Latest / EU-Startups Podcast / Meet the World's Tallest Autonomous Robots

Transcript

Meet the World's Tallest Autonomous Robots · EU-Startups Podcast

• 0:00
  Speaker 1: You are now listening
• 0:01
  Speaker 0: to the EU startups podcast.
• 0:13
  Speaker 1: Hello, and welcome to this episode of these service podcast.
• 0:16
  Speaker 1: as always I'm your host David sending Garcia the editor at you start ups today.
• 0:21
  Speaker 1: I chat with one Yinka.
• 0:22
  Speaker 1: She's the co-founder of Dexery.
• 0:25
  Speaker 1: some of you might know it has bots on us As it was previously called.
• 0:29
  Speaker 1: Now dexter is no stranger to a set use startups we've been covering Just about before they changed names, which was in twenty-twenty two.
• 0:38
  Speaker 1: They've raised almost two hundred and fifty million euros in funding have expanded not just across Europe but uh...across the US and really globally.
• 0:49
  Speaker 1: Uh..they are a robotics and physical AI company specifically warehouse robotics .They develop what they call The Tallest Autonomous Robot.
• 1:01
  Speaker 1: So we were able to chat with Iwana about her own history, how she fell into dexery.
• 1:08
  Speaker 1: The impact of physical AI especially when comparing it companies and LLNs, the value of physical AI when it comes to this conversation especially how she sees overall sector within Europe compared with US.
• 1:37
  Speaker 1: How does she see funding going around for her own sector?
• 1:41
  Speaker 1: And even application gender in these sectors?
• 1:47
  Speaker 1: if she feels there is more space To be made for more women to join the industry what work is left to be done.
• 1:54
  Speaker 1: So I hope you're looking forward to that conversation.
• 1:58
  Speaker 1: Hello, Anna and welcome to The EU startups podcast.
• 2:01
  Speaker 1: It's a pleasure to have you on especially because you made the time onto two short notice to come join us.
• 2:07
  Speaker 0: Thank You so much for having me now very excited to have a chat.
• 2:11
  Speaker 1: Yes yes i am as well.
• 2:13
  Speaker 1: during my tenure of the podcast which has been just about a year so far now, we actually haven't had that many kind of physical AI even just robotics.
• 2:24
  Speaker 1: people come join us.
• 2:25
  Speaker 1: So I am actually interested in this conversation.
• 2:26
  Speaker 1: some of the reasons why i chose you apart from your obvious background to come joining me here...so on that note why don't you go ahead and explain uh just who you are what do?
• 2:36
  Speaker 0: Sure thing, so my name is Awana Jenga and I'm one of the co-founders at Dexery.
• 2:41
  Speaker 0: Day to day i kind look after the commercial and product side of the business sales marketing customer success protocol map strategy but as a founder you do everything in nothing.
• 2:51
  Speaker 0: it also very varied hour.
• 2:57
  Speaker 0: And maybe to touch a little bit on what dexory does in dexories.
• 3:00
  Speaker 0: So we are indeed, in the physical AI space but we're robotics and AI solution for warehousing logistics which is quite an industrial space-to be in But it's probably one of our biggest industries that we have as humankind.
• 3:14
  Speaker 0: Anything around us is at end of day something that spends time in warehouse at least the warehouse.
• 3:20
  Speaker 0: Usually, it's multiple of those sometimes even ten plus warehouses around world.
• 3:23
  Speaker 0: so It is definitely industry that connects us all together and we have a solution That brings visibility and data intelligence to warehouses To be able run a lot more efficiently.
• 3:35
  Speaker 0: And as part of that We use some of the worlds most The tallest autonomous robots kind of a mouthful there.
• 3:41
  Speaker 0: They go up to eighteen meters in air and scan warehouses digitize space rules, kind of the people and machinery inside all different kinds resources.
• 3:53
  Speaker 0: digitize that in real time.
• 3:54
  Speaker 0: And then there's a lot interesting insights we extract from data.
• 3:58
  Speaker 0: so yeah We could definitely go into more detail but just to get things going.
• 4:04
  Speaker 1: That is good starting point.
• 4:06
  Speaker 1: You are no stranger at EU startups.
• 4:08
  Speaker 1: I actually had quick look over past coverage that we've done.
• 4:11
  Speaker 1: And, uh... We go even as far back before you were Dexer when you were bots in us.
• 4:17
  Speaker 1: You are on our radar then and I think so-far at a look of Crunchbase you have raised two hundred forty nine million total funding.
• 4:24
  Speaker 1: So your no slouch That's for sure!
• 4:27
  Speaker 1: I do think there is lot good things to talk about here.
• 4:29
  Speaker 1: that will be relevant For the audience.
• 4:31
  Speaker 1: But Before we get into it Uh..I Do kinda want touch On Your Background There A little bit Because It's Always The Most.
