Latest / Automated with Brian Heater / Andrei Danescu on Why the Market Does Not Want More Robots

Transcript

Andrei Danescu on Why the Market Does Not Want More Robots · Automated with Brian Heater

• 0:00
  SPEAKER_01: We're here today to talk about one thing.
• 0:03
  SPEAKER_01: Race cars.
• 0:04
  SPEAKER_01: Yeah.
• 0:05
  SPEAKER_01: Maybe robots as well.
• 0:07
  SPEAKER_01: How much overlap is there between being a mechanic working on a car and robotics engineering?
• 0:14
  SPEAKER_00: People working at the cutting edge of car racing and having a good understanding of how to push systems to the extreme.
• 0:20
  SPEAKER_00: Those skills are very transparable and very well suited to kind of robotics engineering.
• 0:25
  SPEAKER_01: Some of the like initial ideas, hospitality, retail.
• 0:29
  SPEAKER_01: It's interesting to me that somebody in logistics would hop on the phone and say, hey, you have an interesting product.
• 0:34
  SPEAKER_01: Can you do something completely different with it?
• 0:36
  SPEAKER_00: To be honest, we didn't jump at the opportunity straight away.
• 0:39
  SPEAKER_00: We asked the market.
• 0:40
  SPEAKER_00: During the pandemic, a lot of opportunity came for the logistics industry, but a lot of problems were compounded.
• 0:46
  SPEAKER_00: It's an environment where the dynamics are really high.
• 0:49
  SPEAKER_00: The challenges you have from a technological perspective, to be able to localize and to be able to navigate this environment with the level of accuracy that we need, and we're talking millimeter precision.
• 0:59
  SPEAKER_00: I mean, this stack behind me is by far the most advanced autonomy stack in the world.
• 1:03
  SPEAKER_00: Heads down.
• 1:04
  SPEAKER_01: How are you able to, especially indoors, how are you able to localize with that accuracy given how relatively uniform all the aisles are?
• 1:14
  SPEAKER_01: So we have developed the hello everyone.
• 1:29
  SPEAKER_01: My name is Brian Heater.
• 1:31
  SPEAKER_01: I am the managing editor at the Association for Advancing Automation.
• 1:35
  SPEAKER_01: Uh we have a we have a cool episode for you this week.
• 1:37
  SPEAKER_01: We actually flew down to Nashville to record this conversation with Andre Donescu.
• 1:42
  SPEAKER_01: Uh the UK-based firm Dextery was hosting a big event to mark their expansion into the US.
• 1:49
  SPEAKER_01: It's a fun conversation, and I think you'll enjoy it.
• 1:51
  SPEAKER_01: So uh thanks to everybody who has subscribed to the channel recently.
• 1:55
  SPEAKER_01: Getting those numbers up is exactly what allows us to do the show.
• 1:58
  SPEAKER_01: So please like and subscribe if you haven't already.
• 2:01
  SPEAKER_01: And with that, please enjoy this conversation with Audrey and Mesco.
• 2:07
  SPEAKER_01: Record MOI are changing what's possible in industrial automation from quantum inspection and predictive maintenance to robotics, logistics, medical devices, or space, and manufacturing for that version of MOI conference.
• 2:20
  SPEAKER_01: We'll hear from industry leaders to see real-world AI performance conduction.
• 2:25
  SPEAKER_01: Excellent to meet version of computer version of sensors and computing and automation technology.
• 2:31
  SPEAKER_01: You'll also be able to connect with people who are thinking of what comes next.
• 2:35
  SPEAKER_01: Whether you're building, buying, or implementing these systems, this is a chance to stay ahead of the curve and see how vision and AI are moving from promise to practical impact.
• 2:45
  SPEAKER_01: Learn more and make sure to buy your ticket early at the Advanced Vision and AI conference September 23rd and 24th in Santa Clara.
• 2:56
  SPEAKER_01: So we're here today to talk about one thing race cars.
• 3:00
  SPEAKER_01: Yeah.
• 3:00
  SPEAKER_01: Maybe robots as well, but we can start race cars.
• 3:04
  SPEAKER_00: That was like your that was your passion, right?
• 3:06
  SPEAKER_00: It was.
• 3:07
  SPEAKER_00: Actually, I didn't like cars much before I uh before I turned 18, before I got my driver's license.
• 3:12
  SPEAKER_00: But I've always been a very competitive person and I love sports.
• 3:16
  SPEAKER_00: I really love sports.
• 3:17
  SPEAKER_00: I used to play basketball, I used to play all kinds of sports, and then I discovered cars, and then I liked sports, and also I love engineering.
• 3:26
  SPEAKER_00: Like the engineering's been my my lifelong passion and real love.
• 3:30
  SPEAKER_00: So race cars kind of take all those boxes.
• 3:32
  SPEAKER_00: It's a team sport, um, it's competitive, and it's very high, high degree of engineering complexity.
• 3:38
  SPEAKER_00: And and that's effectively what moved you from Romania to the UK?
• 3:43
  SPEAKER_00: Yeah, so there's not um even now it's it's a little bit better, but realistically, there's not a real industry for motorsports in Romania.
• 3:50
  SPEAKER_00: Whereas in the UK, when you look at it, you have uh the vast majority of the Formula One teams, you have so many different championships.
• 3:57
  SPEAKER_00: Um, there's education programs, there's engineering degrees you can get in motorsport engineering and so on.
• 4:03
  SPEAKER_00: Um so it's like then it was the best chance at getting a job in motorsport.
• 4:08
  SPEAKER_00: Um, so I I couldn't resist it.
• 4:10
  SPEAKER_01: What is the I mean, is there a robotics community in Romania at all at this point?
• 4:17
  SPEAKER_00: I think there is, um, and it's growing.
• 4:19
  SPEAKER_00: Uh it's growing through uh multiple levels of uh complexity as well, from university.
• 4:26
  SPEAKER_00: It started at university, I would say, um, and is now making its way through high school and uh secondary school.
• 4:32
  SPEAKER_00: People are passionate about it, and it's it's definitely uh gaining a lot more um adopters and a lot more people that compete.
• 4:39
  SPEAKER_00: They compete in international events, in the first events and multiple other categories as well.
• 4:44
  SPEAKER_01: It's funny because you and your two founders are both from there originally.
• 4:48
  SPEAKER_01: You all ended up, like I guess, separately in the UK and then started a company over there.
• 4:54
  SPEAKER_01: Now that you have this successful company and your London base, you know, are you able to sort of give back to that community?
• 5:00
  SPEAKER_01: Are you able to mentor people back from where you're from?
• 5:04
  SPEAKER_00: Yeah, absolutely.
• 5:05
  SPEAKER_00: And and we always try, we always try and get involved with multiple different types of competitions.
• 5:11
  SPEAKER_00: I think I personally think they're one of the best ways for uh for the younger generation to learn.
