The Quiet Warehouse Problem AI Drones Can Finally Fix

In this week’s episode of the Sustainable Supply Chain Podcast, I sit down with Jackie Wu, CEO and co-founder of Corvus Robotics, to explore how AI and autonomous drones are redefining inventory management, and what that means for the future of supply chain resilience.

Most warehouses today still rely on people with barcode scanners trudging up and down aisles, a slow, error-prone, and frankly unsustainable process. Jackie’s team has built the world’s first fully autonomous drones capable of flying through massive warehouses, some the size of ten football fields, to scan inventory in hours instead of weeks.

We talk about why inventory accuracy is the invisible foundation of supply chain sustainability: if you don’t know what’s where, you can’t optimise, decarbonise, or even plan effectively. Jackie explains how real-time data from drones helps companies reduce waste, prevent stock loss, and cut emissions linked to overproduction and unnecessary transport.

We also dive into the importance of building resilience through visibility, the limits of traditional automation, and why AI-driven robotics are finally ready to move from lab demos to real-world industrial use. Jackie shares insights on the future of AI in the physical economy, vertical integration, and why autonomy, not just automation, is the next leap in supply chain efficiency.

If you’re curious about how robotics, data, and design thinking can make operations smarter, safer, and more sustainable, you won’t want to miss this one.

Listen now to learn how autonomous drones are helping warehouses take sustainability, and resilience, to new heights.

#SupplyChain #Sustainability #AI #Automation #Resilience #WarehouseInnovation #InventoryManagement #CorvusRobotics #TomRaftery #SustainableSupplyChainPodcast

 If disruption hit tomorrow, would you know where your supply chain was most exposed? In 15 minutes my free scorecard helps you assess 27 resilience statements, calculate your score, and turn the result into three priorities and a 30 day action plan. You can download the scorecard free at tomraftery.com/scorecard.

