Why Your Delivery Robot Still Needs A Mailbox

We sit down with Rithiker Vijay, CEO and founder of Autonomy, to talk about what finally makes delivery robots scale in the real world. The big unlock is not just better navigation, but smart infrastructure that removes the human-in-the-loop handoff and turns pilots into repeatable ROI. 

• why the next decade belongs to physical AI and robotics 
• Nvidia’s ecosystem and why partners matter even for startups 
• Rithiker’s path from college robotics to BMW urban autopilot work 
• why warehousing robots scaled faster than public delivery robots 
• the missing infrastructure layer that public robots need 
• how Arrive Points plus robots create seamless end-to-end logistics 
• hospital deployments and why freeing nurse time beats “labor savings” alone 
• humanoids as the future and why timelines can mislead 
• context-aware autonomy using edge AI and vision-language models 
• rapid proof of value in a large pharma facility workflow 
• autonomous deliveries in remote mining communities in Australia 

Transcript

Welcome And Partner Spotlight

SPEAKER_00 0:04

Hi everybody, welcome to another dance show.

SPEAKER_02 0:07

Hey, how are you?

SPEAKER_00 0:09

I'm good. How are you doing?

SPEAKER_02 0:10

Really, good.

SPEAKER_00 0:11

We have another very special guest today.

SPEAKER_03 0:14

Is it me?

SPEAKER_00 0:14

It's always you. It's always you. You're the number one family.

SPEAKER_03 0:19

I thought so. I thought so.

SPEAKER_00 0:21

Um, no, it's another partner of ours. Um, clearly, without our partners, our network doesn't exist, right? So they're crucial elements. And this is a really important one that was a part of our first successful deployment that we've had.

SPEAKER_02 0:37

And a big part of the future, too.

SPEAKER_00 0:38

Huge part of the future.

Meeting Autonomy CEO Rithiker Vijay

SPEAKER_02 0:40

Who is this guy?

SPEAKER_00 0:40

This is Rithiker Vijay. He is the CEO of the U.S.

SPEAKER_02 0:43

Let's bring him in.

SPEAKER_00 0:47

Um joining us from very far away. It's a pleasure to have you on the Dan show today.

SPEAKER_01 0:53

It's pleasure pleasure, it's all mine because uh, you know, while you always want to be on a David Letterman show, but uh, you know, there you are the celebrity. But uh, you know, on the Dan show, Dan is the celebrity. So hi Dan, you know, nice to have us.

SPEAKER_03 1:07

Rithiker, so happy to have you, man. Love you, buddy. Get welcome, man. Well, what are we doing? Tell us what Kylie, where are we going?

SPEAKER_00 1:15

Yeah, but no, Rithiker uh joined us from India, and um it has been so fun to get to know you and to see how our processes align. We spent a lot of time with Rithiker in Indian at the India AMA Impact Summit. Um and Rithiker, there was so much momentum gained while we were there, and it was just thrilling to see um the whole network together, the drone, the robot, and the arrive point working together and the enthusiasm that surrounded it. So being there, um, you know, with Kieran Paul, our strategist there on the ground in India, and then VP of sales Robert Rex, also on Keith from um SkyAir, from your perspective, what type of momentum did you feel there? Um, what was it like?

Physical AI Needs Real Infrastructure

SPEAKER_01 1:59

So I think uh, you know, it's it's it's a very interesting time in which we are. So, you know, last two, two and a half years have been, you know, really rewarding for all the AI enthusiasts. And I think next 10 years are going to be, you know, for physical AI. And that's what is you know rooting at this point in time, where uh, you know, drones, robots are everywhere. But one of the most important things which uh, you know, as uh you know uh tech folks or you know, you say roboticists, is that we miss out on the infrastructure element, and that's that's something where this partnership is uh it's it's a bliss, you know, because we are seeding it for the future, and that's what I see. You know, there's a huge amount of momentum which we can drive forward.

SPEAKER_03 2:39

You know what I love is that video that you showed me where autonomy is on the stage. You're up Vincent, Vincent, or I'm sorry.

SPEAKER_00 2:48

Jensen?

SPEAKER_03 2:49

Jensen Jensen, I'm thinking of somebody else. Sorry. Jensen Wong is up there, autonomy, you, you know, uh, yeah, it's tell us about that. That's that's huge, rare error, man.