• 4:38
  Speaker 1: One of the most interesting aspects in these chats for me is to hear about how people landed in a role that has brought them success and ability in sector they're in.
• 4:47
  Speaker 1: Now, from what I read about you...I saw that you've held positions before in Telefonica or Google.
• 4:56
  Speaker 1: so then How does this translate into robotics field?
• 4:59
  Speaker 1: You said on sales & product side but still have some kind understanding as well.
• 5:07
  Speaker 1: be successful and imagine.
• 5:10
  Speaker 0: Yeah, it's a very interesting kind of story because I mean growing up... don't think I ever had a plan to actually get into robotics or anything like that.
• 5:18
  Speaker 0: But, I remember early days in my mind for some reason although i have no examples of it with the family and knew they wanted me to do something on their own business.
• 5:27
  Speaker 0: as a child you know playing around with other kids having your little store and everything.
• 5:33
  Speaker 0: It was just something kind of in the back of our minds.
• 5:35
  Speaker 0: but uh...I came from very traditional families working at medsons and doctors.
• 5:41
  Speaker 0: so there's nothing guiding towards this.
• 5:44
  Speaker 0: I always kind of had it back with my mind.
• 5:45
  Speaker 0: So went to university, did marketing PR management business sales so very, very business driven and then from there pretty much started directly into the tech sector.
• 5:58
  Speaker 0: As you said i don't have a typical tech background but a lot of experience in bringing products to market, tech products to markets whether that's software or hardware.
• 6:09
  Speaker 0: Indeed I started Telefonica initially on what was called like Telefonic Digital so all the different products outside of basic telephony and texting messages and stuff like that Telco has.
• 6:20
  Speaker 0: it was apps for loyalty All kinds new ways communicating internationally.
• 6:27
  Speaker 0: this is way back when we didn't have WhatsApp and everything else on our phones.
• 6:32
  Speaker 0: And then I moved over to Google, where I was on the team that launched the first Google Pixel phones and the Google Home hardware.
• 6:39
  Speaker 0: Obviously like now it's very well spread.
• 6:42
  Speaker 0: so there has always been this motion in my career around... finding a cool product idea, working on it with technical teams and then bringing you to market through marketing sales partnerships.
• 6:54
  Speaker 0: And of course after that making sure there's enough sales aspect for this.
• 6:59
  Speaker 0: This is what I do at Dexery.
• 7:01
  Speaker 0: To your point earlier when we started, because as the three co-founders who have complementary skills they always thought business was going to mind in game like Who will buy?
• 7:14
  Speaker 0: How do we take it to people?
• 7:17
  Speaker 1: Okay.
• 7:18
  Speaker 1: And so what was that initial thinking with the company?
• 7:21
  Speaker 1: because you've been around since twenty fifteen and now obviously AI is a huge idea, it's a huge implementation in most companies.
• 7:34
  Speaker 1: It wasn't really something people were talking about in twenty fifteen.
• 7:39
  Speaker 1: So how did the business kind of gear towards that?
• 7:43
  Speaker 1: Was it a product question, was that something where you're seeing most your customers interested on this side?
• 7:50
  Speaker 1: or why do they start adopting from robotics?
• 7:54
  Speaker 0: Yeah so its been something we've pretty much focused on day one which is using autonomous robots to collect data Like, yeah people probably see behind me.
• 8:04
  Speaker 0: I've got a bot.
• 8:05
  Speaker 0: send us sign just up there which is from the very early days our first trade show that we did bots and does like back in like twenty sixteen-twenty seventy ages ago And at the time We were working on a product for the home.
• 8:17
  Speaker 0: So we're gonna use an autonomous robot to navigate your homes.
• 8:20
  Speaker 0: You can keep an eye on the house from far away.
• 8:22
  Speaker 0: you could kind of make sure There's no obviously course kind of security issues but also communicate with the people inside like have this kind view of what's happening at home from wherever you are in the world.
• 8:35
  Speaker 0: We quickly realized that, making a product for our own back then was going to be extremely expensive like Lydarsens and everything else were absolutely ridiculous.
• 8:43
  Speaker 0: so it wasn't something people could actually afford very easily to put into their homes.
• 8:48
  Speaker 0: So we very quickly pivoted from there into B-to-B space From your first conversations with having We're talking to us about a very similar challenge, which was like they got it.
• 9:00
  Speaker 0: They had an understanding of what's happening for instance in their stores or in the terminals and airport because that is also kind follow it, like day after date to really understand what the reality on the ground is.
• 9:15
  Speaker 0: So we pivoted from home into commercial spaces and as we were doing that retail became a big focus for us with the pandemic because our market disappeared almost overnight.
• 9:27
  Speaker 0: but we're quite lucky because those retailers didn't ask us can you actually do this inside of warehouses?
• 9:32
  Speaker 0: Because they have massive visibility problems.