• 5:17
  SPEAKER_00: I, you know, the way I got into autonomous robots was through a uh robot football playing competition in uni.
• 5:23
  SPEAKER_00: Um and then you look at Formula Student if we if we make the parallel to race cars.
• 5:28
  SPEAKER_00: Um that's you know, Formula Student is a it's a really it's a great opportunity to get involved with something that's it's actually quite complex but uh and and relatively open-ended in terms of regulation, but it gives the students the opportunity to really understand what practical value those those skills are.
• 5:45
  SPEAKER_00: So uh we sponsor a couple of formula student teams, we get involved with a number of robotics competitions, um, and now as as our company is also gaining momentum, we're we're gonna be looking at um kicking off graduate schemes, we're gonna be looking at actually sort of like um having open door events to inspire the next generation.
• 6:04
  SPEAKER_00: I think that's really important.
• 6:05
  SPEAKER_01: Yeah, because we see we said you know, race, race cars, formula or one, you know, is your passion, but like robotics really was your first love, right?
• 6:14
  SPEAKER_01: I mean, that's what you were like what it according, I don't know, I don't know how much a trust uh accompanies about page, but according to your about page, you built your first robot at eight.
• 6:23
  SPEAKER_00: You're pretty much spot on.
• 6:25
  SPEAKER_00: Engineering and and especially robotics has been is has been my my my one first true love.
• 6:30
  SPEAKER_00: And um it's also shaped my career.
• 6:33
  SPEAKER_00: Like even throughout school, I was always deep.
• 6:36
  SPEAKER_00: It's it's also shaped what they do, yeah.
• 6:39
  SPEAKER_00: Um but throughout school, that was my interest.
• 6:41
  SPEAKER_00: It was it was math, it was science, it was uh physics.
• 6:44
  SPEAKER_00: Um that was that was really the bit that made me tick.
• 6:48
  SPEAKER_00: And having the opportunity or jumping at the opportunity to start something in this space, uh you if you care deeply about something, you will find a way, and you will find a way to kind of follow that passion through.
• 6:59
  SPEAKER_01: Yeah, I mean, how much how much overlap is there between uh like being a mechanic, you know, uh working on again a car and robotics engineering?
• 7:11
  SPEAKER_00: I think there's quite a lot of there's quite a lot of overlap.
• 7:14
  SPEAKER_00: So, for example, the role I played in in Formula One and my profession, let's say, was um electronics and control systems.
• 7:22
  SPEAKER_00: But because of where I was positioned in the team, I had the opportunity to work with an incredible team of mechanics.
• 7:28
  SPEAKER_00: I had an opportunity to work with engineers that were spending their time mostly away from the car, but also get involved with how the garage was built, how we were actually collecting data, collecting telemetry information, feeding this information to the governing body, making sure that you know all our sensors are calibrated, the cars are legal, and so on and so forth.
• 7:47
  SPEAKER_00: But now, as we look at scaling the business, we we basically recruit and we have people from all kinds of uh different industries, and there's quite a lot of uh I would say motorsport heavy um specialties in wiring, in in mechanics, mechanical engineering, and so on.
• 8:05
  SPEAKER_00: So I think especially when you look at motorsport, they're they're high performance, high performance machines, right?
• 8:12
  SPEAKER_00: And a lot of robots, this is where I think uh, and we can talk about this a bit later as well, but this is where I think people forget that robotics is uh hardware and software and and business models pulled together.
• 8:24
  SPEAKER_00: You know, the the robots actually need to be really good quality once you deploy them in the world.
• 8:29
  SPEAKER_00: And the real world is a difficult environment.
• 8:31
  SPEAKER_00: It's not about having a nice demo in the lab.
• 8:34
  SPEAKER_00: So people working at the cutting edge of car racing and having a good understanding on how to push systems to the extreme, are those skills are very transferable and very well suited for you know for high-end robotics engineering and and other areas that support autonomous systems.
• 8:50
  SPEAKER_00: So, in what sense is a race car software?
• 8:54
  SPEAKER_00: I think in a lot of senses, especially if we look at the um the first race this year, where a lot of people were complaining that the software let them down.
• 9:02
  SPEAKER_00: Uh yeah, I know when you look at hitting the.
• 9:04
  SPEAKER_00: It's one of those things that you start to notice when it fails, right?
• 9:07
  SPEAKER_00: Yeah, exactly.
• 9:08
  SPEAKER_00: But um no, I mean it's similar to modern cars, right?
• 9:13
  SPEAKER_00: Like it's it's very much software-defined.
• 9:15
  SPEAKER_00: There's a lot of control systems that run how this car works.
• 9:20
  SPEAKER_00: Um, you know, a modern F1 car to be able to actually push things to that uh that level of performance, you need to have very tight control loops.
• 9:27
  SPEAKER_00: You need to have state-of-the-art systems that can run the engine, uh, that can run the battery, make sure that you you get to that level of performance without adding a huge amount of mass and so on and so forth.
• 9:38
  SPEAKER_00: So actually, without uh a very, very tight control between your electromechanical, hydraulics, and software control systems, it would just not work.
• 9:48
  SPEAKER_00: You could you can break the car by sending the wrong software to it.
• 9:52
  SPEAKER_00: You can also break the car by not building it properly and by putting the wrong hardware in it, but I think you know that's that's the implication it could have.
• 10:00
  SPEAKER_01: Yeah, I wonder, you know, we talk a lot about um uh upskilling or reskilling, and often like in in this sort of context, right?
• 10:08
  SPEAKER_01: We're we're here in in this warehouse, you know, this idea of people who labor their entire life and working on the floor and teaching them to program robots or to work with robots.
• 10:19
  SPEAKER_01: And I wonder if maybe there's also if there is an analog between um working on cars and doing robotic hardware and software, whether there's an opportunity there as well.
• 10:32
  SPEAKER_00: Yeah, I think I think there's a big opportunity, to be honest.
• 10:35
  SPEAKER_00: Um, I think when you look at the old ways of doing work and the old ways of going about even the logistics industry as a whole, there's a massive labor shortage.
• 10:46
  SPEAKER_00: So we need it's it's not so much about only retraining people, it's also about making sure that we can make the most out of the hours in the day that you spend working, making sure that you you bring the right level of automation and you enhance people's abilities to do work.
• 11:01
  SPEAKER_00: So it's not I wouldn't pigeonhole it down to skills alone.
• 11:05
  SPEAKER_00: Because you can have really good skills, but you might not have the right tools or you might not have the right environment.
• 11:09
  SPEAKER_00: It's about creating a the industry, the logistics industry of the future, where you can actually deploy these technologies and you can be a lot more efficient and a lot faster at what you do, regardless of what that job actually is.
