Robotics, Data, and Optimization with Dexory’s Oana Jinga
Episode Transcript
DD Spotify DD Apple Podcast

In this episode, Blythe and Oana Jinga, Co-Founder and Chief Commercial and Product Officer at Dexory, discuss the company’s tall, autonomous robotics solution that collects granular data from warehouses to create a dynamic digital twin environment.

Jinga shares insights on leveraging this data for optimization, consolidating product locations, identifying discrepancies with WMS systems, and orchestrating other robotics operations.



  • “The way we look at a digital twin is like it’s a dynamic and living and breathing digital object. So it needs to be able to show you at any point in time, what’s the latest that’s happening in the physical world [of your warehouse], how that’s evolving, and what factors are influencing that change.” – Oana Jinga
  • “The robot is there to collect information. But what you can do with that information, is pretty much limitless. You can extract data and different points, you can compare it with other data sources, you can crunch it, you can use it for simulation for forecasting, and so on.” – Oana Jinga
  • “So when we say a digital twin of a warehouse, we literally mean like this digital environment that changes continuously as the warehouse changes, and that you can also then use in the future, like if you do want to simulate and try different scenarios.” – Oana Jinga
  • “From our perspective, we use the robot to get data continuously from warehouses to push them into that digital twin and show as close to reality as possible what’s going on.” – Oana Jinga
  • “We take continuous pictures of the racks, and then we deconstruct those and can use computer vision. And again, some elements of machine learning to see what we’re seeing in that data.” – Oana Jinga

Feedback? Ideas for a future episode? Shoot us a text here to let us know.



Are you experienced in freight sales or already an independent freight agent? Listen to our Freight Agent Trenches interview series powered by SPI Logistics to hear directly from the company’s agents on how they took the leap and found a home with SPI freight agent program.

Tai TMS is designed to streamline your brokerage operations and propel growth for both FTL and LTL shipment cycles. Book a demo with the Tai team today and tell them Everything is Logistics sent you.

Maximize your website’s performance as a sales tool with Digital Dispatch’s website management.

Show Transcript

See full episode transcriptTranscript is autogenerated by AI

Oana Jinga: 0:05

Probably the one thing I wanted to touch on is from our perspective. I mean, the robot is there to just collect information, but what you can do with that information, then it's pretty much kind of endless and limitless, because you can kind of extract data and different points, you can compare it with other data sources, you can kind of crunch it, you can use it for simulation, for forecasting and so on. So normally the way we kind of approach things, our product at the end of the day is the data rather than the robot itself, versus some of the other solutions that I was mentioning earlier, that you referenced, where the actual robot is the one that's actually doing the job.

Blythe Brumleve: 0:42

Welcome into another episode of Everything is Logistics, a podcast for the thinkers in freight. We are proudly presented by SPI Logistics and I am your host, Blythe Brumleve, now today on the show. I'm happy to welcome in Oana Jenga. She is the co-founder, chief commercial and product officer over at Dexory, and we are going to be talking about robotics, data and technology inside of warehouses. So this is a very timely discussion considering everything that, especially around the explosion of e-commerce orders and just the adoption of tech overall within logistics. So, Oana, welcome to the show.

Oana Jinga: 1:16

Hello, thank you so much for having me and, yeah, I really regret to talk about all these topics because obviously they're very, very dear to my heart.

Blythe Brumleve: 1:23

Now I want to take it back to you know, before you got into logistics, you spent I think what was it six years at Google, so you spent a long time there. And so I'm wondering what was the catalyst from working at Google to saying I want to start my own company?

Oana Jinga: 1:41

Yeah, and I think it probably was way back when I kind of had that initial thought in my mind. I don't know why, but even kind of growing up I always had this idea that one day I'll have my own company. I don't think at the time I would ever think what exactly in which sector, like what exactly the company will do and why, but I think always kind of had this inkling and kind of looking back. I don't really know where it came from, because my family it's kind of very traditional, like doctors and lawyers and everything else and like you have to have a very stable profession and all of that, and kind of went a little bit against the flow. But I think my first step into technology was actually one of my first jobs.

Oana Jinga: 2:19

I used to work at Telefonica, so Telco, working on some of the digital products very much in line with, I mean, all the typical apps that we have today, like your chat apps and your video calls apps.

Oana Jinga: 2:32

At the time when I started there they were just very much in their infancy, so I was part of that initial kind of kickoff and boom of communication apps. And then from there I moved over to Google, like you said, where I had a chance to learn from some fantastic people again going back to building products, taking them to market and understanding what people want and kind of feeling that one with technology. So it did kind of prepare me very, very well to having the skills and knowledge and the desire to build something from scratch myself. But probably due to mention, it's not just me, we're three co-founders and the other two co-founders are actually a lot techier than me. They are kind of software developers, engineers and have a very kind of strong background into that. So they kind of came with the angle of robotics and I kind of came with the angle of, okay, robotics, so what, and who's going to pay and buy this? And that's kind of how Dexory kind of came together.

Oana Jinga: 3:22

And that's kind of how Dexory kind of came together, and so I'm curious as to what were some of those, I guess, lessons that you learned at Google that still stick with you to this day. I think I was talking about that the other week with somebody in the team and there's many, many different things. I mean some good and some bad as well. I mean, google is an extraordinary company to work with, but sometimes you're also thinking like, oh, I probably don't want to do that in my own company because it wouldn't work.