• 9:34
  Speaker 0: there's no way What's going on inside the warehouse?
• 9:38
  Speaker 0: You know what goods are coming in or going out, but actually what is happening under the roof it very hard to monitor.
• 9:43
  Speaker 0: And that still means and was meaning back then they were making a lot of decisions blindly or just some kind of gut feel.
• 9:51
  Speaker 0: So I'm going back where we started.
• 9:54
  Speaker 0: It has always been this intention using robots reality, and now it has a word.
• 10:03
  Speaker 0: It's called physically eye world models like to your point.
• 10:07
  Speaker 0: yeah when we started that was as many years ago Now um it wasn't even an idea right.
• 10:13
  Speaker 0: so Um uh it's great to kind of see the world catching up with it And it is.
• 10:18
  Speaker 0: Of course A lot of stuff Has evolved in The meantime.
• 10:21
  Speaker 0: I'm also talking about the sensors are very expensive.
• 10:24
  Speaker 0: easy and cheap to build robots, of course you still have to make sure that they operate the after day.
• 10:28
  Speaker 0: And there's a lot complexity in it.
• 10:30
  Speaker 0: but the market has evolved so much over past decade than we've been in it.
• 10:35
  Speaker 1: How do your... You call them the tallest autonomous robots right?
• 10:39
  Speaker 1: Yes!
• 10:39
  Speaker 1: Yeah.
• 10:40
  Speaker 1: how do they kind of interact with human warehouse workers?
• 10:45
  Speaker 1: because you mentioned COVID Right, so that must have been a period where it was kind of hard to get people in the warehouse and workers back on track doing their jobs they were doing.
• 10:54
  Speaker 1: So what's the interaction there?
• 10:57
  Speaker 1: Like can they do with human?
• 10:58
  Speaker 1: does point blank or is there balance between an operator and robot?
• 11:04
  Speaker 0: Yeah because we use these robots let's say rags, like I said all the way up to eighteen meters in the air.
• 11:13
  Speaker 0: There is an element of human teams that are trying to capture some of this data by doing inventory checks or walking around and trying spot things.
• 11:22
  Speaker 0: but they can only do a certain number what we call pattern locations per hour.
• 11:29
  Speaker 0: so you cannot really be replicated with a human workforce unless A ridiculous amount of people that all they do is walk around the comb boxes, which is pretty tedious and boring.
• 11:42
  Speaker 0: And also quite dangerous because you kind have to go up and down all the way after eighty meters all the time.
• 11:47
  Speaker 0: So we don't necessarily come in To replace something.
• 11:50
  Speaker 0: We come into really augment The business with a lot of real-time data information.
• 11:54
  Speaker 0: hence why The digital twin that you mentioned is like where all of the data gets pushed almost instantly as it's being scanned and processed, then kind of open up to a lot inside.
• 12:03
  Speaker 0: So from our perspective we're going to come in an automate task with very small amount but there are also ridiculous amounts of value for bringing back again.
• 12:13
  Speaker 0: You can't capture what people take high resolution images around.
• 12:16
  Speaker 0: Of course We cannot do this yet With eyes because I don't know when were gonna get too But There obviously was massive amount of data robots.
• 12:25
  Speaker 0: And going back to your question around like, are they interacting with people?
• 12:28
  Speaker 0: They're actually not meant to interact with people that there just collect data in the background.
• 12:33
  Speaker 0: if anything less than getting anyone's way or kind of disturb any one then more successful because their just meant be pretty much capturing this information and get out off pickers and people driving forklifts making sure none productivity is being impacted by the robot collecting do not interact at all with the workforce.
• 12:56
  Speaker 0: But of course, most human colleagues or robots are super proud to have them there and they give their way... There's a whole kind of human-robot interaction field that happens automatically although the intention is just make sure we get out of the way and not disturb your
• 13:12
  Speaker 0: operations.".
• 13:17
  Speaker 1: How do these robots work practically when customer comes to you?
• 13:22
  Speaker 1: Is it that they're retrofitted into the warehouse that has already existed with a customer, or do provide your own facility giant robots around it.
• 13:32
  Speaker 1: What's that
• 13:32
  Speaker 0: process?
• 13:33
  Speaker 0: Yeah, I know so its extremely simple to get up and running.
• 13:37
  Speaker 0: first of all we only put one single robot per warehouse.
• 13:41
  Speaker 0: We actually just published the other day on our case studies in One Of The Sites That Were In.
• 13:45
  Speaker 0: It was like a million square feet.
• 13:48
  Speaker 0: i think something you can't even picture how big some of these warehouses are.
• 13:53
  Speaker 0: So Even For Very Big Spaces We Only Need One Robot Because We're Scanning At A Crazy Speed And Capturing That In A Couple Of Hours And then we can kind of go back and like get it again.