• 11:21
  SPEAKER_01: You know, the the reason why I was asking from uh from a skill set uh perspective was because you know I'm I'm I'm wondering whether uh you have a bit of a head start if you're a mechanic.
• 11:34
  SPEAKER_00: So I think if you if you go into servicing robots, um, because they're they're mechanical systems, electromechanical systems actually, um yes, you can have a you can have a head start, but um, you could also have a head start by being an electronics engineer or electronics technician that fixes complex systems.
• 11:52
  SPEAKER_00: Um good example, I think, uh was uh was back in the day.
• 11:55
  SPEAKER_00: You you have uh, for example, technicians used to servicing big printers.
• 12:01
  SPEAKER_00: And you know, that is a skill that would be directly applicable to robotics because they're complex electromechanical systems, regardless of the end function or the end job that they actually end up uh doing.
• 12:13
  SPEAKER_01: This this uh this barbecue where you and your two co-founders were first, I guess, first started brainstorming about uh the company that became Dexary, was it just a bunch of like expats?
• 12:25
  SPEAKER_01: Was it a bunch of Romanian folks, Romanian tech folks?
• 12:28
  SPEAKER_01: What was kind of the theme?
• 12:29
  SPEAKER_01: How did everybody know each other?
• 12:31
  SPEAKER_00: Um no, so I I never I was never like a big fan of this this concept that sometimes people have.
• 12:38
  SPEAKER_00: You move to a different country and you kind of like want to stick to a certain community.
• 12:42
  SPEAKER_00: I'm like, if you really, if you really just want to stick to people, you know, like why why did you decide to move countries, right?
• 12:48
  SPEAKER_01: I mean, people like some, you know, you like the comforts of home to uh you know a certain extent, I guess.
• 12:53
  SPEAKER_01: I don't know.
• 12:54
  SPEAKER_00: I don't know.
• 12:54
  SPEAKER_00: It's just not something that was that was my thing.
• 12:57
  SPEAKER_00: So I always I always love to make friends and and always love to to kind of meet people and and learn about new cultures and new way of of looking at the world.
• 13:05
  SPEAKER_01: You did end up starting a business with two Romanians, so you know there's something something there.
• 13:10
  SPEAKER_00: That's also because we've we've known each other for we had a a common group of friends.
• 13:19
  SPEAKER_00: Um and then we um we kind of left and decided to pursue different careers, and we we happened to meet back in the UK.
• 13:26
  SPEAKER_00: Um and we thought it'd be a great idea to to have a robot that could collect data, map a physical space, transform this into insights, and then use it for businesses to make better decisions.
• 13:37
  SPEAKER_00: But uh I would say you know, some of the the first engineers, definitely one was from Scotland, and uh, you know, there's there's a bunch of different people that were from um from the UK or um had moved to the UK before us.
• 13:51
  SPEAKER_01: So yeah, I want to walk through some of the like initial ideas.
• 13:55
  SPEAKER_01: Um you know, I know hospitality, uh uh retail.
• 14:02
  SPEAKER_01: Wait, uh what were you kind of looking for in in those spaces?
• 14:06
  SPEAKER_01: Was it similar, you know, like in terms of retail, were you looking to do like sort of front-of-house inventory?
• 14:11
  SPEAKER_01: Yeah, pretty much.
• 14:12
  SPEAKER_00: So basically, it's quite interesting because uh when you look at the foundational idea of the company was to use autonomous robots, collect data from physical spaces, transform it into insights, and then early days place this information on a 2D map.
• 14:29
  SPEAKER_01: So this would maybe be like what like we have Simba here, like that sort of thing, the robot kind of rolling through the supermarket.
• 14:36
  SPEAKER_00: Yeah, so so Simbi Robotics is a good example, but that came a little bit later.
• 14:40
  SPEAKER_00: So we went even before that.
• 14:41
  SPEAKER_00: So if you remember uh the pepper robots.
• 14:44
  SPEAKER_00: Oh yeah, yeah, Softbank.
• 14:46
  SPEAKER_00: Yeah, exactly, Softbank.
• 14:47
  SPEAKER_00: So we looked at a hospitality but fully autonomous robot that could actually guide you around spaces, could go around a hotel um lobby or go around an airport or a museum and so on.
• 14:59
  SPEAKER_00: Um, you know, you could ask it questions, you could provide information, and it would actually show people, you know, what kind of information I've provided in specific areas and actually helped streamline the operations in that space.
• 15:11
  SPEAKER_00: Um I would say it's probably not a groundbreaking use of robotics considering that it didn't pick up very well.
• 15:18
  SPEAKER_00: Uh but then we went to retail, and it's very similar to what you're saying about CMB robotics, is um we were focusing not so much on groceries, we were working with um electronic stores, and it was pretty much a small-scale application of what we have today.
• 15:32
  SPEAKER_00: So we had a digital Twin A 3D layout of the store, and we could position all the products and we can take pictures and show how they're being presented to the customer.
• 15:41
  SPEAKER_01: Obviously, we're not we're not here to uh really litigate pepper.
• 15:45
  SPEAKER_01: Um, but the sense that I always got was you know they had this really cool robot and they were always trying to figure out you know something something to do with it.
• 15:52
  SPEAKER_01: And and that's maybe not the best way to approach building a robot.
• 15:57
  SPEAKER_01: Um I mean, I could make the argument that we were guilty of of that to some extent as well.
• 16:02
  SPEAKER_01: But you didn't, I mean you had sort of a broad, kind of abstract idea, and you like this wasn't like hanging out and you're like, oh, okay, what are we gonna do with this thing?
• 16:10
  SPEAKER_00: No, no, this this is a different part of the story, and this is definitely working, you know, working very, very closely with the market and understanding what problems you actually solve with what parts of technology.
• 16:21
  SPEAKER_00: Um so going back to the Pepper example, I think the problem was exactly what you're saying, is there's no clear use case.
• 16:28
  SPEAKER_00: It was sold as a product to a bunch of companies that were using it as an innovation kind of showcase.
• 16:34
  SPEAKER_00: Um so it never really had um It was a novelty.
• 16:38
  SPEAKER_00: Yeah, it was a novelty, but it didn't really have a strong foundation in a in a use case.
• 16:42
  SPEAKER_00: And when we looked at Texas Review and the current robots that we have behind us, um it was actually during the pandemic when we started getting a lot of people from logistics who saw some of the applications we had for retail asking us, hey, could you guys look at adapting the technology and actually scanning our warehouses?
• 17:00
  SPEAKER_00: Because we're really struggling with getting real-time data.
• 17:03
  SPEAKER_00: So during the pandemic, a lot of businesses moved online.
• 17:06
  SPEAKER_00: So a lot of opportunity came for the logistics industry, but a lot of problems were compounding.
• 17:11
  SPEAKER_00: Because the more you wanted to put through a warehouse, it was difficult to get data, it's difficult to streamline the operations, so the more issues you're generating.