Oana Jinga: 3:50

Um, but um, starting from just what I was saying earlier around that idea on how do you even start when you want to build something, like, where is the insight?

Oana Jinga: 3:58

How does that come from the, the customer, and um, google has this kind of mantra around like know the customer, know the product and then connect the two and make magic. Basically, um, so, um, really understanding how to to read kind of people, needs and wants, like obviously, like all the, the tools to to bring that together, but then also trying to figure out how you turn that into into a solution and the product was probably like the, the biggest learning. Um, but then also a lot of different things around how to manage big teams technology teams as well versus like business teams, and how everyone kind of operates a little bit differently. Having systems like well OKRs, like objectives and key results in place to be able to guide everyone towards the same directions of a company so multitude of things. Yeah, going across all the different areas of a business so a multitude of things going across all the different areas of a business.

Blythe Brumleve: 4:45

And so when you are first coming up with sort of the initial concept of Dexory, what were some of those, I guess, those key customer insights that you were trying to solve for them?

Oana Jinga: 4:56

So, funnily enough, we actually didn't start in logistics. So the background of Dexory is actually retail. So we started with having an autonomous robot that would do stock checking and inventory taking inside retail stores. This was just before the pandemic. I mean, it was going really really well and it's kind of like booming at the time and then hey ho, over 2020, everything kind of closed.

Oana Jinga: 5:16

So we actually had our retail customers from back then ask us well, obviously the stores are closed, but we do have a massive problem in the warehouse. Obviously, at the time we couldn't even kind of get people in and they had to be distanced and everything else. So can you actually do what you're doing in the stores but on the warehouse side and help us keep track of inventory and our stock? So it started with a customer ask and I think ever since, everything we've been building has always kind of been on top of a customer need or a customer question um, yeah, our customer interest. So it's always kind of very important to to try and identify the early days so you mentioned that the the almost the robotics focus shifted from 2019 to 2020.

Blythe Brumleve: 5:59

Has it evolved since then or has it kind of stayed the same since 2020? Yeah, it's, it's constantly evolving. I mean, at the essence it's, it's pretty much kind of stayed the same since 2020?

Oana Jinga: 6:04

Yeah, it's constantly evolving. I mean, at the essence, it's pretty much kind of the same in the sense that we use an autonomous robot to collect information and data inside warehouses, we digitize that into our digital twin platform and obviously present our customers with a lot of insights and information. But the way we collect that data, the type of data we collect, how we deconstruct it, how we present it, all of that pretty much changes, I think, week to week, as we keep evolving the product and keep learning more from our users. So it's a constant kind of trial and error and understanding of the market.

Blythe Brumleve: 6:36

So how does? I'm curious because, for I just, I got introduced to Dexory at the Manifest Conference Future of Logistics and Supply Chain, or Future of Supply Chain and Logistics and I stopped by the booth because you guys had this amazing, you know, Lego kit, which was really creative marketing, and I had to get my hands on it. So I was one of the lucky few to get my hands on it and I'm keeping it in the bag. I'm scared to build it because I don't want to mess it up. I feel like it's going to be a collector's item. So was Manifest sort of maybe one of the first sort of coming out moments for Dexory, or were you on the trade show sort of circuit before then?

Oana Jinga: 7:14

It's a very, very good point because we actually so maybe kind of for the listeners we are based in Europe, so we're based in the UK, just kind of outside London. We have our kind of big facility for manufacturing and commissioning robots and bringing them together, but we do have customers around the world. So Manifest for us was a great opportunity to actually introduce our product or solution and our team to the US market in particular. We did have like a couple of customers already live at the time, but we had never really done something very big for the US market. However, kind of going back to your question, we have done some of the bigger kind of trade shows and events in Europe for probably the past kind of two, three years now, Slowly of course, kind of growing that presence as the company evolved as well.

Blythe Brumleve: 7:56

But Manifest was kind of our first step across the pond. I'm curious as to what the differences between. Are there any differences between the trade shows and events in Europe versus the United States?

Oana Jinga: 8:05

are there any differences between the trade shows and events in Europe versus the United States?

Oana Jinga: 8:11

It's a very good question because I think last well, this kind of like last quarter, when we started with Manifest, we then moved over to, I think we had Modex and we had Logimat in Europe and then we had one in the UK as well Interlogistics so it was very interesting because the core team was pretty much the same going from one to the other and at the end I mean like, oh my god, this experience couldn't have been different and I think it's a.

Oana Jinga: 8:28

It's a very big cultural difference around how, um, I mean, I don't the american culture just embraces innovation and ask questions and they're very curious and want to know more, um, more kind of central european, a little bit more skeptical and technical, and go into a lot of technical details, so that the people that you speak to react to very different things, which is not something we had thought about before, but definitely like a big, big learning after this season, as we call it, because now kind of things calm down in logistics for a while, which is very, very strange. You do a lot of beginning of the year and then nothing for a few months.

Blythe Brumleve: 9:01

Yeah, it's definitely. You have your trade show peak seasons and we definitely just ended, I think, for a good majority of it. You'll have a few over the summer, but then the fall is going to pick right back up and I think Manifest kind of kicks that off, at least in the North American market. I've had somebody on the show previously from Pitney Bowes and they were talking about the differences from warehouse operations in Europe versus the United States and I'm curious if you've seen you know sort of a similar issues or maybe challenges that go on, because for a lot of, from what I understand, for a lot of European customers, the way that they're dealing with warehousing is that it's a lot smaller of facilities or a lot smaller buildings, retail shops, things like that, and so the facilities and the delivery from that point A to B has to be more efficient in a smaller way in Europe versus the United States. Are you seeing some of those similarities as well?