• 14:03
  Speaker 0: So you're pretty much reducing to a couple hours, any data gap that currently could be even months sometimes years until actually figure out something is wrong.
• 14:14
  Speaker 0: so first yeah its just one unit which makes quite simple and easy understand.
• 14:18
  Speaker 0: also there's absolutely no change that needs to happen, the current warehouse that a customer has.
• 14:24
  Speaker 0: All we need is power plug.
• 14:25
  Speaker 0: just get robot in and after our deployment team spends about two or five days on site depending upon how complex it gets everything up running.
• 14:35
  Speaker 0: then they start getting data.
• 14:36
  Speaker 0: so there's been focus from beginning making sure solution can be implemented seamlessly super-super fast.
• 14:51
  Speaker 1: Would that be?
• 14:52
  Speaker 1: the most difficult part of the kind journey at The Company is Had has been figuring out this customer application because you're already a series C, I think if my notes are right.
• 15:03
  Speaker 1: And so obviously you figured your market and figure our product.
• 15:08
  Speaker 1: but throughout their journey there had to be a growing pain in getting it to work.
• 15:12
  Speaker 1: was it convincing people that they should have a giant robot in their facility.
• 15:20
  Speaker 1: What was the kind of process for you?
• 15:22
  Speaker 0: Yeah, it's everything right because we rebranded to Dexory and introduced this particular product just about over three years ago as like in twenty-twenty-three.
• 15:33
  Speaker 0: so um... It still relatively new if I think the logistics sector, which obviously it's a massive market but still a little bit more traditional.
• 15:41
  Speaker 0: And of course once you make some changes implications can be huge in terms like that doesn't go well.
• 15:47
  Speaker 0: customers get upset, transportation gets messed up.
• 15:49
  Speaker 0: there is whole kind of implication to it.
• 15:51
  Speaker 0: so I think yeah without singing our praises we managed do so much in three years because of that solution being as seamless as i said earlier.
• 16:00
  Speaker 0: So I think he was a combination coming out of Europe, even a small kind of company in the UK trying to actually create a brand within the sector.
• 16:09
  Speaker 0: It's it's very big endeavor because you were relying on these massive customers and see a lot of names from our website like your DHL and MERSC.
• 16:18
  Speaker 0: I did some of the biggest logistics companies in the world getting their trust to allow you in there sites doing such an important part at work.
• 16:28
  Speaker 0: obviously that takes time.
• 16:29
  Speaker 0: so number one is first of all this who we are.
• 16:32
  Speaker 0: yep, the solution works.
• 16:33
  Speaker 0: Trust us on this and start to kind of build that
• 16:35
  Speaker 0: in.".
• 16:36
  Speaker 0: And with a first case studies then that starts get a snowball effected.
• 16:39
  Speaker 0: people find out about you in the market and so one.
• 16:41
  Speaker 0: but then on the technical side there are number things we have to figure out.
• 16:45
  Speaker 0: So when it comes to robots themselves like make sure data is extremely correct because any AI product at end what they were offering was not for the robot itself.
• 16:57
  Speaker 0: They're just there to collect information.
• 17:00
  Speaker 0: What's important?
• 17:00
  Speaker 0: if that information is wrong your insight and what you're offering on the other side is going to be wrong.
• 17:05
  Speaker 0: So there's a whole element of making sure that robots obviously are super accurate, when talking about these massive spaces they still have like centimeter accuracy even after navigating one million square feet which anyone working robotics knows how difficult it is.
• 17:21
  Speaker 0: but then thirdly does all elements make sure earlier data we provide brings value?
• 17:28
  Speaker 0: simple use cases, then we observe how the customers are using that data to extract all kinds of additional bits of information.
• 17:35
  Speaker 0: So I'll give you an example like when we initially started were looking at stock accuracy.
• 17:39
  Speaker 0: so we're comparing what robot was scanning to what's customers had in different systems they were using in their warehouse pointing any discrepancies quickly telling them it can action these five locations right now and fix stuff.
• 17:52
  Speaker 0: But very, very quickly we moved on from that because people were using the pictures to find ways in which they... for example, kind of create more space by bringing different bits of stock together.
• 18:02
  Speaker 0: How they're trying to reduce walking distances but bring in products together and how there are trying to produce congestion?
• 18:08
  Speaker 0: There's so many kinds of ways of optimizing the site that this data was now allowing them do which wasn't like a possibility before because it just didn't have access information.
• 18:17
  Speaker 0: So then quickly came into product And very recently we launched our context review adapt Which is full play on agent take for seeing a problem might arise further down because we can see what goods are coming into the site, or what promotions are happening.
• 18:33
  Speaker 0: We're helping them stay on top of that even before it hits their supply chain.
• 18:37
  Speaker 0: so I think going back to your question is like one thing getting in at the market as start-up It's always very hard especially when you go industrial having reliable data source.