• 17:20
  SPEAKER_00: So it was kind of like a diminishing returns problem.
• 17:23
  SPEAKER_00: So the way we actually ended on this version of our technology is working very closely with the market and understanding what are the needs, what do people really want.
• 17:33
  SPEAKER_00: Uh, and what's the form factor in the technology that answers those needs.
• 17:37
  SPEAKER_00: That's why we have ground-based robot that is very high because you can scan at an incredible speed, it will do well over 10,000 pallets an hour.
• 17:45
  SPEAKER_00: You can deploy it in any operation without having to add additional infrastructure.
• 17:49
  SPEAKER_00: But very importantly, we we have also built the digital platform where you can consume this information and where people go to get those insights and really understand how they can run their warehouses and how they can run their factories and how they can run their logistics operations.
• 18:05
  SPEAKER_01: See, this is something that's really interesting to me because you know I I was imagining it sort of happening in the opposite way where you know you you this obviously this pandemic is happening.
• 18:17
  SPEAKER_01: Um there isn't a hospitality market in the same way there there was before.
• 18:23
  SPEAKER_01: You know, maybe maybe people well people certainly aren't going into stores like they were before, and and you know, you're reaching out to uh to logistics folks, but it it sounded like it happened the opposite way.
• 18:35
  SPEAKER_01: And it's interesting to me that somebody in logistics would hop on the phone and say, hey, you have an interesting product.
• 18:42
  SPEAKER_01: Can you do something completely different with it?
• 18:45
  SPEAKER_00: I mean, you have people with with vision in a lot of industries.
• 18:49
  SPEAKER_00: We happened to have a good relationship with a company that was doing fast-moving cargo handling.
• 18:54
  SPEAKER_00: They had seen some of our concierge robots at Hitro.
• 18:58
  SPEAKER_00: Um, so we were kind of like the first uh top of mind when they had an idea for this.
• 19:02
  SPEAKER_00: You were just kind of in the right place at the right time.
• 19:04
  SPEAKER_00: Right place at the right time.
• 19:06
  SPEAKER_00: And to be honest, we didn't jump at the opportunities straight away.
• 19:09
  SPEAKER_00: So we had a technology that would work in a in a retail store.
• 19:12
  SPEAKER_00: It's maybe what, a hundred meters by a hundred meters.
• 19:15
  SPEAKER_00: Then we wanted to put in a warehouse that is a thousand meters by four or five hundred meters.
• 19:20
  SPEAKER_00: So it's orders of magnitude more complex.
• 19:23
  SPEAKER_00: It's taller, it's a lot harder, and it's a lot harsher as an environment.
• 19:27
  SPEAKER_00: So we we kind of like took a step back and we wanted to really understand is this just gonna be a nice to have a gimmick in a market, someone's cool idea about how they could use autonomous technology, or is this really gonna be having the opportunity to turn into a product that can transform the industry?
• 19:44
  SPEAKER_00: So we did actually a lot of work with the the first version of the robot, what we had for retail, which was clearly not gonna be capable of scanning the heights we needed or returning the data we needed.
• 19:55
  SPEAKER_00: But it was that that listening to the customers, asking the right questions, that early market validation that determined us to really completely pivot.
• 20:03
  SPEAKER_00: And it was a hard pivot.
• 20:04
  SPEAKER_00: We never looked back and focused fully on the logistics industry.
• 20:08
  SPEAKER_01: So just to back up a little bit, so you did have uh robots out in the world prior to getting into this space?
• 20:16
  SPEAKER_00: Yeah, so we had a couple of uh pilot deployments and early deployments with an airport space with Hitro and around Hetro with a couple of companies that wanted to use them as um sort of like robots in the reception area to help speak multiple languages and guide people.
• 20:31
  SPEAKER_00: But also we had a couple of uh pilots in retail, exactly what you're mentioning earlier.
• 20:35
  SPEAKER_00: So being able to scan the environment and being able to collect that information.
• 20:39
  SPEAKER_00: So we had a bit of an understanding of how hard it is to actually put a real autonomous robot in the world.
• 20:44
  SPEAKER_00: Um, but I would say this is this is orders of magnitude different.
• 20:48
  SPEAKER_01: It yeah, it's it's different.
• 20:50
  SPEAKER_01: I and but and there are uh there are obviously parts of it that that I assume are much more difficult, but there are also perhaps parts that are, I don't want to say easier, but a little more straightforward, right?
• 21:03
  SPEAKER_01: In in working behind the scenes versus working like front of the store and working in a busy airport.
• 21:08
  SPEAKER_00: I think you're working in a you're working in an industrial environment with people that you know have a certain level of training, you know, they're used to machinery, although there are forklifts which which move a lot faster and so on and so forth.
• 21:20
  SPEAKER_00: But I would say it's also an environment where it the dynamics are really high, but also the similarities are very, very high.
• 21:30
  SPEAKER_00: So if you look at a warehouse, it has you know hundreds of aisles, they all look the same.
• 21:34
  SPEAKER_00: So the challenges you have from a technological perspective to be able to localize and to be able to navigate this environment with the level of accuracy that we need, and we're talking millimeter precision.
• 21:45
  SPEAKER_00: I mean, this stack behind me is by far the most advanced autonomy stack in the world has now.
• 21:51
  SPEAKER_00: So if you think about it, you have to scan this warehouse continuously every day, and you cannot have errors because then your cloud point and the results you bring to your Customers will mean nothing.
• 22:01
  SPEAKER_00: You have to have a level of accuracy that makes it more an autonomous scanner than a simply autonomous robot.
• 22:07
  SPEAKER_00: You have to be able to deploy in an environment where people must do their jobs.
• 22:11
  SPEAKER_00: So, you know, they you cannot get in the way of people's work.
• 22:15
  SPEAKER_00: They have to move pallets around, they have to be able to kind of retrieve materials, feed a production line, push this forward to other customers, and so on.
• 22:23
  SPEAKER_00: So the implication of the technology not working at the level of performance that you expect or the customers expect are also way, way higher than having this deployed in a front of house where there's more people around it and maybe not more people around it, but there's untrained people around it, I would say.
• 22:42
  SPEAKER_01: How are you able to, especially indoors, how are you able to localize with that accuracy, again, given um how relatively uniform all the aisles are?
• 22:54
  SPEAKER_00: So we have developed the entire tech stack specifically for this use case.
• 22:58
  SPEAKER_00: So when you look at these robots, maybe I should have mentioned this as a business we're full stack.
• 23:03
  SPEAKER_00: So we design every part of this product in-house.
• 23:06
  SPEAKER_00: Design, engineer, test, hardware and software.
• 23:10
  SPEAKER_00: Hardware and software, manufacture the robots in-house, and we also work directly with customers to do the installs.