Transcript

Copy

0:03

I think the Venn diagram of you know, people who work on cutting edge AI, and people who understand how dusty, operationally heavy, warehouses are that, you know, the overlap of that Venn diagram is not, not that, not as huge as you would think. Good morning, good afternoon, or good evening, wherever you are in the world. Welcome to episode 89 of the Sustainable Supply Chain Podcast, the only podcast focused exclusively on the intersection of sustainability and supply chain. I'm your host, Tom Raftery, and I'm delighted to have you here today. Before we get started, a quick reminder. You can now support the podcast and unlock the full back catalog by becoming a Sustainable Supply Chain Plus subscriber. For just five euros or dollars a month, less than the price of a fancy coffee, you'll get access to all 80 plus past episodes of this podcast and more than 380 episodes of the Digital Supply Chain Podcast. The most recent four episodes remain free, but the archive is exclusively for subscribers. Plus members get a shout out on the show and a direct line to me for suggesting topics and even shaping where we take the podcast. Next, you'll find the link in the show notes or at tinyurl.com/ssc pod. Now onto today's conversation, and it's one that quite literally takes flight. We're heading inside the warehouse to explore how AI powered autonomous drones are transforming inventory management, and in doing so, building smarter, leaner, and more resilient supply chains. My guest is Jackie Wu, CEO, and co-founder of Corvus Robotics, a company creating the world's first fully autonomous drones for warehouse inventory scanning. Jackie and his team are tackling one of the quietest yet most costly vulnerabilities in global supply chains, inventory accuracy. Instead of workers spending weeks walking the aisles with barcode scanners or clipboards, corvus, drones can scan an entire facility, sometimes the size of 10 football fields in hours, delivering near realtime visibility and control. We talk about how this technology strengthens resilience by reducing waste, uncovering hidden inefficiencies, and turning warehouse data into actionable insight. We also discuss what happens when AI automation and sustainability collide not as buzzwords, but as the backbone of a more adaptive, efficient and circular supply chain. If you are interested in how visibility, precision, and resilience come together to future proof operations, this episode is for you. Jackie, welcome to the podcast. Would you like to introduce yourself? Yeah. Thanks Tom. Hey, super excited to be here. Super excited to be spending time with the audience. Yeah, I'm Jackie Wu, CEO and co-founder of Corvus Robotics. We are an AI and robotics company based in Mountain View, California. And we build the world's first and only fully autonomous drones that help warehouses, scan inventory. So if you think about world's supply chain for food, for ag, for everything else. You know, kind of every commercial object that sat around you, it sat in a warehouse at one point. You know, your, your laptop, the building construction materials, your phone, your watch, whatever, clothing, right? And these companies in, in the physical economy all have to do inventory. And the way that every company around the world does inventory right now is pretty much, you know, a person with a handheld barcode scanner going around with a clipboard, checking off items. And it's slow and laborious, and it takes them, you know, months to do this. So our team over here we have built an autonomous flying robot using AI and lots of sensors in order to fly around these massive warehouses and check inventory for you. And it does this, you know, substantially faster than people. We can get turns, you know, every couple days instead of once a year or something like that. And, and you kind of have to imagine these warehouses being these massive facilities, the size of 10 football fields going up four stories high. So, so, you know, it's no trivial task for a team of human inventory checkers to go and do it. Whereas the robot can do it and doesn't get tired and you know, it's, it's a job that people don't particularly want to do. That's a little bit about us and what we've been building as our first product. And you know, happy to dive into the rest of the show. Superb. It's funny when people mention warehouses and the size of warehouses to me, I'm often reminded of the last scene in the movie Raiders of the Lost Arc, where, you see this pallet being wheeled into this warehouse, and then the camera backs back and back and back, and you see just thousands and thousands and thousands and thousands more of identical pallets in this warehouse that just goes on forever. So yeah, I know exactly what you're talking about, but. Yeah. let's