• 18:49
  Speaker 0: but thirdly adapting product and keep building more value to our customers.
• 18:55
  Speaker 1: Okay, well on this topic.
• 18:56
  Speaker 1: then they talk about kind of entering a market initially as the startup As we mentioned before that was back in twenty fifteen when were talking over just more than a decade now.
• 19:06
  Speaker 1: So you've obviously had some time into see competitors to see the ecosystem To see peers and how they evolve.
• 19:13
  Speaker 1: so with that in mind I had it look over our past coverage Of kind of the physical AI warehouse robotics space And that's just for this year.
• 19:22
  Speaker 1: That I had a quick look at.
• 19:24
  Speaker 1: From what I've seen, this is just a quick look.
• 19:26
  Speaker 1: There's two hundred and forty four million in funding that kind of just circled around the same area The Dexerys working on.
• 19:33
  Speaker 1: so with that in mind How do you see that space within Europe?
• 19:39
  Speaker 1: Is it growing into correct way?
• 19:41
  Speaker 1: Do feel from twenty fifteen to now know.
• 19:44
  Speaker 1: how does that compare?
• 19:45
  Speaker 1: when we're looking at international markets will get the US which you have expanded to, and that's another point of the conversation.
• 19:52
  Speaker 1: But what do you see the European sector looking like right now?
• 19:57
  Speaker 0: What I was going say is that million or billion because...
• 20:01
  Speaker 1: Well from what we've covered- And i just want put out there We don't cover everything as much as we'd liked
• 20:06
  Speaker 0: it
• 20:07
  Speaker 1: Just for twenty six from companies.
• 20:10
  Speaker 1: That's the million is what I saw, but yeah.
• 20:12
  Speaker 1: I'm sure it can be very much higher.
• 20:14
  Speaker 1: yes
• 20:14
  Speaker 0: Yeah because i was kind of like uh...I talk a lot about physical AI Because obviously its our bread and butter And has been for a long time.
• 20:20
  Speaker 0: Um It's-it's A Very interesting space because you Can break it down into a Lot Of different sub sectors?
• 20:27
  Speaker 0: Then your point Is exactly Like okay how Much funding is going in Across everything or specifically In The robotic side of it?
• 20:34
  Speaker 0: Because there is a whole element.
• 20:37
  Speaker 0: Everyone's comparing physical to digital AI, which of course in digital AI and especially LLMs because that was the one thing a lot people understand now with child GPT and cloud then everything else is like you have the whole internet let's say.
• 20:49
  Speaker 0: To train language models on so they become better when interacting with them.
• 20:54
  Speaker 0: When we move into the physical world The number-one point it's like You need physical data So that you can train other physical agents Like robots That know how to react to physical realities.
• 21:05
  Speaker 0: And if you kind of think about physical AI as the whole thing, like... You have to include perception systems that understand world and models in this world.
• 21:14
  Speaker 0: The way those are being stored and processed then spread out to fleets.
• 21:20
  Speaker 0: so there's a massive sector which is why I was asking actually put everything together kind of global level.
• 21:29
  Speaker 0: If you then look only at the robots and only on the robot in warehouses, then obviously get smaller and smaller.
• 21:36
  Speaker 0: And I feel like automation is nothing new.
• 21:39
  Speaker 0: It's been around for ages.
• 21:40
  Speaker 0: But what the robots can do now?
• 21:42
  Speaker 0: How much better they are how they react to all the flexibility requirements of different spaces.
• 21:48
  Speaker 0: That's an interesting side of things.
• 21:50
  Speaker 0: it a very hot topic And actually, I feel Europe is doing really well.
• 21:55
  Speaker 0: Um...I mean i'm super proud for us to be part of a great community and we cover so many different use cases across the European ecosystem from moving things like scanning things to humanoid aswell that have some fantastic humanoid stories here aswell.
• 22:11
  Speaker 0: um..and it's only gonna get better & better.
• 22:14
  Speaker 0: The number one thing would say Let's solve is actually kind of like open up the books and where some of the big grounds are coming from.
• 22:22
  Speaker 0: It's not European investors investing in the European companies, it's investments from Asia... Of course the US but Middle East especially there just a lot money for Middle East coming into the European side of things.
• 22:34
  Speaker 0: So I would always kind of urge European investors to really be a lot bolder when it comes to full-stack companies because, ten years ago and we started off course.
• 22:42
  Speaker 0: everyone was running the other way.
• 22:43
  Speaker 0: When you're talking about supply chains physical production all that stuff like oh my god!
• 22:48
  Speaker 0: That's very capital intensive but don't want get anywhere near it now its'a hot topic.
• 22:52
  Speaker 0: still people are like ugh!
• 22:53
  Speaker 0: But the unit economy should be better or finance will have to invest in this physical staff not going to have.