SPEAKER_01 3:01

So Nvidia has been a great partner right from the you know, first year of our our inception. We have been part of their inception program. And uh, while you know uh people often look at Nvidia as a you know semiconductor company, you know, five years back, but uh at this point in time it is uh it is very, very important you know, partner for any future technology which is coming in. So uh Jensen actually came in uh you know uh pretty much uh you know two years back in India, and uh there we had a lot of folks uh who were you know uh in the whole summit. So Mukesh Shambani, who is the you know, richest man in uh uh in Southeast Asia and you know even top five globally. So uh you know suddenly Jensen you know put a slide on a big stage and uh robotics partner, they were only two, three companies, and autonomy was one of them. So I think uh it's it's a you know uh I think it's very important to understand that no matter you know whether there's a trillion dollar company, but they value their partners so much that uh even a startup matters to them. So really value the relationship which we have with Nvidia and uh we're loving it. And I think that's just that's something which uh which took us to the AI summit where we were part of the Nvidia you know uh uh booth and uh we we showcased what uh multimodal you know autonomous systems look like.

SPEAKER_03 4:19

Hey, hey, you're you've been around this industry.

From Robotics Competitions To BMW

SPEAKER_03 4:21

Tell us about your history, how you got into this industry, and and then we'll then we'll go from there.

SPEAKER_01 4:27

So, you know, my journey started from you know college robotics competitions, and uh, you know, it's it's always fascinating to be part of something which is uh not regular curriculum. So uh, you know, uh three, four years spent like a tremendous amount of time in robotics, then got into uh unmanned ground vehicles. Uh, you know, that was my first job uh for defense applications, then forwarded into warehousing automation, so where mobile robots were introduced back in like 2014, 2015. And then, you know, towards 2016, uh there was a huge you know rally around autonomous cars, and that's where you know I got in. So, you know, I was I moved to Germany. I I lived there for for some time, and there I was leading BMW's urban autopilots uh while working in a company called Aptive, which was a L4 autonomous cars company. So I think uh, you know, what we have seen over the period of time is uh autonomy has different flavors, like you know, defense, warehousing, you know, uh on on road vehicles. But we saw an opportunity which was like uh, and uh, you know, there was a white space between indoor and outdoor logistics because there's so many robots on the warehousing side. There are on-road vehicles or even delivery robotics companies, which are pretty much outdoors only. So we picked up a sweet spot that we can, you know, uh what we can solve is the indoor-outdoor aspect. And uh, you know, fast forward, you know, at this point in time, we are so happy that uh our wonderful team has got us into you know airports, you know, that that's what that was the starting point. But uh, you know, that led us to hospitals, last mile, uh, and also, you know, industrial facilities. So uh so great to have a team. And uh every part of the journey,

Why Public Robot Adoption Stalls

SPEAKER_01 6:08

I think sometimes we don't realize, but uh, you know, that is contributing to where we are today.

SPEAKER_00 6:13

So you've been in this for 10 years. Um what have you seen as far as the growth in this industry? And from 10 years ago to today, where do you see the momentum today and how it's just becoming much more mainstream?

SPEAKER_01 6:28

I think uh, you know, uh last 10 years we have seen enormous amount of uh growth in warehousing robotics. And uh one of the major roles of uh you know uh scaling such a technology was because it was a controlled environment, the infrastructure was not a problem at all because uh everything is in control pretty much, right? Now, fast forward, uh the robots are not uh you know in the warehouse anymore, they're out in public, and uh you can you can see drone deliveries, you can see robot deliveries happening across. Uh, but uh what is missing? Because uh, you know, uh the introduction of the systems have been there since last couple of years, but the massive adoption has not taken off like the warehouse, uh, you know, warehousing robotics have done so far. So I think the missing piece is infrastructure. And uh we see that uh you know the infrastructure which is made in today's world, uh, you know, it is not meant for, you know, it is meant for people, it's not meant for robots. So something needs to be changed, and that's why, you know, uh uh any key uh to you know adoption of uh you know public robots or robots in a public domain, infrastructure will play a very, very key role.

Removing The Human From Handoffs

SPEAKER_03 7:38

So our our yeah, please go ahead I was gonna say talk about the differentiation you're seeing when you when you introduce an arrive point and a robot together and what that's doing in the sales cycle and how you've now actually implemented um having arrive AI in every presentation you guys are doing, right?

SPEAKER_01 7:56

Yeah, I think uh it becomes super important for uh for folks like us, like you know, because uh when we are putting up these robots, there is still a human element uh involved, you know, to to uh place the packages inside the robot, to pick up the packages inside the robot. And uh, you know, on the you know, uh as such, it's it's it's it's still efficient. But uh why the real adoption is still you know yet to be seen. There's no robotics company or no delivery robotics company, I'll say, which has put like 10,000, 20,000 robots still dead. So really uh the the area has not evolved such. And the reason is uh you know, uh the human in the loop. So what if the robots are down there and you know they're waiting for a person to come and pick up the delivery? I think that's when you know, when uh we met Arrive like a couple of years back, and I still remember that was National Retail Federation event in New York. So, you know, NRF uh arrive and autonomy was just neighbors to each other uh when we were exhibiting. And uh, you know, the thought just occurred that hey, what are we doing? I mean, why shouldn't why shouldn't we work together? Because we can really deliver packages uh into the mailboxes, and those mailboxes can be the you know uh pivotal point for end-to-end logistics. So, you know, that's how the journey started here.