Oana Jinga: 9:57

I mean yes, yes and no. I would say it kind of depends where in the United States and then what in Europe as well. Because, um, I mean, if we go back to the basics, the one thing we've definitely noticed, like in the us you have warehouses that are much larger horizontally, whereas in europe, and especially the uk, people are building vertically a lot more because of, yeah, the cost of real estate. So you would have warehouses that are 18, 20 meters in the uk I mean, most of the us, let's say they're about, like I don't know, 12, 14, even that would be a tall one. So there's like an interesting kind of element of infrastructure to start with. The other element, of course, because of space, and you will have more kind of smaller, let's say, distribution centers. I wouldn't necessarily kind of call them more warehouses, because some of them will be be quite small, like you said, it's like the back of a retail store and people are distributing from there, um. But on the other hand, I mean there is a lot of trade and commerce and movement of goods happening across europe, so there are also massive, massive parks that have literally kind of come up, um, over the past couple of years and they keep kind of expanding and growing. So most of the motorways around cities, like you'll see huge industrial parks with really big warehouses as well.

Oana Jinga: 11:03

Um, so I wouldn't say it's necessarily a trend. From what we're experiencing, there's definitely a lot of like really big sites as well. Like the biggest site we're actually deployed in has 110,000 pallet locations, about 1 million square feet. It's in actually Eastern Europe. So, yeah, it's a site with Maersk. I can kind of like very openly talk about that and they're distributing their goods across kind of Eastern and Central Europe. And then also we have sites in the US which are much smaller. So it's yeah, it's not necessarily a trend, but the element of real estate, I think, is what kind of dictates a little bit the possibilities of where you can build over here versus in the US.

Blythe Brumleve: 11:40

Yeah, great point. Because when I was at Manifest the location that it was at before it's not going to. It's going to be at the Venetian for next year, but the location it was for this year there were some folks that were on the trade show floor that could not build their booths as high because the ceilings were or chandeliers were kind of getting in the way of them, showing off you know, their different sort of robots. That were robots that were picking and packing and things like that, and so the next phase is going to be the higher ceiling. So that was interesting from a logistics perspective of those types of people that are setting up their boots. So it's really interesting that it's almost the same thing for warehouse locations as well.

Blythe Brumleve: 12:20

Are you in freight sales with a book of business looking for a new home, or perhaps you're a freight agent in need of a better partnership? These are the kinds of conversations we're exploring in our podcast interview series called the Freight Agent Trenches, sponsored by SPI Logistics. Now I can tell you all day that SPI is one of the most successful logistics firms in North America, who helps their agents with back office operations such as admin, finance, it and sales. But I would much rather you hear it directly from SPI's freight agents themselves, and what better way to do that than by listening to the experienced freight agents tell their stories behind the how and the why they joined SPI? Hit the freight agent link in our show notes to listen to these conversations or, if you're ready to make the jump, visit spi3plcom. You've mentioned a couple times the digital twin environment. How do robotics sort of sync up or complement a digital twin environment? Are they the ones scanning the warehouse and reporting back the data? Is that kind of a summation of how they play together?

Oana Jinga: 13:27

So from our perspective, the robots are there to just collect information and then they digitize everything and on the edge, on the actual unit, and push the data back into a cloud platform where you kind of have that digital twin. And when we say digital twin, I think maybe it's worth just clarifying what the definition is, especially from our perspective. So sometimes people call a digital twin any kind of representation of a physical object in a digital format, any kind of representation of a physical object in a digital format, so a CAD file or, let's say, like a construction model of a warehouse. A lot of times people say that's a digital twin but it's not really. It's a model, so it's been done once. It kind of like never changes and you kind of go back to it from time to time to try and figure out where the fire escape is or something like that, because that will not move. But the way we look at a digital twin is like it's a dynamic and living and breathing digital object, so it needs to be able to kind of show you at any point in time what's the latest that's happening in the physical world, how that's evolving, where the factors are influencing that change. So from our perspective. When we see a digital twin of a warehouse, we literally mean like this digital environment that changes continuously as the warehouse changes and that you can also then use in the future, like, if you do want to simulate and try different scenarios. You kind of use it to to see how that reaction obviously will will evolve over time based on whatever kind of happens to it.

Oana Jinga: 14:44

So in in our perspective, we use the robot to get data continuously from warehouses to push them into that digital twin and show as close to reality as possible what's going on.

Oana Jinga: 14:54

But moving over to other types of robotics as well, the one element that digital twin helps the one that we create is that we can then feed this data back to other types of robotics and help them best kind of operate and orchestrate within a warehouse.

Oana Jinga: 15:08

So let's say, if you have picking robots and they would go to a location to get, obviously, a product, but actually that location is empty because it hasn't been replenished, I mean we flag that pretty much kind of real time and make sure that there is enough product there so that the picking robot goes to the right location. Similarly for autonomous forklifts, like the 3D environment that we have and the information we have about pallets and goods can feed into their kind of data points to know where they're going, what they're picking, where they're putting things as well. So it's kind of two ways. Obviously, from strictly our perspective, we build the digital twins and obviously we enable customers to get information from them, but then from us to other robotic systems as well. They become that kind of orchestration layer and platform for any kind of technology.