• 23:16
  SPEAKER_00: I mean, you would have to manufacture this because like who else is going to be.
• 23:22
  SPEAKER_00: It's a choice you make, right?
• 23:23
  SPEAKER_00: Yeah.
• 23:23
  SPEAKER_00: So if you think of uh Autostore or if you think of the Ocado technology robots, similar kind of picking and packing robots, right?
• 23:30
  SPEAKER_00: They didn't make those robots in-house.
• 23:32
  SPEAKER_00: They they work with external uh external partners.
• 23:34
  SPEAKER_00: There's a lot of and this used to be also the industry trend.
• 23:38
  SPEAKER_00: There's a lot of investors that would push companies that had ideas, or you have to work with an external manufacturer.
• 23:44
  SPEAKER_00: We have fought very many battles around this, and uh we were very bullish on the full-stack approach from the beginning.
• 23:50
  SPEAKER_00: Because I think if you have something truly new and unique that you want to bring into the market, you have to be the expert of how that product is built.
• 23:59
  SPEAKER_00: Because that gives you the opportunity to work with your customer, to learn faster, and then you have this loop that repeats itself, and you can make your product better, and then you have a happier customer.
• 24:09
  SPEAKER_00: If you rely on someone else doing the hard work, you're not gonna be fast, and you're also giving all the learnings away, and you're not gonna be able to, you're just not gonna be able to serve the market really well.
• 24:19
  SPEAKER_01: Giving it away to the to the supply chain.
• 24:23
  SPEAKER_00: Yeah, I mean, giving it away to the manufacturer who wants to make your product, they don't want to learn how to serve your customer best.
• 24:30
  SPEAKER_00: They want to make your really good product to your specification.
• 24:33
  SPEAKER_00: If you think about it, a parallel you can you can derive here is SpaceX.
• 24:37
  SPEAKER_00: You know, they didn't just make their rockets and gave it gave them to someone else, or they didn't just have this concept of reusable rockets or whatever it might have been at that point in time and say, well, actually, we don't we never made rockets before, so someone else can make better rockets than us, so we'll give it to someone else.
• 24:52
  SPEAKER_00: You know, you commit to it and you have this belief that you are gonna be able to transform this industry and to make a really unique and impactful product and bring it to the market.
• 25:02
  SPEAKER_00: I think that's one of the most important bits.
• 25:04
  SPEAKER_00: And I think it's also a part that it gets forgotten a lot in robotics.
• 25:09
  SPEAKER_00: People rely on some external magical entity that would just do all the boring and the hard work for you.
• 25:17
  SPEAKER_00: And you know, if you want to do robotics, you cannot write a couple lines of code, and or in the same time, you can't just focus on making a few bits and bobs, a few hardware parts, and then hope that everything else will be sorted.
• 25:30
  SPEAKER_00: It is a full-stack discipline by design.
• 25:33
  SPEAKER_01: Yeah, but but also uh there are certain things that you can get away with that you can't get away with really at huge scales, right?
• 25:41
  SPEAKER_01: I mean, like SpaceX, you know, it's not that many rockets that they're building, you know, and and and one of the things that you pointed out to me the last time we spoke is that um, you know, your your clients, your partners that you're working with, it it's one, it's effectively one robot per warehouse, right?
• 25:59
  SPEAKER_01: You know, you're not it's not locust robotics where each each warehouse has like a hundred of these.
• 26:04
  SPEAKER_01: So like you can really sort of like focus and narrow down on building each one of these and customizing each one of these in a robot robots in a way that you couldn't if you were really building like a full fleet of them.
• 26:17
  SPEAKER_00: Well, to be honest, um Locus, I think they build a lot in-house as well.
• 26:20
  SPEAKER_00: Yeah.
• 26:21
  SPEAKER_00: Um and yeah, you're right.
• 26:22
  SPEAKER_01: Yeah, and they and they and I know they like repair, but but but yeah, I mean, like when when the it it it is a matter of scale to a certain extent.
• 26:29
  SPEAKER_00: Aaron Powell It's also I think it's a really important matter of business model.
• 26:33
  SPEAKER_00: Because I think the the SpaceX example is really good.
• 26:36
  SPEAKER_00: If they had to mass produce these rockets, and then would they would never have this concept of reusing them or some of them, or at least kind of like driving and having this mission to get there.
• 26:48
  SPEAKER_00: Trevor Burrus, Jr.
• 26:48
  SPEAKER_00: Which is interesting because Locus does that too.
• 26:51
  SPEAKER_00: Like they refurbish their own their own robots.
• 26:53
  SPEAKER_00: It's also, I think, because the business model for a company like Locus is the fact that you know you have peaks and then you might want more robots and then you send them back because you don't need so many Seasonality.
• 27:04
  SPEAKER_00: Yeah, peaks and seasonality, and it gives you that flexibility.
• 27:07
  SPEAKER_00: Um but I think you know that's it it plays into the full stack element for robotics.
• 27:12
  SPEAKER_00: You have to understand what the market wants, right?
• 27:14
  SPEAKER_00: The market didn't want to buy rockets from SpaceX, they wanted to put their payload into orbit.
• 27:18
  SPEAKER_00: I think they put more payload into orbit this year alone than well in 2025, sorry, than the rest of the world combined.
• 27:25
  SPEAKER_00: And yes, they yet they do not mass produce the rockets.
• 27:28
  SPEAKER_00: So I think you know, if you it goes back to our first part of our conversation, right?
• 27:33
  SPEAKER_00: It's listening to the market and understanding what they want and how does the product that you're bringing serve them.
• 27:39
  SPEAKER_00: And that's really, really important because if you are just focusing on maybe we could sell these robots as well.
• 27:44
  SPEAKER_00: And they say, you know what, we don't want to deal with the complexity of deploying and integrating and servicing them over time.
• 27:50
  SPEAKER_00: You know, that's that's that's a fact.
• 27:53
  SPEAKER_00: But that's not really answering the market's problem.
• 27:56
  SPEAKER_00: The market doesn't really want to buy more robots, they want to solve their problem, which is we want real-time data of what's going on in our warehouses.
• 28:03
  SPEAKER_01: Yeah, something I didn't realize that the term digital twin is used in racing as well.
• 28:09
  SPEAKER_01: I think it's used in many industries.
• 28:11
  SPEAKER_01: Yeah, sure.
• 28:11
  SPEAKER_01: But yeah, I did, but how how is it applied in racing?
• 28:14
  SPEAKER_00: So, for example, you have um you have multiple parts of the team running things in parallel.
• 28:20
  SPEAKER_00: So if you have a if you have a race weekend, even before the race weekend, you have um a lab car, which could have various levels of complexities.
• 28:28
  SPEAKER_00: You could have just a bench system where you have all the systems on an actual car and you can replay data from any track.
• 28:36
  SPEAKER_00: You can have that bench system as a hardware in the loop, software in the loop simulator.