SPEAKER_00 9:14

Tell us more about where you are deployed, specifically in the healthcare campus arena and the successes that you've had, and then how the arrive point can help scale those successes.

Healthcare Robots That Save Nurse Time

SPEAKER_01 9:28

Sure. So I think uh one of the wonderful you know deployment which we often talk about is uh you know uh Hancock Hospitals, which is like which is where we have both arrive points and uh you know Autobots uh which are deployed together, uh having an end-to-end seamless delivery system. Uh but apart from that, uh within healthcare, we have been working very closely with Wellstar hospitals or Wellstar Healthcare. And uh there we, you know, Wellstar is also an investor in autonomy. So that uh double verifies what they look into, you know, when they see robots and uh you know the healthcare facilities in general. And uh, you know, apart from that, there are a couple of more hospitals, you know, a few of the largest ones in uh you know uh in Florida and uh globally, but uh you know, Florida, we have a healthcare facility where we have deployed, and uh there are more coming. So the pipeline is getting bigger and bigger. Uh, but what we have seen is like uh the problem which we are solving in healthcare, it's not just labor shortage, right? So for any delivery system, uh there's a you know uh typical labor shortage which is getting resolved by robots, but it's a compounding problem in healthcare because uh you know, even the existing staff, they spend almost 48% of their time in administrative and logistics tasks, right? So that's that's a lot. So what we are solving today is uh the efficiencies uh of uh you know the clinical staff so that the nurses can spend more time on health, you know, patient care. And uh, you know, for the future, uh, we are eventually solving the labor shortage as well, because by 2030, there'll be almost half a million uh nurses which will be not available uh you know in the US hospital. So so that's a big problem to solve.

SPEAKER_03 11:13

Hey, what about what about

Humanoids Are Future Not Present

SPEAKER_03 11:15

humanoids? Everybody's talking about humanoids. Talk talk about that and where that's going and how you're addressing that and why why not now and all those good things.

SPEAKER_01 11:24

Yeah, I think uh you know, uh when I was a kid, uh, you know, the first reference to robots was a movie called Terminator. You know, so so so that's that's how the fascination for humanoids is uh you know everywhere. But you know, that's not uh that's not the reason why in today's time we are talking about humanoids. So um the world is made for people, and uh, you know, people are trying to emulate the same persona as a human into robots. Uh, but uh, you know, there's so much more to humans than robots. So that's why you know what we see here is a typical, a classical approach of what happened to autonomous cars, right? So back in 2009, you know, we were just having autonomous cars next year. So, you know, it it was just next two years. And here we are, it is 2026, uh, where still you know a full rollout, you know, uh west to east coast is has not happened. So while you know humanoids is a fascinating, you know, uh story, uh fascinating uh, you know, aspect of bringing autonomous systems to you know to the household or to the processes which are very close to human function, but uh it's still it's you know there's a long way to go. And this is not the time to deploy, but this is certainly a time to be ready for it. So that's where you know the one thing which I love the most is both you know at our eye points, you know, they they're agnostic. I mean, whether there's a wheel robot or a humanoid or a robot dog, it it doesn't matter. So so uh the idea is to be future ready. Same goes with us. I mean, today we have a software plot platform called autumn.ai, which is working as an orchestration layer, agnostic to the type of robots apart from us. So I think the mantra is to be future ready, but uh when I say this, humanoids are the future, they're not the present for foreseeable future.

SPEAKER_00 13:14

Um, what is at the present is are the ground robots.

Context Aware Robots With Edge AI

SPEAKER_00 13:17

I'd like for people to understand more about the capabilities. What can your robots do?