Blythe Brumleve: 15:50

That's super interesting that your data is syncing with other robots or robotics within the warehouse itself. Now where is that data being stored? Is it within your system? Is it within the WMS, or are they two different, complementary systems?

Oana Jinga: 16:06

So they're two different, complementary systems in the sense that WMS has a very different purpose, so it's being obviously designed to hold information about suppliers, dates, things come in, when they move, how much quantity has to go where, whereas what we do is, rather than kind of going horizontally across kind of all these different types of data points about goods, we're also going kind of vertically within, really inside the warehouse, so really understanding kind of racking structures, occupancy levels, quantity and volumes of every single kind of data point. So all of that information, together with, like I don't know, high resolution images, for example, you kind of suddenly put them back into the WMS because it hasn't been designed to store and keep that kind of information. So our system kind of works very much in conjunction to WMS and becomes kind of that one source of truth across the warehouse.

Blythe Brumleve: 16:53

And because the shape of a Dexory robot is very vertical and so, with other sort of robotics within the warehouse, it feels like from the people I've talked to and the things that I've seen is that robotics within the warehouse are doing a very specific job. Is that accurate?

Oana Jinga: 17:14

Yeah, and I think that's probably I mean, sometimes we're being asked about, like, what competitors do you have there in the robotic space?

Oana Jinga: 17:21

And I think the answer I usually use is if you kind of go into your kitchen, you have a lot of different appliances and they're all kind of designed and meant to do different things.

Oana Jinga: 17:28

Like you would not use a dexory robot to pick something or to move a pallet, the way you wouldn't use, like a locus robot to scan racks up like 50 meters in the air, because they're not designed to do that.

Oana Jinga: 17:40

So I feel like the importance of kind of the shape factor and the way like a unit is built has to always kind of keep into account exactly what the purpose of that unit is and what it's meant to be doing. Hence why, like yeah, we took the approach of having kind of these really tall units that can scan warehouses pretty much kind of top to bottom in one go. So with that we can do about 10,000 to 15,000 padded locations an hour, because it kind of acts like a giant scanner, if you think about it, kind of moving around the warehouse, and we have to be able to kind of moving around the warehouse and we have to be able to kind of get to that height to keep that speed up and then the consistency of the data. But of course, like, if you don't have to go up there for other use cases, then you don't need a tall robot.

Blythe Brumleve: 18:21

Yeah, I was always wondering what I would imagine, because the theory that I've heard is that, you know, robotics are very efficient and very good for one specific task, and so you see a lot of different types of robots within a trade show floor, for example, and so I'm always curious as to how do you make those different decisions on what to get or what to invest in. And that makes a whole lot of sense that you would get this certain kind to do this specific job and you would get this other kind to do another specific job, because warehouse would get this other kind to do another specific job, because, you know, warehouse labor is incredibly challenging to fulfill. I think it was amazon that said a couple years ago that they were, in the us at least, that they were in danger of actually running out of people to hire because it's such a demanding job and there's such a you know, a high turnover rate. I see you shaking your head, that's, it seems like you agree with a lot of those statements.

Oana Jinga: 19:13

And what's funny is that, I mean, a lot of our customers are telling us that the problem is that Amazon is taking all the people. So if Amazon's struggling, then you can imagine how everyone else is who can't necessarily afford to kind of keep throwing money at the problem and, yeah, kind of restrained by other types of resources as well. So absolutely, and I think, kind of going back to your point a little bit earlier around kind of different use cases, probably the one thing I wanted to touch on is from our perspective. I mean, the robot is there to just collect information, but what you can do with that information, then it's pretty much kind of endless and limitless, because you can kind of extract data and different points, you can compare it with other data sources, you can kind of crunch it, you can use it for simulation, for forecasting and so on.

Oana Jinga: 19:54

Um, so normally the the way we kind of approach things. So our product at the end of days is the data rather than the robot itself, versus some of the other solutions that I was mentioning earlier that you referenced, where the actual robot is the one necessarily doing the job. Um, because we talked about picking and carrying and so on, I mean you, the robot is the one kind of performing that task and that's kind of where it ends. Of course it's a lot of complexity behind the technology, but for us capturing the data is just step one. It's like what you can do with it afterwards is where the magic kind of happens I was saying earlier, and where the most value can be extracted from.

Blythe Brumleve: 20:27

Well, could you elaborate a little bit on some of those actionable tips, because I think it's one thing to get data, but what do I do with that data next, after I see it?

Oana Jinga: 20:37

Exactly exactly.

Oana Jinga: 20:38

And I think the most basic element around, like we, I mean, at the core of our technology is the fact that we can instantly find discrepancies versus warehouse management system data or other kind of systems data. So the first step is like okay, once you kind of see a discrepancy, what are you going to do about it? Is it a physical movement of a good or is it actually like a system change that you have to do? But then it gets a lot more complex and interesting because, like, if you do have, for example, multiple locations with the same product, what we do is we kind of recommend consolidating those product locations and kind of making more space in the warehouse.

Oana Jinga: 21:10

If your distance from a pick to reserve location is too long and that kind of might lead further on to delays, we can also kind of recommend bringing them together based on the occupancy levels that we see around that pick location. So it's again kind of correlating different data points across the warehouse and being able to kind of make recommendations to our users to actually know what to do about it. Because, like you said, if you just kind of present the report at the end of the day it's like okay, fine, somebody like, then has to get their head around, like, what does it actually mean? Whereas we take the next step around, like we actually interpret that and we tell customers okay, if you do X, y, z, then this is the benefit that you're going to see and potentially would it be worth it or not in the end to take the action.