• 28:40
  SPEAKER_00: So you hook it up with a simulator that feeds it even more information that you can then put a driver in, which gives you the ability to actually preset or pre-define some of the setups for that track.
• 28:51
  SPEAKER_00: Then as you go to the track, you can run the simulator in parallel with the initial testing.
• 28:57
  SPEAKER_00: You get the data from the test, and you can fit it into the lab car, you can fit it into the simulator.
• 29:02
  SPEAKER_00: You have this digital equivalent of the real car.
• 29:05
  SPEAKER_00: And a lot of the work that, for example, I used to do, I'll look at how these sensors come into play and really understand what's the level of performance that we're gonna be getting, uh the level of performance that we're gonna be getting on track from the driver, because the driver is still a person, you know, they get tired, they have better days or less good days, the track conditions change all the time.
• 29:28
  SPEAKER_00: So being able to fit this into a digital equivalent gives you the ability to run, I wouldn't say infinite simulations, because there's obviously a cost cap in F1, but much, much cheaper simulations and um scenario analysis.
• 29:40
  SPEAKER_00: And that was one of the ideas that inspired what we're doing here today for logistics is that you can run all this what-if scenarios and alternative ways of running your warehouse without having to touch basically a single atom of the operation.
• 29:55
  SPEAKER_01: That's something I hadn't considered because I I I had thought that that maybe the kind of um the big value prop of of building a digital twin is let you like effectively just have a map of where everything is.
• 30:07
  SPEAKER_00: Um I think that's the probably that's the first step.
• 30:11
  SPEAKER_00: Yeah.
• 30:11
  SPEAKER_00: And as you get more data, the other um huge value that you you get out of it is you can look back in time.
• 30:18
  SPEAKER_00: So from pretty much from day one when you first deployed this technology through through current times, but also what we're building now is this ability to run multiple scenarios and leverage AI with in conjunction with the data sets we have, so we can run much more sophisticated analysis for the future, um, to the point where the where the warehouse digital entity can actually provide different suggestions and uh different analysis on on how to best optimize looking ahead.
• 30:46
  SPEAKER_01: Yeah, and and I'm guessing that uh the the companies that you're working with now, I mean, that's useful data to have when uh building new warehouses.
• 30:56
  SPEAKER_00: Um it's it's useful for building new warehouses, but the biggest impact we're seeing is on optimizing existing operations.
• 31:03
  SPEAKER_00: Um because I think, and I'm a big believer in solving current problems.
• 31:08
  SPEAKER_00: It's really difficult to build a business that relies on the market catching up or some something happening sometime in the future.
• 31:15
  SPEAKER_00: A lot of the customers that that use our technology, they're very interested in how to optimize their existing operations.
• 31:22
  SPEAKER_00: So you have these big warehouses, you have these big spaces, and you could you can put a lot of product through them, or you could change the layout, so you can pick a lot faster.
• 31:33
  SPEAKER_00: Before we actually brought this technology to market, there's a lot of guesswork.
• 31:37
  SPEAKER_00: Now the big advantage is that you can look back on data and you can look pretty much on every single scan, look at what the state of the warehouse was, and then make these decisions and make these informed changes of the environment.
• 31:50
  SPEAKER_00: And you might even want to make them, you know, twice a year, three times a year.
• 31:54
  SPEAKER_00: They're big, they're big, they're big changes.
• 31:56
  SPEAKER_00: If you're starting to move racks around and add more levels in your warehouse because you have the confidence that you know we're underutilizing the space, it gives you the opportunity to actually put more value, you can make more money from the from the same from pretty much the same operation.
• 32:11
  SPEAKER_01: I I was coming over here, my my Uber and the driver was telling me that he was like, oh, the GPS like doesn't know where anything is because this is all completely new.
• 32:22
  SPEAKER_01: Like this was just like fields and and trees previously, so the GPS hasn't caught up.
• 32:28
  SPEAKER_01: So like like in one sense, like this it this is kind of greenfield in that like this is like just a brand new warehouse, but really it was it's a brand new warehouse that was kind of built to be kind of an old warehouse, right?
• 32:41
  SPEAKER_01: I mean, it was built to be like sort of a classic warehouse, ex it was built to be an existing warehouse.
• 32:47
  SPEAKER_00: Yeah, I think you know, people like like I was saying, people build them for the problems that they have today.
• 32:52
  SPEAKER_00: And the big thing about the industry is also this this construction, they're built on a 30, 50-year-old year lifespan.
• 32:59
  SPEAKER_00: You're not building one to change it next year or in three years' time, right?
• 33:04
  SPEAKER_00: You kind of put a big investment up front and a lot of commitment to that type of warehouse or that type of facility.
• 33:10
  SPEAKER_00: And people need a level of flexibility.
• 33:12
  SPEAKER_00: I think there's a lot of examples in the market where you know people thought that they could automate a large portion of the warehouse, but then the market changes.
• 33:21
  SPEAKER_00: You know, the things that we buy as consumers, the requirements for the industry, you know, stuff changes and they don't actually get um they don't get the flexibility they want from the automation.
• 33:32
  SPEAKER_00: And that I think is really important.
• 33:34
  SPEAKER_00: That's one of the reasons why we're focusing on the data intelligence layer, because having really good grasp of the data, of the insights from that warehouse gives you the opportunity to optimize what people spend their time on, what they focus on, the kind of automation you want to bring, the kind of processes you want to put in place.
• 33:52
  SPEAKER_00: Basically, you can run different operations for for different seasons if if that's what's best suited for that environment.
• 33:59
  SPEAKER_00: Changing, big physical changes in warehouses are expensive and they're really hard to implement unless you have this clarity and this um confidence that what you're doing is actually the right thing for what the operation needs.
• 34:13
  SPEAKER_00: So, how how did we end up here in Nashville?
• 34:16
  SPEAKER_00: That's uh that's that's actually quite an interesting story because we so we're based uh outside London in the UK, and we run quite a tightly centralized operation, so we serve all our customers from there.
• 34:29
  SPEAKER_00: Um we s we started to get a lot of traction in the US market, and we didn't have anything, we didn't have an office.
• 34:35
  SPEAKER_00: Uh, we had a lot of demand from the market.
• 34:38
  SPEAKER_00: What do you attribute that to?
• 34:39
  SPEAKER_00: Um, I think the biggest part of that is the fact that the product market fit was really strong.
• 34:46
  SPEAKER_00: It's as simple as that.
• 34:47
  SPEAKER_00: Like and then there just wasn't uh there wasn't an equivalent or there wasn't an option.
• 34:51
  SPEAKER_00: There isn't an equival equivalent.
• 34:53
  SPEAKER_00: That was that was the other big thing.
• 34:55
  SPEAKER_00: The way we solve the problem and and the way we sort of like listen to the market and listen to the customers and tick those boxes is a very unique approach and a very unique product in the market.