SPEAKER_01 13:22

So I think uh, you know, uh autonomous navigation is not a new problem. I mean, it has been solved in various industries, be it warehousing, manufacturing, robots have been running in these areas. What has changed is that uh, you know, now these robots are in a much more unknown and you know dynamic environment. So uh the robots need to handle that aspect. Uh when we were running these robots in different environments, uh like uh we started off with airports, but airports or you know, outdoor sidewalks or healthcare facilities, uh each area or each domain has its own set of rules, set of uh expectations from the robot. So really the robots can only function if they start understanding the context of the environment. And that's really you know, uh, you know, we understood after you know spending about a year, you know, where our engineers were on site, you know, uh fixing things or making sure that the robots are understanding the context. Uh, we we started uh you know uh using uh VLMs on the edge hardware uh back in like 2025 onwards, where we are trying to understand the context, where the robots are running, what's what's uh happening, and then have an associated behavior uh to it. Very simple example if uh you know a robot sees an obstacle, the expectation is that the robot will avoid it. But uh what is happening in our case is if the robot is uh in a hospital and if it sees a person in a wheelchair, it will not try to avoid, rather, it'll give the right-of-way. And that is like typically understanding context and what we're trying to

Proving ROI In A Pharma Facility

SPEAKER_01 14:58

do.

SPEAKER_03 14:58

Hey, we we all just got back from Atlanta recently, did a big uh rapid um use case uh of what uh autonomy or I've ai can can uh deliver in a in a large manufacturing uh pharma facility. I think that thing went great, didn't you? I I think they were really impressed how quick quickly we deployed the fact that we were willing to take the risk, right, to to roll that out in just a few days, and it went off seamlessly. If you know, we won't give the name of the customer, but it was very well received. You want to talk about that a little bit and what we proved?

SPEAKER_01 15:36

I think uh, you know, what you touched upon, Dan, is you know, it's something which expands on the utility of the system, right? It's not just like uh, you know, last mile delivery. There are industries which are struggling from you know uh delivery points. And a very good example was a pharma company. Uh, we won't name them right now, but you know, uh when we talk only about robots, uh the whole workflow was different. It was efficient, but it was not, you know, not giving the ROI which they should have gotten by uh providing these robots. But the moment we you know introduced uh you know the arrive points into the overall workflow, it starts making sense that uh you know uh there's no person waiting for the robots, there is no robot waiting for you know people to come and attend. So, you know, it it is the handoff system between the robot and the arrive points, make it so seamless. And there are you know industries, uh, you know, earlier we just thought like you know, healthcare is one of the biggest, but we see industries also evolving where you know you want to completely remove uh or free up the time of a person for the logistics so that you know it can be taken care of by the robots and the infrastructure. So I think uh it's the layer two of uh you know automation uh in general for for a lot many organizations. And uh whoever is having robots at this point in time, you know, uh the customer is understanding now that uh we need more. I mean, this is you know level one, level two is to have a smart infrastructure in

Can Do Partnerships And Real Investment

SPEAKER_01 17:08

place.

SPEAKER_03 17:08

Hey, one of the things I want to say as a CEO of Arrive AI is you know, we we want to be out there and we want to build the brand and show what we can do, right? And we're all in this new industry that we're creating, not a lot of monastization happening right now, a lot of small companies. Uh, you know, everyone has the same plate. You know, you're you're dealing with you know constrained human resources, you know, capital constraints, all those things, right? So nobody has the bandwidth to really deploy for somebody else. And what I mean by that is anytime we've talked to anyone, you know, that they want to know how they're gonna how it's gonna be paid for. That's always the thing. And you know, one of the things about us is uh, you know, we we're willing to shoulder our own responsibility, right? We're out there uh doing whatever it takes. We understand that you know it takes money to make money. You gotta you you can't wait for somebody to pay your way. And and you know, that that has been a big um headwind in the industry because everyone wants everything to be paid for. And the reality is you you have to pay your own way. And one of the things I want to say about you, you guys are the arrived of robots. I I really want to give you a shout out and some love because you have the same can-do attitude that we do. You understand uh the the synergy that can be accomplished when we work together is far greater, it's a huge differentiator. You know, my analogy is we're two NASCARs and we're drafting together and we're separating from the pack because because we're working so closely together. And and you just don't see that. If everybody wants to wait around in until somebody comes and pays their way, they're gonna be left behind and we're gonna be out here owning this market. And um, you know, I want to say that uh we recognize how great autonomy is, and we recently made an investment uh in a in uh autonomy, right? Uh it hasn't been announced or anything, but I'm gonna say it here for the first time. Um so we're excited. We love uh the can do attitude, the mutual uh, you know, um desire to work together and to really create. And there's no holdback, there's no pretenses, there's no send us a check or whatever. It's just, you know, you guys are all in like we are. Uh we're starting to do trade shows together. Uh, you know, we Salil, you're you have a great guy there uh in business development that we've got a great relationship with. Uh we've seen you guys in India, we've seen you here all over the states. I think they're uh we're really starting to build a um a really big uh paradigm here. The the joint companies that I'm really excited about. And I I see uh I see the amount of inbound leads uh that that we're both seeing. Uh people it people are are they getting it, they're starting to understand where where we've been coming from for a while. And you know, I always say if you have an early idea and people don't think you're crazy, you're too late. I think we're right on time now. And and I'm just excited to hear the validation by everything you just said. Kylie, what do you think?

