditionally a lot of these warehouses have conveyor belts. The conveyor belts are really efficient, but they’re really expensive and they’re really inflexible. So what an AMR, an autonomous mobile robot gives you is flexibility because it’s controlled through software.

So that’s where we’re able to bring all of these robots from different vendors and give someone a complete view. The term in the industry is a single pane of glass, one way to see all of your IT, or in this case OT, in one view. I think that’s where this is going, and you see really pioneering companies like FedEx really advocating for this ability to control different robots.

Danny:

– That’s super interesting. Thank you so much for getting, I think you said geeky with me. Getting into the technical weeds. I appreciate that. Actually one thing in particular that I had no idea, never heard this before: one terabyte of data per hour, one robot could generate. That’s a ton of data, especially if you’ve got a fleet of hundreds of these things. Transmitting that and getting that, storing it and then analyzing it seems—That’s crazy. That’s a lot of data.

Florian:

– A lot of times, Danny, all of those terabytes of data, you just need this one piece of data, the proverbial needle in the haystack. I think that’s what we do is, we try to turn all of that data into insights that people can take action on.

Danny:

– Right, absolutely. Well there’s a ton there. A couple other questions that I’ve got coming up then, how do you engage with your customers? Are your customers—obviously you talked about robotics companies. So I imagine, would your customers be robotics companies, end users, and integrators? Do you work with one, all—how does that work?

Florian:

– We work with different players in the ecosystem. At the end of the day, because of our orientation towards operations, we really care about maximizing the potential of every robot. As a robot vendor, of course you want a more efficient robot, a robot that fails less often or that requires fewer interventions. So we can provide a lot of data for them to improve on their robot. As an end-user, you’re really running the robots. In fact, I like to say, and this is a little shocking for companies in the robotics space, the end users, they don’t really care about the robots. They care about their process and what the robot can do to it. Their needs are going to be a little different. They’re all based on the same foundation of data, but what they’re trying to do is, they’re trying to drive this SLA higher and higher and higher.

What’s interesting is there’s this middle ground between those two cases that I mentioned which is this idea of robotics as a service. Increasingly you see companies that, instead of selling you a robot and now you do whatever you want with it, they’ll essentially sell you a service that is backed by their own robots. We’re starting to see—this is very new—third party companies that will say okay, I’m not going to make the robot. I’m not an end-user. I’m just going to run these robots at scale. So we can help them with our infrastructure to do that like I said at scale, across robots, across sites, and across customers.

Danny:

– Yeah, that sounds super interesting. I have heard, definitely over the last little bit or so, robotics as a service. I think it’s a super fascinating play. Certainly with all the labor issues that we’ve got going on now, certainly a lot of companies are turning towards automation in any way—there’s a huge shift in the digital transformation that is happening. It was happening before, but again because of the pandemic in all areas, you’re seeing the supply chain, manufacturing, all up and down, logistics, and beyond. End users, you were mentioning retail and grocery for example. I love the store that you talked about, how the response with the local grocery store in the neighborhood, you said being Silicon Valley, what do we do? It’s fascinating, but we’re continuing to see more of that. And we’ll see more. What does the future look like for you?

Florian:

– I like to think of this future where we have millions of robots around the world that are positively impacting billions of people. Each robot can have this multiplier effect, whether it’s a robot in a farm making—so one of the companies that we’re familiar with is a, basically a robot that goes in a farm, identifies weeds, and then literally zaps them with lasers. Instead of having to have herbicides and spray it, you can turn that to zero. Another customer of ours has been able to grow leafy greens with a 90% reduction in the use of water. Again, all of these are things that are not moving the needle a little bit. They’re moving it a lot. We’re seeing a lot of electrification also which a lot of times goes with autonomy together. We see that reducing, for example, the carbon emissions. These little robots that deliver around Mountain View, they’re replacing somebody driving around in their car, getting out of the car, dropping the groceries, and getting back in the car. I think these things have huge impact.

But to take it a little bit further, where we see the future is the collaboration between robots and people. I think we’re only starting to see the shape of that. I like to talk about this what we call a roboteer, like a puppeteer, but for robots. Maybe your job isn’t to go deliver food or isn’t to unload a truck. You’re monitoring. You’re the boss of a group of robots that are doing those things. Just like we said, the robots may not work fully autonomously, fully independently all the time. There’s always going to be corner cases or areas where somebody needs to intervene. But now maybe you can do–– instead of sitting in a forklift and driving it around–– maybe you can supervise 5 forklifts or 10 forklifts. You don’t even need to be there. You could be doing it through our platform or through other people’s platform. You could be in an air-conditioned room sitting in an ergonomic chair and responding to events as they happen. I think that vision of collaboration between people and robots and obviously the cloud and AI that come into that, I think that’s a great, in my mind a very optimistic vision of the future.