• 23:02
  Speaker 0: You can do as much digital AI if you want, but no one's gonna put it in practice.
• 23:06
  Speaker 0: obviously that's kind of the bottleneck forever.
• 23:09
  Speaker 0: so I think to go back your question The number-one thing then is to happen.
• 23:13
  Speaker 0: i feel this like European investors just need To be bolder and take bigger bets earlier In the stages of these companies.
• 23:21
  Speaker 1: Yeah right?
• 23:21
  Speaker 1: I find This question always interesting because It's Always different depending on if I'm asking a founder, or if i'm asking an
• 23:29
  Speaker 0: investor.
• 23:30
  Speaker 1: You guys kind of pass the buck from one to another you know?
• 23:34
  Speaker 1: It's always only other ones that it is not working out but...I think things are going the right way as you said.
• 23:40
  Speaker 1: A hundred
• 23:40
  Speaker 0: percent
• 23:42
  Speaker 1: yeah!
• 23:42
  Speaker 1: Yeah, it's always interesting to me to see that kind of difference in narrative between the founders and the investors.
• 23:53
  Speaker 1: But but its a symbiotic relationship And It seems moving into right space would then be on this whole Europe angle.
• 24:03
  Speaker 1: as I mentioned briefly That you have expanded to US something glass they saw there was some opening in Nashville just from something I saw there.
• 24:12
  Speaker 1: So what's that been like as a European company moving into the US?
• 24:16
  Speaker 1: Has it been an easier process because we hear so much about how the U.S is so easy and regulation holds Europe back, and they're so much money in the
• 24:25
  Speaker 1: u.s.,
• 24:27
  Speaker 1: has it been more easily less friction processed for dexery ?
• 24:32
  Speaker 0: First of all ,we've been kind global since day one I would say because working in logistics and supply chain, our customers are global.
• 24:40
  Speaker 0: So from the beginning we knew that we had to be able to operate anywhere.
• 24:44
  Speaker 0: so it goes back through your question earlier like The reason is so easy to deploy an install and ship everything else It's Because We Knew That We Had To Be Able To Operate Everywhere.
• 24:52
  Speaker 0: so were actually live across about seventeen different markets And we had our first
• 24:57
  Speaker 0: U.S.,
• 24:58
  Speaker 0: well, deployments with actually European customers that took us to their US sites.
• 25:03
  Speaker 0: I think it was early twenty-twenty four so way before We even have an office or physical presence in the u.s itself and i always kind of tell people There's no longer this need like you have To kind of start from scratch in a market build A local team push on marketing Like.
• 25:17
  Speaker 0: obviously The whole world is global now especially when It comes to logistics.
• 25:20
  Speaker 0: just let your customers take you into a new Market see how that's going and then start kind of building locally.
• 25:25
  Speaker 0: So, that is what we did for North America And with it open our Nashville HQ in March.
• 25:31
  Speaker 0: It was like a really cool event.
• 25:33
  Speaker 0: We had so many people coming over A lot existing customers new prospects as well.
• 25:39
  Speaker 0: I think it was the right time to do because now have significant presence there.
• 25:42
  Speaker 0: when comes customer base Have very big team across whole country But also from their were serving Canada Mexico which are important markets.
• 25:52
  Speaker 0: So I would say like, yeah the advice on when we give people is that you don't have to think about it as a bottom stop.
• 25:57
  Speaker 0: Slowly invest lot of money into your marketing and everything else.
• 26:01
  Speaker 0: just try find customers who can take with them in their market.
• 26:05
  Speaker 0: then start building around that.
• 26:07
  Speaker 1: Okay.
• 26:08
  Speaker 1: so what's next?
• 26:10
  Speaker 1: You've been around for over ten years now, you're a series.
• 26:13
  Speaker 1: see if its global since day one.
• 26:19
  Speaker 1: What comes next for a company like you?
• 26:20
  Speaker 1: Is an IPO of interest to the founding team or do you prefer being a private company.
• 26:27
  Speaker 0: Yeah, and first I just wanted to clarify between twenty fifteen when we incorporated in like twenty-twenty where it started.
• 26:34
  Speaker 0: look at logistics.
• 26:35
  Speaker 0: It was very kind of slow bootstrap process.
• 26:38
  Speaker 0: obviously We did have some solutions from retail that were going well.
• 26:41
  Speaker 0: That's growing with had some European Union funding which is fantastic At time because he kinda helped us build the team against off the ground.
• 26:47
  Speaker 0: but they didn't really start Properly into the right direction until we launched a logistic solution with just twenty-twenty three.
• 26:53
  Speaker 0: So it's one of those typical kind overnight success that took a decade.
• 26:57
  Speaker 0: But it was specifically when will launch to brand Exury and this product.
• 27:00
  Speaker 0: like I said, what kind of three in about years ago?