take a step back, Jackie, and let's talk to me a little bit first about the origin story of Corvus. You know, how did you wake up one morning and decide to kick this off? Yeah, I mean especially apropo'cause you're wearing, I think your Indiana Jones hat. Yeah, I was in grad school at Northwestern in Chicago for robotics and kind of a warehouse nerd in that I've been to warehouses in 10 countries. It's like a stop that I do sometimes when I'm abroad. And what I realised was that these warehouses all around the world, in Asia, in Europe in South America, in North America, all basically look the same. These warehouses, you know, to your point, they're just endless racks just going on stretching as far as you could see. And it's just, you know, it's, it's pallets, it's cases, inventory coming in and out for all different types of goods, all different types of industries. And pretty much all of them, you know, something like 97, 98% of them all do check their inventory in this very manual way, like around the world. It's just people with barcode scanners. And so, you know, in grad school, the, the thought was. Why can't we do it better with the latest wave of AI and robotics coming out? There's gotta be something that we can do to improve this process by 10 times in order of magnitude, if not a hundred times. Why were you a warehouse nerd? How come you'd visited warehouses in all these countries? What's that? What's that all about? I think early on, you know, one of the, as a kid or as a teenager, I think what motivated me was thinking about like, what is my life about and what is the biggest kind of impact I can make, right? And one of the things that really resonated with me was, the physical economy. And you know, we're kind of seeing it now with some of the political events going on. But as a Chinese American right, I've seen the ancestry going back to see my grandparents. It's like China has developed materially very quickly over the past few decades. And in doing so, it was a lot of the physical economy, which is how every country has industrialised. Like people need stuff like, right, like lifesaving medicines, food, perishables, household appliances, all your newest gadgets and gizmos. People like forever since the dawn of time and up until now people need, you know, physical items and all these items need to be tracked. So this is like a, sort of like a core part of the world economy from the most far-flung island nations to the smallest European, you know, micro states. Everybody has, every country has warehouses and it's the backbone of, you know, the supply chain is the backbone of the world. And so. When I was thinking about, you know, in grad school, you know, how can I apply the stuff that I'm learning to kind of help update the economy, right? it's like this is clearly something that is not going away. That is, that will always be here and it will be supply chain is always going to be here. And checking inventory is, is always going to be a function. And I think one of the insights that I had early on in talking to a lot of these operations warehouse, supply chain managers is that inventory accuracy, like everything flows downstream of inventory accuracy. You kind of have to know what is where in order to operate a warehouse, right? To know what to pick, what to put away, what to replenish and so on. And so, you know, it was kind of, that starting point that was like, okay, if this is really important, maybe this is something that we can address, right? And it's a perennial need. And how big a blind spot is inventory accuracy right now? I mean, do you think it's something that's overlooked or is it something that's really important? Yeah, I think the Venn diagram of you know, people who work on cutting edge AI, and people who understand how dusty, operationally heavy, warehouses are that, you know, the overlap of that Venn diagram is not, not that, not as huge as you would think. Yeah. You know, usually you're either a robotics nerd or a supply chain and operations nerd and not both. And I, I think inventory is one of those things. Yeah. You know, maybe now with some of the newest sort of focuses in the market on re industrializing and, and kind of looking at Hard Tech, it's a little bit more popular. But when we started, I mean know, going through Y Combinator in 2018, we were the only company focused on this space. You know, and, and no, no one people were like warehouses, like, but you know what the heck? Right? We could be building a. B2B SaaS app or something. So I think it's, it's just, it's, it's so core and so prime to all of the physical things around us, but we don't think about it. None of us really think about all of the sort of frontline workers, essential workers and all the warehouses keeping them operational day to day,