Blythe Brumleve: 21:50

Oh, interesting, and I'm curious are there any other sort of data points, maybe like purchasing habits or purchasing flows that go out, you know, throughout the year? Are those impacting some of the data decisions that Dexory is making?

Oana Jinga: 22:05

Absolutely, and we're kind of like mixing, I think, different kind of data points in terms of like external factors, like you were saying, like is it peak season, is it kind of Christmas approaching and we kind of see different items kind of potentially being in higher needs. We can take information that already exists in kind of ERP systems or even the WMS or simple things. Like one of the use cases we have for the food industry is if a product kind of approaches the best before kind of the date, then you could kind of highlight them they haven't moved in a while to potentially kind of be the first ones to be taken out of the warehouse. So it's a combination of factors that we take into account. Some of them might be completely external and seasonal, some of them might just be quite specific about a particular item or good in the warehouse.

Blythe Brumleve: 22:47

And you've mentioned Maersk a couple times here. They're one of the customers of Dexory. What does that relationship sort of entail? You mentioned a warehouse in Eastern Europe. I also was listening to an interview where it was mentioned that they wanted to have stacking or sort of storage optimization. Can you break that down for us?

Oana Jinga: 23:09

Yeah, and it's fantastic because they've been a great kind of partner to us and we have numerous kind of slides live with them. We're all in all of their locations in the UK. Like I mentioned Eastern Europe, we're kind of in conversations of launching in the US very soon as well. So it's kind of grown very quickly and I think that's kind of testament of our collaboration and the fact that we've been kind of very open to explore the use cases that you discussed as well with them.

Oana Jinga: 23:30

So, um, I think, yeah, where that comes from is kind of different angles. Um, one is a little bit towards I was saying earlier. I mean, they're very much embracing, using that data to then also make decisions around how to best organize their space and be much more efficient with um, with with the resources that they have at hand. Like, yeah, they can kind of bring in multiple customer goods different times and making sure that they also are able to report back to that customer um what stock situation they have at hand. Like, yeah, they can kind of bring in multiple customer goods different times and making sure that they also are able to report back to that customer what stock situation they have.

Oana Jinga: 23:57

But we're also like and we're going to announce this in a little bit more detail soon. We are moving away from just obviously racking to all the other types of kind of racking configuration as well sorry, goods configuration, because it's not racking like things on the floor, pick faces and smaller mezzanines and so on. So I think also being able to have that capability of doing different types of sites and even if a site has different types of configuration, kind of covering both angles and being able to suggest moving goods from one part to the other, was something they were very interested in as well. So, yeah, very excited about that collaboration and they're fantastic at asking for things, which is good for us because we learn and then we have us kind of providing back to them.

Blythe Brumleve: 24:36

Do you find that what Maersk is experiencing on, you know, sort of the enterprise level is a lot of the same issues that maybe SMBs you know small business, small to medium sized businesses are experiencing as well, just at a different scale?

Oana Jinga: 24:53

well, just at a different scale.

Oana Jinga: 24:54

To a large extent, yes, I think what MERS kind of has the advantage of is really understanding the other parts outside of the warehouse obviously the transportation side and obviously shipping there's no need to kind of explain that so they're probably a little bit more in tune and connected to what happens outside of the warehouse to be able to then be much more efficient inside the warehouse, which obviously some of the smaller players because they'll be very much focused on the contract logistics angle or just the site itself probably wouldn't have the same level of expertise.

Oana Jinga: 25:20

But in essence the problems themselves would be very similar. They're also kind of struggling to attract talent and also kind of like onboard people as quickly as possible, retain people and then from there making sure that obviously their side data is as accurate and as good as possible, because without having kind of people doing the checks and keeping on top of that and also with a high level of retention, you will have a lot of mistakes and issues that naturally happen and that will be across the board, whether there's a smaller or large company.

Blythe Brumleve: 25:47

Brokering success demands a battle-ready strategy. Thai TMS equips freight brokers with the ultimate battle station for conquering a tough market. With Tai, brokers gain access to a comprehensive platform where rate intelligence and quote history converge on a single screen. It's not just a page, it's a strategic command center designed to help brokers win. Tai equips your team with all of the data they need to negotiate with confidence and allows them to communicate directly with carriers and customers from a simple control base. Revolutionize the way your brokers perform by giving them a competitive advantage with Tye TMS. For more info, go to tai-softwarecom.

Blythe Brumleve: 26:27

Backslash battle stations and we also have a link for you in the show notes to sign up for a demo. Backslash Battlestations and we also have a link for you in the show notes to sign up for a demo. There is I heard a stat the other day that only 7% of US-based warehouses have robotics integrated into their system and into their building. Based on how you've worked with customers, I'm curious as to why that adoption rate is so low. Is it something that is growing, because it is sort of fairly new? And then, what are the challenges of you know, sort of onboarding or making that decision that you know it's time to invest in robotics?