• 27:03
  Speaker 0: That would start you get funding again because we found our product markets hit and everything else.
• 27:06
  Speaker 0: You might be working on an idea for long time.
• 27:09
  Speaker 0: Then you find out nugget or play actually put turn it into this In my creative business.
• 27:13
  Speaker 0: so you could argue like Are you looking a decade back?
• 27:16
  Speaker 0: Which, one hundred percent like all that experience from the beginning helped us build what Dexer is today.
• 27:20
  Speaker 0: But for everyone else's view, Dexers are only about three and half years old so... Yeah, it's always like what is under the bottom of the iceberg.
• 27:30
  Speaker 0: But what's coming next?
• 27:32
  Speaker 0: We have some pretty pre-ambitious goals for at the end this year.
• 27:36
  Speaker 0: Of course our team has grown quite a lot.
• 27:38
  Speaker 0: but all that I was saying earlier about data.
• 27:44
  Speaker 0: we are sitting on the biggest database in warehouses like warehouse operations, I'd say.
• 27:50
  Speaker 0: Like we have over one point six or seven billion locations scanned so far quite a few hundred sites that we're capturing day after, the other days.
• 27:58
  Speaker 0: So that allows us to really understand how warehouses operate and really build those warehouse models.
• 28:04
  Speaker 0: going back my point earlier no one else has access too.
• 28:06
  Speaker 0: so We are looking at what else can we extract from this?
• 28:08
  Speaker 0: To help our customers like connect the dots between different facilities, different markets they have but also working with quite a lot of robotics players in the field to see if data helps them become better.
• 28:18
  Speaker 0: because again having whole machine big, expensive.
• 28:23
  Speaker 0: It's complicated and some of the others in the field don't necessarily have to do that because we already have the infrastructure for them.
• 28:30
  Speaker 0: so looking at how we can orchestrate more or more automation by using our data
• 28:36
  Speaker 1: How does it play into?
• 28:36
  Speaker 1: I read my research but it popped out too.
• 28:41
  Speaker 1: me was you had a focus on ethics in robotics and unbiased AI.
• 28:48
  Speaker 1: So how does this kind of the goal that company play into?
• 28:52
  Speaker 1: How do they even play to what you do
• 28:54
  Speaker 1: concretely?".
• 28:55
  Speaker 0: Yeah, there's a lot different elements to it.
• 28:57
  Speaker 0: so first off all I mean if we kinda break down on the bias side of things If you train NEI with data sets very similar from very similar environments Of course like your gonna have certain results are going be quite repetitive based on what data you've given them.
• 29:15
  Speaker 0: So when it comes to very practically speaking the customer base that we have, um... We are working across various different types of customers because we intentionally wanted to do there from beginning with our retailers and our pharma companies who help automotive players like pet food.
• 29:29
  Speaker 0: It's just an end-to-the point earlier as well some facilities at one million square feet somewhere like just a couple twenty or thirty thousand square feet.
• 29:37
  Speaker 0: so where trying to capture is much different type of data and diversified environments, in different types of goods.
• 29:44
  Speaker 0: And geographies so that you have this kind a very strong database that can then obviously train models to be able to understand based.
• 29:52
  Speaker 0: okay where am I now?
• 29:53
  Speaker 0: I have all these previous information so i know how to react it.
• 29:56
  Speaker 0: So there's an element of that bias.
• 29:59
  Speaker 0: when he comes to the ethical side as well is like We make sure that we have, first of all very international team.
• 30:07
  Speaker 0: Very diverse teams from the perspective of engineers who are working in but also at a leadership level.
• 30:13
  Speaker 0: so again based on our user base which is a global user-base The way to build out products keep it in mind and then test things as well kind of uses the team across the board.
• 30:23
  Speaker 0: So going back where customers are makes us do our best think like they might think or use their product Focus our putting on that.
• 30:34
  Speaker 0: and also when it comes to yeah, the other side of I think switches.
• 30:39
  Speaker 0: The regulation side you're talking about earlier trying get involved in those conversations then Be there be top-of-mind for regulators not Not just a kind of push back from them but also to make sure that if they are building certain Frameworks that their builds based on the reality on the ground.
• 30:56
  Speaker 0: Like again, we see so many different kind of sites across the markets.
• 31:00
  Speaker 0: I know there's a lot of work happening here in UK for instance at the moment on robotics and We try to advise them all like.
• 31:05
  Speaker 0: this is what people are telling us on the ground.
• 31:07
  Speaker 0: This is exactly where were seeing rather than if theory should be like this.
• 31:11
  Speaker 0: Well just let make sure that we have practical element in it.
• 31:14
  Speaker 0: So yeah will trying get involved with those conversations.
• 31:19
  Speaker 1: I saw something.
• 31:20
  Speaker 1: You were in Downing Street, weren't you?
• 31:24
  Speaker 1: What was that on?