starting their day at 6: 10:22

00 AM that get us all the stuff that we have. Sure, and I mean, manual inventory counting as you said, that guy with the scanner and scanner. Yeah. Yeah. why is that still the norm for a lot of warehouses in 2025? You asked me, Tom, I ask who you know, I, I, I, I'm reminded of these like old Egyptian cuneiform tablets, some of the oldest human texts, and it's like, you know, they finally decipher them, and they're inventory lists. It's like what has been in my grainery, right? Like, I have this many stockpiles of whatever. That's basically where we are, you know, in a couple thousand years. The latest innovation has been a handheld barcode scanner and barcodes. And so, the vast majority of warehouses are these very manual type of places where they're also struggling with, at least here in the US with sort of labor costs in the labor pool. Most people don't wanna work in warehouses. This is not like thrilling, exciting jobs. Checking inventory as your full-time job going up and down aisles. To your point, those thousands and thousands of pallets, they all look the same. Every aisle looks the same. You're supposed to find one among, you know, needle among the haystack. It's it's not a fun job. And so I think one of the promises of AI and robotics is to hopefully right, free us from the drudgery of work that's just very dull and boring, dull, dirty, and dangerous. You know, you, when you think about warehouses that go up, four stories tall 40 feet in the air people are, people can't fly. So they're taking these scissor lifts and strapping themselves on with a safety harness going up there and it's wobbly. And you're hoping it's not stuck. And you know, it's, I don't know, like I got a fear of heights, man. Like I wouldn't wanna do it. And how did you settle on drones? Because I've had another company on the podcast before who's doing a similar thing, but they're doing it with these robots that have a very long vertical arm and multiple cameras on the arm going up so they can scan multiple rows of a, a warehouse at the same time. I think that's a good point. You know, when we started, we looked at, we were solution agnostic, so we were looking at different ways to solve this problem, right? Naively we thought, what if we put a ton of cameras all over the warehouse. And it turns out that, you know, if your facility is the size of 10 football fields, that's a lot of cameras. That's a lot of wires, it's a lot of maintenance and you can't really do that. So then we thought, you know, sort of what about a ground robot with this sort of mechanical giraffe of a, of a neck with a bunch of cameras on it, to your point, right. And I think that is a valid solution. I think it's actually a very good solution. But it occupies a slightly different niche for our customers, in terms of price point and in terms of operational complexity. If that big thing is in an aisle that is blocking the aisle, no one is going in the aisle. And the cost is about an order of magnitude more than the drones. So sort of, you know, if you think of like a two by two matrix of sort of price and performance, you know, you, you see that type of item, you know, somewhere over here. If you get more expensive, you see maybe like an ASRS system even more expensive. You think of sort of those fully three D-A-S-R-S systems like, you know, Exotech or, you know, they look like a beehive of, of robots, right? It's just orders of magnitude of increased complexity and increased sort of performance. But one of our insights was the cycle counting team for checking inventory, even though it's really important. It, it occupies a very small percentage of the warehouse's, total budget and labor budget. Labor force, right? So if you've got a hundred people in a warehouse, the majority of those people are picking, you know, and and receiving. And they're doing things that actually move the inventory. You usually only have a handful of folks checking inventory. And so you kinda have to justify the ROI within that budget 'cause that's what people are going to be benchmarked against, right? So there's, there's other ROI beyond the labor, but that's what people are gonna be thinking about. When we were starting Corvus it was thinking what is the smallest, cheapest, most, economical, least disruptive type of form factor in robotics today. That can still go up 40 feet in the air. And that happened to be the latest research coming out on micro aerial vehicles or, or drones today. So, you know, these are, these are small, you know, it's, it fits on the palm of your hand. It's a couple pounds a couple kilograms. And it easily goes up as high as you want it to. 40 feet, 60 feet.'cause of, well, it flies, right? So you know, it's a, it's a very small form factor that doesn't block the aisle. So it kind of occupies our own little niche that's kind of tailored to the realities of the warehouse, the the warehouse's budget and ROI constraints. And if I have a massive 10 football field, warehouse, to your point, do I need a fleet of drones? Will one drone do? Is it somewhere in the middle? Is it an it depends answer? Yeah, I, I mean, it totally depends you know, on, on the frequency that you want the counts. So you, you gotta think, these are like big companies. You know, some of the biggest companies in automotive, aerospace, consumer packaged goods, healthcare, you know, whatever, right? So they have these like mandates internally from some kind of finance team of how frequently they have to do this. And so, you know, if it's more frequent, so you could kind of picture this massive facility with a, with a lot of pallets, having a small fleet of drones, you know, less than a dozen kind of flying by, you know, kind of waking up, doing their morning duty, doing the afternoon duty, flying overnight when no one's in the facility and all working in unison to make sure that the inventory of that facility across many different aisles are, you know, at the same time being updated with inventory checks. Now we've also had some very small facilities. We joke that smallest unit that you can place is one drone. You cannot order half of a drone. I cannot chop it in half and deliver it to you, right? So, you know, for some of the small, smaller facilities, that's something that we do qualify. We gotta make sure you know that either the, the inventory is high value enough to justify, having a drone or that you, you need, it counted very frequently, you know, for sustainability or for expiration type of reasons. And so you know, it's quite flexible in that the lowest point is, is, you know, can be quite, quite