• 35:05
  SPEAKER_00: And I think you know that the big difference to doing business in the US versus Europe, for example, is that people really their business, the business environment here is really competitive.
• 35:16
  SPEAKER_00: You know, people want to win and they want to make money and they want to get that advantage to other competitors.
• 35:22
  SPEAKER_00: And that's why we were so successful.
• 35:24
  SPEAKER_00: Like we've opened up the US market, which is actually a big achievement for a lot of European companies.
• 35:30
  SPEAKER_01: They they they they I constantly hear how difficult it is to break into the US.
• 35:35
  SPEAKER_01: It is difficult because it's a ruthless market.
• 35:38
  SPEAKER_00: You know, if you have something really unique, very, very high quality that solves the problem, you're gonna get a lot of adoption, a ton of adoption.
• 35:47
  SPEAKER_00: It's also big and it's diverse and it is big and diverse, but it is moving fast.
• 35:52
  SPEAKER_00: And when you're a startup, when you're a young company that wants to actually bring something into the world, that's those are the customers you want to be working with.
• 36:00
  SPEAKER_00: The customers are committed to winning in their space.
• 36:03
  SPEAKER_00: And I think you know that was a that was a big deal.
• 36:05
  SPEAKER_00: So we started to accumulate this uh locations where we're deploying in the US, and a lot of it was actually centered around Nashville, around Tennessee.
• 36:14
  SPEAKER_00: So we started scoping for um, you know, what would make sense, where do we want to set up a it's not just an office, where do we want to set up a testing space, where do we want to set up a development center?
• 36:24
  SPEAKER_00: Um, it's kind of like an excellent center where you can do demos, where you can bring people and actually discuss what the future of the industry looks like.
• 36:32
  SPEAKER_00: Um so that's how it uh it kind of like narrowed into uh for for Nashville.
• 36:38
  SPEAKER_00: And it's also it's a great city.
• 36:40
  SPEAKER_00: Um, great music, great food.
• 36:42
  SPEAKER_00: I mean, it takes a lot of boxes.
• 36:44
  SPEAKER_01: Yeah, is in terms of uh uh I'm really curious because I know obviously like you know, we've been talking a lot about how you know Austin has become this big uh hotbed for robotics.
• 36:55
  SPEAKER_01: Um you know, the the scene in Atlanta has become really interesting, obviously around the University of Georgia.
• 37:02
  SPEAKER_01: Um have you been recruiting?
• 37:05
  SPEAKER_01: Have you been finding a lot of uh folks locally?
• 37:07
  SPEAKER_01: So we're I mean we're just opening up this facility.
• 37:10
  SPEAKER_00: Yeah, and the Yes.
• 37:14
  SPEAKER_00: Actually, this is really exciting for me.
• 37:16
  SPEAKER_00: Um But we we started to engage with with a number of uh of different entities, and we're seeing good talent here.
• 37:21
  SPEAKER_00: We're also seeing something very interesting, which is um a lot of the the big tech companies are deciding to open up bigger offices here and to actually move uh big parts of their operation in the area, which which is obviously helping.
• 37:34
  SPEAKER_00: Um so it's it's looking really good from that perspective for us as well.
• 37:37
  SPEAKER_01: Yeah, one of the things I was writing about a a lot in um you know, especially at the beginning of the pandemic, was how much Amazon's big robotics fleet was like pushing everyone else into robotics because everyone else had to say competitive.
• 37:50
  SPEAKER_01: I mean, are are you noticing that that's sort of a big that that continues to be a big thing that's pushing people into automation, pushing competitors into that space?
• 38:01
  SPEAKER_00: So I mean that's uh it's a difficult answer, right?
• 38:04
  SPEAKER_00: Because I think they're they're making some big cuts in their in their robotics departments, Amazon.
• 38:08
  SPEAKER_01: Yeah, but but in terms of yeah, yes, I mean they're making big cuts across the board, but um, you know, the thing, you know, they're not they're not making the cuts to the robots themselves.
• 38:18
  SPEAKER_01: Yep.
• 38:19
  SPEAKER_01: And and the things that the the robots themselves afforded is you know, same-day, next day delivery, and that's what they have to compete with.
• 38:25
  SPEAKER_00: That's exactly, yeah, that's exactly what's actually driving people to look at technology and not just automation, but different technologies, and I think AI plays plays a big role here because it's not so much Amazon as an entity, it's the fact that it changed the consumer behavior and mentality.
• 38:43
  SPEAKER_00: And people are now wanting the same day delivery, next day delivery to be able to actually be competitive in this market.
• 38:50
  SPEAKER_00: You know, that's that's what makes it really hard.
• 38:52
  SPEAKER_00: That's where you have to, and you have to have a holistic strategy.
• 38:56
  SPEAKER_00: You can't just say, I'm just gonna deploy some autonomous XYZ type of machinery, and it's magically gonna cure everything.
• 39:03
  SPEAKER_00: You have to look at how you're changing your operation, how are you adapting your systems, and how are you making it cohesively work?
• 39:09
  SPEAKER_00: Um and this is what we're seeing from from our customers as well.
• 39:12
  SPEAKER_00: It's not so much, oh, we want to deploy the Dexuri product in one warehouse, it's how does this work and how does this help us shape the roadmap to other technologies that we're looking to adopt.
• 39:27
  SPEAKER_00: So it's good to have a digital twin because a big part of what we do with the information we're gathering is a digital infrastructure for future deployments if you have autonomous forklifts or even humanoid robots or other type of AI technology that relies on this physical understanding of space.
• 39:46
  SPEAKER_00: You know, you already have that.
• 39:48
  SPEAKER_00: So it that's a big advantage.
• 39:49
  SPEAKER_00: And you know, it's it's really about how you create the system to enable you to get to whatever goal or business goal you have.
• 39:56
  SPEAKER_00: Not it's really not about, and this is this is really exciting in logistics, it's not about running a pilot or it's not about running a nice um technology demonstration in a remote part of your company.
• 40:09
  SPEAKER_00: It's really kind of getting behind and deploying this at scale.
• 40:12
  SPEAKER_01: So you're saying that that building this very uh this precise digital twin, this map is gonna make it much easier to uh to implement and roll out other form factors in the space.
• 40:25
  SPEAKER_00: Yeah, absolutely.
• 40:26
  SPEAKER_00: Big uh you know, big believer in that because you have it's virtually a warehouse archiving system.
• 40:34
  SPEAKER_00: You have a digital copy of what your location looked like, what every single palette location, block stack location, pick phase location looked like in that warehouse at any point in time.
• 40:45
  SPEAKER_00: So you have an incredible training data set, you have an incredible um lake of information to leverage for AI, to leverage for machine learning and other automated technologies to help shape how you can write, how you can drive that operation to 100% efficiency.