Oana Jinga: 27:06

I think it's yeah, it's multifaceted, it's multifaceted. The one interesting element from our perspective is, a lot of the times we're one of the first robots to come in because we actually require very little to do upfront, there's very few demands from our customer. All we need is a power plug and then we can pretty much be on our way. We don't even need a Wi-Fi connectivity if that's not available, and also we do not without going into too much detail, but our business model is robot as a service, so we don't require a massive upfront investment to kind of bring a big automation in which makes it a lot kind of seamless and quicker to get customers comfortable to start their journey in robotics actually with us. Because now going back to your second part of the question, I think number one is people fear making a massive investment that will then kind of have a pretty kind of long term to provide a return on that. So if you have to bring in a I don't know significant automation system for sortation or like conveyor belts or whatever it is, to kind of get things from the door to the racks, that will be potentially in the hundreds of thousands or maybe even millions to get started on. So you have to be very, very sure that you want that in, and especially if we're talking 3PLs, like third-party logistics providers, they'll have to be confident that whatever customer they have on site will stay enough to be able to then see the returns of that investment. So that upfront cost is a big hurdle. And then the second part is I feel like people did kind of make the decision that they want automation. It's just like having the confidence again to start and really understand that again.

Oana Jinga: 28:44

Technology has evolved a lot and it's probably much easier and quicker and less kind of intense from an internal resource perspective to actually get some of the technology out than they think. So there's some misconceptions around, like it will take months and months of integration and I'll need 20 different partners to work together. Yeah, I need to kind of revisit my staffing and everything. So there's a lot of kind of misconceptions around that. And I think most of the companies that have evolved in the industry are way beyond that and they can get up and running much quicker with very few resources. So yeah, it's kind of like time and money at the end of the day, which kind of makes decisions a little bit harder. But I do think, again, from my knowledge, that the US is actually quite far ahead versus some other markets. Of course, if you go more into Asia, some of them might be more advanced, especially when it comes to manufacturing rather than logistics. But yeah, I think others are even more behind.

Blythe Brumleve: 29:30

Yeah, I think others are even more behind. Yeah, I imagine that if you're facing that, that might be the catalyst for these folks to adopt robotics, because they face so many different hiring challenges over the last few years and it's probably not going to be going away anytime soon. When somebody does make that decision, OK, I'm going to invest in robotics. What does that sort of onboarding initial you know, couple weeks, maybe a couple months look like From?

Oana Jinga: 29:56

our perspective it's very, very seamless, like we can be up and running in about five to seven days, depending a little bit on the size of the site and, like I said, all we need is a power plug.

Oana Jinga: 30:04

So it's a very, very kind of fast journey. So normally what we do is we, depending on where we kind of start a conversation with the customer, because a lot of our customers obviously have multiple sites and it might be like a regional conversation or directly a local conversation with a site. We try to get our heads around operations very quickly, understand what's happening on the ground, get an idea of the layout like where would the robots can, how would it go around, what would be the routes, and then we pretty much kind of ship a unit. It comes in two boxes and it takes like just an hour to put together like that lego block. You should time yourself, by the way, when you build a lego, because there is a bit of a competition going oh, wow quicker if you can do that quicker than our team can put together an actual robot.

Oana Jinga: 30:48

Um yeah, so it comes into into the boxes and then, like yeah, they put it together and in a couple of hours we do a map of the site by literally just driving the robot around. Using the, the sensors on the robot, it kind of captures its kind of environment in 3d and 2d and from there we just run a couple of kind of tests to make sure that we're capturing all the kind of barcodes and types of boxes or whatever it is. Do some training with the site team and you're pretty much kind of ready to go. So it's a very, very fast process. Again, I can only speak for our perspective and our technology, because we focus so much on trying to make this as easy as possible for the end user so that they can very quickly then get access to the data that we collect and actually see the impact of that.

Oana Jinga: 31:30

Obviously, some other systems and solutions might take a bit longer because they might require reconfiguring of spaces or just even kind of workflows being a little bit different, because I referenced kind of picking robots. I mean, from there you have to kind of understand a little bit how that will work with the people around and who's going to maneuver the robots and obviously put goods in or whatever kind of system you're using and so on. So we don't touch on any of that because from our perspective, if no one looks at, touches the robot, then that's perfect. It means we're doing a good job and it's working in the background and it's collecting that information.

Blythe Brumleve: 32:02

So when you a typical warehouse, it sounds like they would, I say, a typical warehouse. I'm not even sure if like a typical warehouse exists, but for most, I guess, warehouses. Do you find that it only takes one Dexory robot to sort of map the entire system, or is it kind of you know a suggested amount per square footage?

Oana Jinga: 32:20

Yeah, yeah, so it depends a little bit on how often and how much data the customer wants, but on most of our sites we only have one robot. Like I was saying earlier, we do a huge amount of kind of data collection an hour. So within a couple of hours even the biggest sites you will have a full wall-to-wall kind of inventory check and data collection from the racking. If you have sites that are very fast moving and this high velocity of the goods, then you might kind of make sense to have two units so that when one is kind of charging the other ones out and vice versa. But apart from that, like even the biggest of spaces will still be able to cover at least once in 24 hours, which for most of our customers is enough, considering that currently probably takes them weeks, even months, to actually get the accuracy and visibility of data.

Blythe Brumleve: 33:03

Now we've made it 30 minutes, close to 31 minutes, into the interview and we haven't mentioned AI once, which I think is sort of the perfect time to start talking about it. I'm trying not to no, now we got to ask, now we got to talk about it. I imagine that AI machine learning. How do you define AI and its implementation with in Dexory?