• 31:25
  Speaker 1: is this the same idea of getting involved with regulation talks?
• 31:29
  Speaker 0: Yeah, I mean that particular one and like we've been a few times which is kind of yeah great to have the privilege of obviously kinda here getting your voice out there.
• 31:38
  Speaker 0: And becoming an expert in your field right but um i think you had that particular One was more focused on talent and again We are struggling with talents In our field.
• 31:46
  Speaker 0: so trying To figure how can we attract More people into The robotic space then into companies Like ours because There's a lot of shiny things out there, especially when it comes to all the big AI companies that are obviously paying the big bucks.
• 32:02
  Speaker 0: A lot of great talent is going in this way.
• 32:04
  Speaker 0: so we're trying figure how to make sure they stay relevant into space.
• 32:09
  Speaker 1: I think actually that leads naturally to last topic before we wrapped us up and You talk about diversity, you've talked about talent and I think that kind of leads us into a gender question as well which i think is important.
• 32:24
  Speaker 1: Just if look at our own catalog.
• 32:26
  Speaker 1: we have been trying really hard to get more female voices involved with us in highlighting female founders hopefully bring more attention that they do exist, and you know it's not some unicorn in the wild.
• 32:37
  Speaker 1: It is actually quite common but there still work to be done.
• 32:42
  Speaker 1: The reason why I even prompted us was because we had a look at our catalog when we realized ninety-five percent of this has meant maybe we should make more of an effort here to highlight other kinds of voices.
• 32:54
  Speaker 1: so then on your end what does that ecosystem looking like?
• 32:58
  Speaker 1: Do you feel its fairly gender neutral workplace, some work balance when it comes to robotics and physical AI or is there still kind of work to be doing on that front?
• 33:09
  Speaker 0: Yeah.
• 33:09
  Speaker 0: It's a great question because I feel our sector has actually probably lot more diverse than people realize.
• 33:15
  Speaker 0: And even if i look at our own teams like we have A lot of women working in very different type Of roles and I think maybe Robotics Is a bit about reputation Because People just Think About you Have To Be Hardware Engineer You Have To be Software Engineer have grown up with a screwdriver in your hand and the search type of profile.
• 33:34
  Speaker 0: But actually robotics is business at end-of-the-day, which has to have different functions for businesses.
• 33:39
  Speaker 0: so you don't necessarily need an engineer who works in robotics.
• 33:42
  Speaker 0: You can definitely work on commercial side things like customer success.
• 33:46
  Speaker 0: There are many different angles but even when it comes The engineering roles like people focus too much on the product you're building and not as much On.
• 33:57
  Speaker 0: The end customer, and a problem your solving with that adaptable products.
• 34:02
  Speaker 0: So if I'm looking at the logistics sector is actually very diverse sector.
• 34:05
  Speaker 0: We have so many female users across different types of teams.
• 34:08
  Speaker 0: are we interacting with it?
• 34:09
  Speaker 0: I think people just keep driving with the tech when they're talking about robotics rather than actually talking about the benefits and use cases, which I feel will attract a lot more kind of female talent to the field because you are realizing these problems that understand.
• 34:26
  Speaker 0: So I think there's a big kind of element and then, I was talking about this the other day as well around like we just need to change little bit of topics that were bringing forward.
• 34:35
  Speaker 0: And of course examples are also women in field.
• 34:39
  Speaker 0: make sure don't focus on technical side things but actually building business around it.
• 34:47
  Speaker 0: But I think things are going in the right direction.
• 34:49
  Speaker 0: And even today, I think I saw on LinkedIn a post about how many billionaires have been created by AI there women which is great to see.
• 34:57
  Speaker 0: it's probably like one of what we're gonna call sector or an industry whatever you want to call their capability that has created The biggest number of female billionaires.
• 35:07
  Speaker 0: That's all the article was saying.
• 35:10
  Speaker 0: and uh i think that kind Of shows again that maybe with AI taking away that bias around like you have to be highly, highly technical and kind of studied.
• 35:18
  Speaker 0: And we had a PhD in computer science.
• 35:21
  Speaker 0: You can actually build so much more with those tools and focus again on what problem your solving which is fundamental human problems rather than gender specific one.
• 35:32
  Speaker 1: Okay Yeah well I think that puts nice little bow over our chat here.
• 35:36
  Speaker 1: It's it's...I think We've covered lot your whole history here, but also love interesting looking forward to the future and your opinions on physical AI and robotics.
• 35:45
  Speaker 1: And all that which I think is yeah it's pretty unique i think so far in terms of conversations ive had out there.
• 35:51
  Speaker 1: So thank you for that very much Juanan For joining me today.
• 35:55
  Speaker 0: Thank You so Much.
• 35:56
  Speaker 0: Love you to chat.
• 35:57
  Speaker 1: Okay we'll keep in touch then.