Danny:

– That’s really interesting. I like that, and it’s interesting. People always harp on, “Robotics is taking people’s job away,” and the answer we keep hearing is—I love this; I haven’t heard this, roboteer. It makes a lot of sense. You’re seeing that’s a—I do certainly believe that is where the future is going, too, when you look at a lot of these jobs as well. Obviously we’ve got major issues right now for a whole host of reasons, but the trend for years though before this has been, a lot of people, they don’t want to work. I don’t want to work in a warehouse, or I don’t want to work in manufacturing or logistics. There’s been a struggle trying to attract, so if there’s a way to say listen, we’re already seeing a decline in that. If we can bring in a solution over here from a technology standpoint to be able to assist people in those jobs, it’s creating an increased net positive effect.

Florian:

– Yeah, and you say I don’t want to work. I think it’s—these are tough jobs. I had the opportunity to tour some facilities, and without automation your job may be to literally walk 15 miles a day within the warehouse pushing a cart back and forth. Let me tell you, that’s not good on your knees. It may be super healthy because you’re walking a lot, but it’s not necessarily the best environment where you want to do that. What we’ve seen is, we’ve seen this tremendous increase in productivity while also improving the health of the workers where you can have, instead of somebody pushing a cart, the cart knows where to go. When you think about a robot, some people may have this idea of a humanoid robot. Our robots have very boring shapes. They resemble something that has existed all along, but now they add autonomy to it. So if the cart can go where it needs to go, now you’re focused on picking the right products and putting them on the cart, not on walking back and forth. So that has the dual effect of better, less wear and tear on the body and also greater productivity. Some companies will see 300% productivity increases by introducing an AMR. I think it can very much be a win-win.

Danny:

– Certainly, absolutely. Alright, my last question is a very difficult question. Since you brought up Star Wars earlier and you’re working in the robotics space, I’m curious. What’s the difference between a robot and a droid?

Florian:

– Ah, I think it’s just a terminology thing. Everyone knows R2-D2.

Danny:

– Who’s that? Sorry; no, I’m just playing.

Florian:

– So that is actually called an astromech droid. This is me showing off my Star Wars nerdiness. They do have different models. Not all of them are R2-D2. You do have, in the Star Wars universe, different types of droids that do different things. They already had this orchestration problem in Star Wars, and maybe they’re using InOrbit—you know, InOrbit, star—maybe there’s something there.

Danny:

– Maybe, and it’s a great parlay from the question of the future. So there we go. Listen Florian, I really have appreciated our conversation. Thank you for obliging. I loved your answer. Thank you so much on that, and for just sharing all of your insights about the industry, what you guys are doing, sharing your story and how you got to where you are today. It sounds like you guys have a very bright future, no pun intended with the Star Wars analogy. But I really do think that you guys are solving some big challenges and creating some unique opportunities in a space that is exploding. I’m going to be very excited to follow you guys. So for those companies who would like to learn more about you, I’ve got the website here, or the URL, inorbit.ai. I want to make sure that’s correct. We’ll have that in the show notes.

Florian:

– You got it.

Danny:

– And Florian, again, just thank you so much for your time with us today.

Florian:

– Danny, it’s been a lot of fun. Thank you so much.

Danny:

– I have enjoyed it too. Alright. Alright, well that wraps up today’s IndustrialSage Executive Series with Florian Pestoni who is the CEO and cofounder of InOrbit. You can go check them out at inorbit.ai. I loved today’s episode. Great story; a lot of great insights here.

So if you like today’s episode, you need to go to IndustrialSage.com if you’re not there right now, and you can subscribe because you’re missing out on great content like this. Not only do we have the Executive Series. We also have news, and we’ve got a lot of other great content coming down the pike. You don’t want to miss it; I guarantee it. So subscribe; don’t miss out. And I’ll be back next week on another episode of IndustrialSage.

Thanks for reading. Don’t forget to subscribe to our weekly newsletter to get every new episode, blog article, and content offer sent directly to your inbox. You can also subscribe wherever you download podcasts so you can listen on the go!

Sponsored by Optimum Productions

  continue reading

241 episodes

Tous les épisodes

×
 
Loading …

Bienvenue sur Lecteur FM!

Lecteur FM recherche sur Internet des podcasts de haute qualité que vous pourrez apprécier dès maintenant. C'est la meilleure application de podcast et fonctionne sur Android, iPhone et le Web. Inscrivez-vous pour synchroniser les abonnements sur tous les appareils.

 

Guide de référence rapide