lower than a lot of traditional automation, which involves a lot of, you know, metal and steel and kind of costs there associated. But also we could flexibly increase to, you know, a fleet of vehicles. And do they talk to each other, peer to peer to share information and avoid collisions and things like that? And, and as well, do they then share the data they've got, I assume they have to share it back to an ERP system or a WMS or something like that. How does all that work? Yeah, so the, the, the drones have a lot of software and AI on them that talks to each other and to the landing pads. The landing pads are actually sort of, mini supercomputers that have a lot of processing and data storage on them because, you know, the drones are collecting a lot of pictures and videos and sort of 3D you know, environmental models like a digital twin of the facility. And so that takes storage. And on the backend, all of this syncs with the local WMS warehouse management software, which is usually a branch of, or, you know, attached to the ERP. Right? So SAP has their EWM or, know, you know, some folks are using separate ERPs in Manhattan or something, and even some of the more niche wms, homegrown WMSs, we've integrated.'cause at the end of the day, the integration is quite simple. It's just, what inventory is where, when? Right, so it's, it's, it's just a couple fields. But what's interesting is much more of the rich data that we collect just hasn't been possible to collect before. Before when it was, you know, a guy with their cuneiform tablet, sorry you know, notepad and barcode scanner that's what they're writing down. But now there's all this sort of pictures, videos, historical pictures. Looking back a month ago, something moved here, computer vision and machine learning analytics to look at occupancy and count and tie high and all these other kind of things. It, it's a very interesting age that we are in now with computer vision and machine learning and being able to analyse all this data. Like this wouldn't have been possible a decade ago or even five years ago, I think. I gotta ask, what's the most surprising thing any of your clients have found when they've gone shooting drones around the their warehouses? I think for one of our, you know, customers can't say, you know, can't say who, but more heavy industrial. So, you know, some of these pallets are worth a lot of money. You know, easily six figures. On the first day, on one of the first flights, we found a pallet that, you know, was, was high up. So, you know, people hadn't seen it and they were like, oh yeah, that is worth, you know, six figures. You know, X, X, X, X, x, X. And we just reordered that a few weeks ago 'cause we couldn't find it. And it's like first of all there's your ROI thank you very much. Second of all, like, yeah, you know, if we were flying this every single day, you would be able to find, you know, all of the stuff substantially faster. It's a, it's a, it's a thing, man. Like, you know, we, we talk about sustainability, you know, supply chain resilience, but, at the core of it, you, you kind of have to know what is actually in your supply chain and how it's moving. And the previous data collection methods was very much garbage and garbage out. I gotta think as well, if you are flying through warehouses, you're identifying probably not just what's in the rows, but also where there's nothing. So that should mean that you can fill the warehouses even more. That was possible, possible before. You gotta be able to optimise the occupancy of the warehouse, no? Yeah, absolutely. That's one. I mean, you are entirely right Tom. That is one among some of the very interesting sort of analytics things that are kind of starting to emerge as we're collecting more and more of this data across, you know, different companies, kind of, you know, American footprints, right? If we're in multiple facilities or across their entire footprint, it's like, you know, we can, we can optimise what is in that facility and how do you store it more efficiently, right? These are your items that are picked the fastest. Maybe we should have 'em on the ground level instead of, you know, 30 feet in the air. And then also like, how do we optimise cross facility kind of shipments? You know, there's sort of, a lot of folks have dedicated software and rules for this, but we could tell you exactly on the ground, like how many pallets of this you have today and how many pallets are in the, the other regional warehouse. And oh, you've got a big demand coming up from sort of historical analysis. This is what we gotta do. So there's a lot of interesting things that are coming to play now that we have this sort of early critical mass of data that we can play with. And that kind of leads me to think that sometime in the future for the supply chain, it is going to be a lot more AI driven, a lot more data driven in a way that just wasn't possible before. You, you're kind of seeing that with, you know, with AI and LLMs of course, but all of it is math and data science with a new name and maybe more data. But I think a lot of the world's physical economy will become more and more optimised through AI. And we want to be a key piece in collecting that data. And does this then allow for the likes of real time inventory knowledge because you talked about before people doing inventory once a year and now presumably if they have drones flying around or even a drone flying around, if it's a small facility, they should have pretty much up to the hour, up to the day at least, inventory lists. Yeah, basically, I mean, there's a little bit of nuance there in that I would say that you don't want to over collect inventory data, so you know, you wouldn't want to scan the same location a thousand times if it hasn't moved. Right. If, if it's, you know, if you have, you know, if your warehouse has on average, you know, a couple hundred moves or something, and you're scanning a hundred thousand locations a day, then it's like, what are the other, you know, 999,000, you know, scans for, right. You're, you, it's, this is, this is an overbuilt system. But you know, to to your point, it's like now we have the inventory visibility and the sort of expediency of, of that so that we could find root causes. Before, if you were scanning things once every six months, once every four months. That means if a problem happened on average, you know, it, it happened three months ago. Like what? You know, I, I have no idea where that case of, you know, what, whatever liquor or something left. But now you know, with AI and, corvus drones, you can say, Hey,