Oana Jinga: 33:28

Yeah, it comes at different levels. So maybe, if I break down the technology a little bit, obviously on the robotic side there's a lot of kind of machine learning and computer vision behind the robot navigating in this massive, massive spaces with, like I'm saying, no markers on the floor, no beacons or anything like that to guide them around, just literally kind of using the racking, the ceiling and the floor, so that in itself, from a robotics perspective, there's loads of different kind of tools and resources that we're using to really kind of make that as seamless as possible. It then moves into how we're collecting data, so we take continuous pictures of the racks and then we deconstruct those and can use computer vision and again some elements of machine learning to see what we're seeing in that data. And then it moves into the digital twin and how can you actually kind of use all kinds of formulas and mathematical understanding of the space to be able to optimize that space and move forward.

Oana Jinga: 34:19

And there's quite a few other things that are coming soon that I can't like go into too much detail on. That I can't go into too much detail on, but we're bringing a lot more tools and understanding on that Because at the end of the day. I mean, we get a ridiculous amount of data from every single site because of just the nature of these kind of warehouses being very, very big, and we get that on a daily basis. So there's a lot of kind of foundation there to start building and analyzing and extracting insights and information that's useful for our customers from that data, which currently they wouldn't be able to get because they just do not have the data in the first place. So it's a very, very exciting kind of topic and area for us because, yeah, at the end of the day I mentioned like data is our bread and butter and what the product is and how you can make that more useful for for users, key to our success and next steps.

Blythe Brumleve: 35:08

Do you find that large language models? You know that everybody talks when they mention AI today, sort of the casual will typically be referring to the large language models ChatGPT, you know, and the like. Does that play a role in robotics yet, with in Dexory yet, or is that maybe coming?

Oana Jinga: 35:27

uh it does and it will continue to evolve. Um, I think, um and I mean we're not kind of playing the technical expert here because I've got people in the team that are kind of much closer to to the technical details here but I think a lot of the things that you're seeing at the moment out in the market are very much kind of focused on consumer use cases or still barely kind of scratching the surface around. You know, kind of creating memes and videos and and obviously um, fake, um, homework for for taking back to school. So I'm not saying they're obviously kind of simple technology, because what's behind it is insane, but in terms of the use case it's still a little bit kind of superficial in terms of, like, exactly what the full potential potential is. So what we're seeing over the past kind of I would say, six to eight months, is a huge, huge focus on, like, what does it actually mean for b2b use cases?

Oana Jinga: 36:13

And from our perspective, I mean, if you think about it in kind of pure essential terms, um, for us kind of the language is actually what exactly is happening inside the site and how that site is evolving and based on what. So what is influencing it? How, like what exactly kind of external factors, like you were asking me earlier, um can drive to different behavior. So it doesn't have to be language in the pure sense of literally speaking words. It can be like the language and the environment of a particular kind of customer or site, and how can you then extract more from that and then understand how we will potentially react to different types of kind of factors in the future? So I think that's where kind of the focus is moving now. When it comes to AI, in terms of like, okay, we are looking at the basics, but what does that mean in particular industries, sectors, and how can you define that?

Blythe Brumleve: 37:03

based on it. Yeah, it's definitely such a rapidly evolving landscape in a variety of different ways, but you've hit home on a few different notes that really ring true for me, and that's making data actionable. It's one thing to feed people a bunch of information, but being able to take those next steps and what does that look like for sort of the common warehouse worker that's going to be using these systems From the lens of you know we've talked about starting a company. I'm curious as to you know, what you envision for growing the company. Is it more on the software side of things?

Oana Jinga: 37:45

talking about just product and, yeah, that that's one of the biggest kind of focus areas that we have at the moment.

Oana Jinga: 37:48

Um is like what else can we Dexory doing on the digital twin platform to really kind of make that more and more useful and take it to the next level?

Oana Jinga: 37:54

Um, there's obviously a lot of different angles to growing a company. It's like from growing the team to like going to new markets to um, as obviously we talked earlier on the product side as well, expanding the product range, and for us it's been an insane kind of past 18-24 months, I would say ever since we kind of raised our seed round. Then we moved into a series A and the team has pretty much kind of quadrupled in less than a year. So there's all kinds of interesting challenges and opportunities that come from that. But we're just very grateful to actually be able to provide a lot of value to multiple customers globally and learning from them and understand, like, how the different operations they have vary. So I think a lot of the growth will come from us just being more and more established and having more insight into our market, which will then obviously feed back into the product. But it's multifaceted, I would say.

Blythe Brumleve: 38:44

Oh for sure. Yeah, definitely, there's a lot of uncontrollable situations and then controllable situations and you got to navigate both of those different paths, especially as a co-founder. And last couple of questions here as a co-founder in logistics and technology, what advice would you give to somebody else who may be thinking about, you know, either entering logistics or starting their own company, adopting technology? What would be sort of that general overview of advice?

Oana Jinga: 39:14

it goes back to where we started the conversation in terms of knowing the user and the problem. We spent a lot of time on warehouse floors shadowing people, trying to understand why they're doing certain things, asking a million questions to really kind of get our heads around their environment and their challenges and even outside of kind of the use cases that we do, to really be able to build something that then answers those elements. And as our team kind of grew, I mean we now have a kind of product team that looks after that and then so on. But obviously they're still a bit kind of newer to the sector than myself and my two co-founders were when we actually kind of tapped into it and we were there spending the time on the warehouse floor. And we're still coming back to them and say I actually don't think that makes sense because operationally this is what people are doing and all the time they're like how do you know? That's like well, because I I was like I was there and I kind of spent the time and I took my time to ask the questions.