• 41:02
  SPEAKER_00: That's really the name of the game.
• 41:03
  SPEAKER_00: I mean, that's why we deploy these technologies.
• 41:05
  SPEAKER_00: We want to operate at the maximum amount of efficiency we can so you can maximize everyone's time and make sure that everyone's kind of you know bringing the right level of input to make the business successful.
• 41:17
  SPEAKER_01: Yeah, when uh I was first describing the the robot, I think I called it kind of like a brute force approach.
• 41:25
  SPEAKER_01: Like, you know, everyone was trying to figure out how we can inventory these really tall uh shelving systems.
• 41:33
  SPEAKER_01: Um, you know, there was that that company out of Israel that did the squid robots at like, you know, which we went up the shelving units, and obviously, you know, you know, you you mentioned a lot of the um uh like the actual like shelving systems themselves, and there's a couple of companies doing uh you know drones, which is uh at very least interesting, and you just said, hey, let's build a 60-foot telescoping robot.
• 41:58
  SPEAKER_01: Well, we didn't start with 60 foot, we started a bit smaller.
• 42:01
  SPEAKER_01: Yeah, yeah, you yeah, you started, yeah, you started very modestly with a 40-foot.
• 42:05
  SPEAKER_01: Yeah, exactly.
• 42:06
  SPEAKER_00: To be honest, it was actually, and I don't want to necessarily claim uh you know, claim the glory for that for that one because we we asked the market.
• 42:14
  SPEAKER_00: So we went and we spoke to customers and we said, but what do you really want?
• 42:17
  SPEAKER_00: What do you really need?
• 42:18
  SPEAKER_00: Uh, because people don't need flying drones in their warehouse, they don't need big telescopic robots, they don't need squid robots that climb the rack.
• 42:25
  SPEAKER_00: What they wanted is a solution that enables them to deploy without changing the environment.
• 42:30
  SPEAKER_00: So the the speed, the time to value was incredibly important because they need to solve the problem quickly.
• 42:36
  SPEAKER_00: They want to scan as much as possible in the shortest amount of time, which is where having a telescopic approach gives you the ability to scan the entire rack, 10,000, 15,000 pallets an hour.
• 42:47
  SPEAKER_00: That's really what the market wanted.
• 42:49
  SPEAKER_00: And what they also wanted was to have access to the insights.
• 42:53
  SPEAKER_00: So not the raw data.
• 42:54
  SPEAKER_00: This this thing can do well over a thousand gigs an hour, but the actual insights are a lot smaller, and we process everything on the robot.
• 43:02
  SPEAKER_00: Because again, as we were speaking to the market, people are saying, you know, our Wi-Fi is actually not that great.
• 43:07
  SPEAKER_00: We might have dark spots.
• 43:08
  SPEAKER_00: So you need to have you need to have the ability to run fully autonomously and bring this information to them.
• 43:14
  SPEAKER_00: And then the flip side of that was how do you consume the data?
• 43:18
  SPEAKER_00: That was the other big question that we asked.
• 43:20
  SPEAKER_00: Okay, cool, we will give you guys everything you want: pictures, 3D scans, you know, millimeter accuracy, you're gonna have the whole view of what's going on in every single square foot of that warehouse.
• 43:34
  SPEAKER_00: How do you consume the data?
• 43:36
  SPEAKER_00: And that was the other big part of it, is like, okay, the digital twin idea is really when when we were very bullish and said, oh, look, there's nothing else in the market that enables you to consume the insights.
• 43:46
  SPEAKER_00: You can't just rely on having an Excel spreadsheet or a CSV file.
• 43:50
  SPEAKER_00: You have to have this physical understanding of what's going on in the space.
• 43:54
  SPEAKER_00: And that's really why it ended up being and looking as it is today.
• 44:00
  SPEAKER_01: Do you think that there's a space for drones in warehouses?
• 44:04
  SPEAKER_00: Um I think Corvus are doing something very interesting in the freezer space.
• 44:10
  SPEAKER_00: From what we're seeing in the market, um, customers are not happy with the speed of scanning and um with the fact that uh they have to close down the aisles and the risk they pose of also like uh having flying objects in the warehouse.
• 44:23
  SPEAKER_00: And I don't think this is a drone specific, like a drone company technology, it's just aerodynamics.
• 44:29
  SPEAKER_00: You know, you can go into a narrow aisle and then you have missing pallets, and then it completely messes the um the control systems on a drone, or more so than ever, you have hygiene issues, you have hanging cling film, and those are gonna be sort of like sucked into the propellers, and they're gonna impede the speed at which the system can actually do its job.
• 44:50
  SPEAKER_01: One of the things that we were talking about before the camera started rolling is I asked you, I was like, I, you know, I've I've written about your company several times.
• 45:00
  SPEAKER_01: I don't know what the robot's called.
• 45:02
  SPEAKER_01: And you said that's because the robot isn't really called anything, and then that's by design.
• 45:08
  SPEAKER_01: And and uh the sense that I've always gotten from you is like, you know, you you I don't want to say you're annoyed, but like you don't you don't want the focus to be on the robot.
• 45:18
  SPEAKER_00: You don't want all the press coverage to be on the robot.
• 45:21
  SPEAKER_00: That's true, that's factual.
• 45:22
  SPEAKER_00: And and so like I I can make the parallel to SpaceX again.
• 45:25
  SPEAKER_00: It's like you I don't think they have a name for every single one of those rockets.
• 45:28
  SPEAKER_00: They might have code names.
• 45:30
  SPEAKER_00: But for us, the big thing is is the full we're a full stack business.
• 45:33
  SPEAKER_00: I don't see us as being a robotics company because we don't just make the robot.
• 45:38
  SPEAKER_00: We make all the perception systems that transform the data the robot gets into insights.
• 45:43
  SPEAKER_00: We make all the control systems, the autonomy stack.
• 45:46
  SPEAKER_00: We're not an autonomy technology provider company.
• 45:49
  SPEAKER_00: I really see us as being a full stack company, a full stack product.
• 45:54
  SPEAKER_00: Then we this is how we bring value to the market.
• 45:58
  SPEAKER_00: Well, we're at time.
• 45:59
  SPEAKER_00: Thank you so much for joining us.
• 46:00
  SPEAKER_00: It's been a pleasure.
• 46:02
  SPEAKER_01: Thanks so much for tuning in.
• 46:04
  SPEAKER_01: Thanks to Andre and Dexary, thanks to you as always.
• 46:08
  SPEAKER_01: Uh, if you're enjoying the show, please like and subscribe wherever you get your podcasts.
• 46:13
  SPEAKER_01: And don't forget to subscribe to our weekly newsletter at automated.fm.
• 46:18
  SPEAKER_01: Thank you so much, and we will see you at the same time next week for another episode of Automated.