here's pictures of yesterday at 7: 24:01

00 AM when something changed, right? And, and we can kind of root cause and do that kind of inventory research a lot better, which is very, which is, you know, it, it's a shift in how the world economy will move. My hope is that the world economy like that, we, we will contribute to positively to economic growth, which would be amazing. And in an age where a lot of startups are outsourcing their hardware, what's the case for owning the full stack? Because you do it all yourselves, right? Yeah, I mean, we, we build in-house and we've done that you know, since before it was politically cool. So that's our little pat on, on, on the back. You know, we, we build here in California. It was actually out of necessity rather than you know, sort of how do you do this, right? Is the question that we asked ourselves early on. Because, you know, initially we tried every drone out there. There's this sort of naive view that like a drone is a drone. That's why people who like, don't know anything about the drone industry, right? It's like, yes. You know, DJI makes great consumer hobbyist drones for when you go skiing, the drone will follow you and take amazing footage that you, you could, you show, show your dog or whatever. And you know, when we started we tried every off the shelf drone out there, whether it was DJI, Skydio, there's this French company, Parrot. You know, this is South Korean company Unify. We've tried all these drones. A lot of them were amazing, but for their specific use case, right? Whether drone light shows or this kind of ham camera photography stuff and there's just nothing out there that would actually satisfy what the customer wanted for autonomous warehouse inventory, drones. So we had to build our own. And building our own meant that we would control our own AI stack on the hardware and software. And this is key to delivering actual autonomy in a space where, you know, autonomy is really hard. You could think of the Corvus drones as essentially a self-driving car that fits on the palm of your hand. That flies, right? And, and so we kind of have to build that from the ground up from the sensors, the compute. You just sort of calibrating all of it with the software and the AI in order to get this all to work. And that's been a, that's been a large endeavor. So that's been kind of key to our success and initial scale. I would say in having the, the years of, of painful work, building it ourselves. Yeah. Sure, And I. Obviously building it yourselves. the drones have inbuilt sensors, not just for photographing the shelves, but also for collision avoidance. They're fully autonomous, so I assume it's wireless charging. When they go back to the drone pad. I assume you can auto schedule them as well so that they go out maybe at nighttime when there's nobody there. Or if it's, if it's 24 hour one, they probably know the aisles that are being used at any particular point, in any particular day, avoid them and go for the other ones, or, you know, talk me through all that. Yeah, you're, you're totally right, Tom. I, I, I think you have a product mind and that, and that's very nice. And, and I mean, you are absolutely right in thinking about, you know, what would the customer actually want for a, you know, if you, if you're a warehouse manager, what do you want this thing to do? You want it to be pretty much hands free. Right. If you were paying your warehouse workers a certain dollar per hour, you, you don't want them to also be managing a drone. That's not the point of you hiring this robot. Right? And so you want it to be fully autonomous you know, fully recharging, intelligently planning. No one should even have to hit, go on a scheduler. It just auto schedules. Or you could request it to schedule, you know, if you want to check a particular aisle that day, or particular location or whatever, right? Collision avoidance is essential. You need to be able to, so we've got 12 cameras all around it to see everything around it. In, in every sort of, and collusion avoidance software. And so I, you know, these are kind of all table stakes at this point. But you know, in, in the drone space, in this drone niche that I talked about, sort of from micro aerial vehicles, we are currently, as far as I know, the only player in the world that that that does this for, you know, it's, it's fully autonomous and there's nobody involved and there's no sort of location stickers or beacons or whatever. Which is, which is pretty interesting. And kind of that base is the foundation, you know, that technology is the foundation for a lot of the you know, some of these new product lines that we've been launching which we'll talk about in, you know, perhaps a future conversation. Yeah. So, you know, there's, there's a lot of work behind the scenes to get it all work on the AI side, but the payoff is massive. If we can build an actual autonomous system inside of, you know, the, the world's busiest industrial environments. And are there any big sustainability wins that you can talk about? I mean, I, I know we talked about that one pallet that that one customer found, but any other big sustainability wins. yeah so I will, you know, limit to what is public so don't get myself in trouble. You know, but for some of our food and ag customers and medicine, pharmaceutical customers like these guys, they have products that have expiration codes. You know, this is not like a chair, which you can, you know, kind of sit in the warehouse for, you know, five years and, you know, it's still a chair. You know, it might be outta style, which is a different thing. But coming up to an expiration coming up to a month before the expire Expiration date, you know, you, you can soon cannot sell that item. And, and that impacts people's health, right? And, and so these are something that that is pretty important for sustainability and reducing waste reducing just the massive amount of you know, things that go into landfill. Okay. I mean, if you cannot sell it. That is trashed. Right? And, and so that's a monetary loss, but also a loss for sustainability that, you know, could have been, we could have done that more efficiently. And so I think it's, it's very good that much of the industry is looking at a lot of these sustainability initiatives both from, you know, reducing material handling emissions, you know, this kind of stuff. I mean, the drones are electric, right? It's, you know, it's green energy in that sense, but also in terms of the amount of loss and wasted products. So we're happy to make our contribution there. And how are things things like shifting tariffs and sourcing strategies, influencing inventory decisions right now. Oh my gosh. Yeah. The, the tariffs. You know, I, I think for a lot of companies it is some uncertainty. In the robotic space you know, many of our vendors, you know, key vendors are American, but some are overseas because there's certain components that are only made overseas and component parts and just there's no alternative. So over time we hope to build more and more of the sort of the, upstream components ourselves which is kind of