Oana Jinga: 40:05

So that would be like the number one advice is kind of really really know your user and spend as much time as you can with them.

Oana Jinga: 40:11

And then it goes into like very practical kind of entrepreneur type of advice, like just make sure you don't follow kind of guidance from people that you don't feel like they're where you want to be in a few years time. So like always kind of follow the ones that you want to follow their footsteps as well, and other elements around like hire very, very good people. As the team is growing, you have to have some fantastic kind of individual around you to be able to support that growth, and especially as a founder, when so we're just over like well, over a hundred people now. So making sure that we have a strong leadership team that we can start delegating things to and letting some of the things go is fundamental, because otherwise you can't scale Like you can't just be. Everything has to go through one person. So there are loads, loads of points of advice there, but it all goes back to like knowing your user very well.

Blythe Brumleve: 40:59

Yeah, that's great advice that is applicable to anyone who is running a business and also learning how to delegate. That is applicable to anyone who is running a business and also learning how to delegate. That is so challenging as a founder to sort of let it go and trust that somebody else is going to care as much as you do and you know just.

Oana Jinga: 41:22

Lastly, is there anything else that you feel is important to mention that we haven't already talked about? I mean, I'm always kind of passionate about the topic, so I will bring it up. Obviously, I am a woman in tech and robotics field, and also kind of in logistics, which is becoming a lot more diverse than when I kind of started to look at it a couple of years ago. So that's fantastic and there's so many initiatives to support that. So I think, yeah, that probably is another angle kind of going back to your question earlier around like advice angle. Kind of going back to your question earlier around, like advice.

Oana Jinga: 41:47

Um, I love that. I'm kind of seeing more entrepreneurs and more kind of women, especially in the logistics sector, um trying to disrupt the market, come up with ideas to all kinds of amazing things, um, so, uh, yeah, definitely wanted to say always kind of here for anyone who wants to have a conversation about that. I'm also, like, always very open to uh speak to, to people that could be mentors for myself as well. So, very, very kind of key to myself and also to us at Exory.

Blythe Brumleve: 42:10

Yes, very, very well said. I used to joke about the fact that the line, the bathroom line at conferences and logistics used to be very short, oh yeah.

Oana Jinga: 42:20

Oh yeah, but now it's getting longer, which is nice to see. Yeah, absolutely, and I think that's probably the number one thing that you notice versus any other type of event that you go to.

Blythe Brumleve: 42:32

It was funny. At one of my first logistics conferences there was a break after a big talk and the bathroom line for the men was so long. But then on the girls' side, I just walked right in and I was like wow this is unusual.

Oana Jinga: 42:45

I was like this is nice, right?

Blythe Brumleve: 42:46

I'm like Hmm, I was like this is nice, but also maybe not, maybe, maybe we need to do something about this. So uh kudos to you for for leading that charge and being and and setting the example. Um Ohana, is there anything else, uh, where folks can maybe follow your work? Uh, sign up for a demo, withxery. All that good stuff.

Oana Jinga: 43:06

Sure, I mean, the number one kind of channel that we use is LinkedIn, so you'll see everything that we're doing over there. Definitely kind of reach out to anyone on the team or kind of drop us a message Always very happy to show people the product, regardless of what kind of work angle or sector they're coming at it from. So, yeah, just give us a follow and you'll be able to stay on top of all the different news backstory wise, appreciate it.

Blythe Brumleve: 43:28

Thank you so much. This was fascinating conversation. Thank you so much for having me Absolutely.

Blythe Brumleve: 43:38

I hope you enjoyed this episode of Everything is Logistics, a podcast for the thinkers in freight, telling the stories behind how your favorite stuff and people get from point A to B. Subscribe to the show, sign up for our newsletter and follow our socials over at everythingislogisticscom. And in addition to the podcast, I also wanted to let y'all know about another company I operate, and that's Digital Dispatch, where we help you build a better website. Build a better website. Now.

Blythe Brumleve: 44:03

A lot of the times, we hand this task of building a new website or refreshing a current one off to a co-worker's child, a neighbor down the street or a stranger around the world, where you probably spend more time explaining the freight industry than it takes to actually build the dang website.

Blythe Brumleve: 44:27

Well, that doesn't happen at Digital Dispatch. We've been building online since 2009, but we're also early adopters of AI automation and other website tactics that help your company to be a central place to pull in all of your social media posts, recruit new employees and give potential customers a glimpse into how you operate your business. Our new website builds start as low as $1,500, along with ongoing website management, maintenance and updates starting at $90 a month, plus some bonus freight marketing and sales content similar to what you hear on the podcast. You can watch a quick explainer video over on digitaldispatchio. Just check out the pricing page once you arrive and you can see how we can build your digital ecosystem on a strong foundation. Until then, I hope you enjoyed this episode. I'll see you all real soon and go Jags.

About the Author

Blythe Brumleve
Blythe Brumleve
Creative entrepreneur in freight. Founder of Digital Dispatch and host of Everything is Logistics. Co-Founder at Jax Podcasters Unite. Board member of Transportation Marketing and Sales Association. Freightwaves on-air personality. Annoying Jaguars fan. test

To read more about Blythe, check out her full bio here.