an, an interesting technique to kind of vertically integrate. But ultimately you look at, you know, these large companies Skydio for example, or you know, this, this large American company earlier this year, you know, and, and they make these amazing, you know, their autonomy is incredible. But they faced supply chain challenges from Asia earlier this year. Like they could not get their batteries, they could not ship any units. And, and so kind of similar for some of the other, you know, robotic manufacturers for, you know, whatever it is, actuators to fly controllers or something like that. And so it is a question of, you know, the tariffs and, and the supply chain instability. Uncertainty always is, is there, but it's a form of, you know, strength, building up strength and being able to, build all this in-house and kind of vertically integrate more, more and more of that stack. Where do you see the biggest opportunity for AI and automation in the next five, 10 years? And flip side of that, where do you think the hype is outpacing reality? Tom, I think that's a super interesting question because you know, you're seeing these, these massive investments by all these large tech companies. NVIDIA is the world's first$4 trillion company, right? Like that's massive, right? 10 years ago, we wouldn't have thought it was possible, and you're seeing these massive bets, and so far most of them have been in software. Right. They've been in LLMs, they've been in, you know, for sort of the, the software, you know, for specific verticals that build on top of that and, to support this and that. But you're starting to see a lot of this investment in hardware too. And there is a particular approach for the building of hardware with, you know, the vertically integrated AI stack and data sort of feedback loop of data capture and iteration that companies that have not just the smartest engineers, but also sort of the ability to collect that data are, are going to win out massively. Like it is going to be a power law type of return where like it. It, it it'll be night and day, basically a step function in, in, in night and day in being able to have these hardware devices out actually collecting data versus sort of an R&D project with, you know, cool reinforcement learning in the lab. And, and I think that's where some of the hype is, you know, respectfully, you know, but, some of these companies that have raised a lot of money but their products are in the lab where they're, you know, they're not quite doing anything useful. So some of these companies remind me a little of the autonomy, you know, self-driving space, you know, maybe close to a decade ago now. There's not going to be that many self-driving winners, just like they're not as many as are being massively funded. You know, same as for some of these humanoid companies. Sure, What are your plans for Corvus then for the next five, 10 years? Ooh you know, we are in a good position and lucky position to be in a position of financial strength and business strength and stability, which is very nice. We can, choose how we want to, you know, what kind of paths we want to take. but there's some very ambitious things that we see with the core technology that we've built for autonomous warehouse inventory scanning drones. On the AI and computer vision, machine learning hardware, a lot of the key components that we've been building for, the past several years are kind of all coming into place. And, and, and so there's some sort of bigger plans around sort of in, in this wider space where we think we can make a unique contribution to the world. And for, for supply chain, for making, you know, the physical economy more efficient. I, I, I think that is the next like biggest opportunity in the world, which is fascinating to be working on. Interesting. Interesting. Have you ever had a warehouse worker give your drone a nickname? Yeah, we've you know, that's always a, a fun one. I think one of the facilities that had two, they called them Bonnie and Clyde. You know, some of the other, you know, after like Disney characters or Disney princesses, like we joke about like, you know, maybe one of these days we're gonna put some googly eyes on, on the drones and you know, let's make them a lot more fun. I think, I think it's an opportunity to delight customers and to, to make people feel happier. So, you know, we try to have fun with, with folks whenever we can. Yeah, of course. Yeah. I, we have a, robot to clean the pool and we call it Wally. So, you know, those kind of things. They're always good. Yeah. A left field question for you, Jackie, if you could have any person or character, alive or dead, real or fictional, as a champion for drones in warehouses, who would it be and why? Hmm. Probably Lee Kuan Yew the sort of founder of Singapore. There's a lot of mythos there, but I think know, there's some very interesting ways in which he thought about, building. I think there's some analogs there to, start up land, right? You know, here I am running a humble business, but I think there's some analogs there in, in state building and how to sort of optimise the economy for the wellbeing of, of its people that I think would be very interesting to learn from. Nice. Okay. Very good. We're coming towards the end of the podcast now, Jackie, is there any question that I didn't ask that you wish I did or any aspect of this we haven't touched on that you think it's important for people to think about? I think you've largely touched on it on, on, you know, everything that I wanted to, I think. Some of the manufacturing pieces of, of the importance of building in-house, sort of the future of AI in the physical economy and, and the importance there. And clearly sort of the importance of inventory accuracy and warehouses in a way that provides ROI for customers. I mean that's, I mean, this is all stuff that I think about, everyday. So, you know, really glad to be here and I appreciate your time, Tom. Fantastic. Jackie, if people would like to know more about yourself or any of the things we discussed in the podcast today, where would you have me direct them? They should go to corvus robotics.com C-O-R-V-U-S robotics.com. There's live videos. You can see the drones fly autonomously. Read some of the case studies, everything's on there, and reach out to our team and we'll take care of you. Fantastic. Jackie, thanks a million for coming on the podcast today. It's been really fascinating. Thank you, Tom. Appreciate it. Okay. Thank you all for tuning into this episode of the Sustainable Supply Chain Podcast with me, Tom Raftery. Each week, thousands of supply chain professionals listen to this show. If you or your organization want to connect with this dedicated audience, consider becoming a sponsor. You can opt for exclusive episode branding where you choose the guests or a personalized 30 second ad roll. It's a unique opportunity to reach industry experts and influencers. For more details, hit me up on Twitter or LinkedIn, or drop me an email to tomraftery at outlook. com. Together, let's shape the future of sustainable supply chains. Thanks. Catch you all next time.