SPEAKER_00 20:10

I second that.

SPEAKER_03 20:11

Okay, good, good. What do you think, Rithager?

SPEAKER_01 20:15

No, this is like you know, uh, as we always discuss one plus one is 11. So I think that's that's that's how it is. So I think that's that's a little monster here.

SPEAKER_03 20:25

I took it away from you. No, no, no, that's great, man. Hey, how's a beer doing? Sorry?

SPEAKER_01 20:31

How is a beer? Okay, yeah. So yeah, my son is good doing fine. I think uh he he he still needs to learn one plus one is his is two. He he can't go for 11 right now. So that that's for us.

SPEAKER_00 20:43

That would be confusing for him.

SPEAKER_03 20:44

Like, what the skip to 11, man, come on, it's it's new math. Let's do it.

SPEAKER_01 20:49

You know, you know, one of the fascinating things which I learned uh the most is you know, uh uh about the love for cars which Dan has. And I think uh it's it's really important to have certain uh passions that shows uh different color of uh a founder, because uh what that means is that uh you know it's it's not just uh uh things which you need to accomplish, it's things which you love and you keep doing it. So I think that's it, that's a very important trait for any founder that uh you need to love what you do. And uh, you know, whether there's a logical reason, everybody will say don't do it, but you'll do it anyways. And I think that separates us out from a perspective that we bring in technology which is which people have not seen or thought that it's too far, right? So I think I think that's that's something which is a good trait.

SPEAKER_03 21:37

I always say by the time he thought about it, I did it. You can't wait, you gotta move fast. If you mess up, you you celebrate it and move forward, right?

SPEAKER_00 21:44

To Rhethiker's point, though, you didn't have to do this. No, you wanted to do it, it was a fun adventure for you. You didn't have to do this. Your livelihood and your like passions in life don't ride on this. You wanted to, you know, that's like that's a really great perspective.

SPEAKER_03 22:00

I think it is, but it we're gonna do something so huge here. I'm gonna be able to get a few more cars.

SPEAKER_00 22:05

Yeah.

SPEAKER_03 22:07

So that's the goal.

SPEAKER_01 22:08

Let's the next one's gonna be autonomous. I think I suppose the next one's gonna be a good one. I'll never he likes driving them too much.

SPEAKER_03 22:15

Yeah, I I want that visceral, loud, you know, I want to be in control, you know. So that's that's what it is. Um, but I'll be passing you guys out there. You guys will be sitting sleeping in the backseat of your autonomous cars, and I'll be zooming by.

SPEAKER_00 22:31

Well, Rethika, anything else you'd want

Mining Deliveries In Remote Australia

SPEAKER_00 22:33

to add?

SPEAKER_01 22:34

I think uh, you know, uh it's it's just a starting point. And you know, one of the key things which I do want to highlight is today, you know, the press is out. So so we are the world's first to do autonomous deliveries at mining areas, mining residential areas in Australia. And uh, we just went live uh with Sodexo a couple of months back, and uh, you know, I'll share the article. So it's a good shout out to Sodexo and Rio Tento, uh, who has been a terrific partner in bringing autonomous uh you know deliveries to one of the hardest jobs on earth, which is mining. You know, for those workers, uh it becomes super, super important. And trust me, uh it makes uh you know uh the teams, you know, just swell that you know we are we are able to contribute and run these robots in so remote situations in Australia, uh, because even the network is you know, uh it's not there, it's a few places, but uh, you know, uh there's a good traction of uh orders which we are delivering each day. So that's great. Yeah, yeah.

SPEAKER_03 23:33

No, that's awesome. Hey, I think to sum it up, the trajectory is straight up, all right? Uh, ton of inbound leads, everyone's getting this unified ecosystem, and the the there's nowhere to go but up, right? What goes down must go up. That's what I always say, right? So let's do this, man. Hey, it's great seeing you. Thanks for joining us, man. Um, say hey to the family, and I can't wait to see you here in the States very soon, okay?

SPEAKER_01 23:58

Absolutely. Thanks for having me. Thanks, yeah. Thanks, Raddy.

SPEAKER_00 24:02

Yeah, bye, Rithiker.

SPEAKER_03 24:03

Bye, man.

Wrap Up And What Comes Next

SPEAKER_00 24:06

Uh some he he's just been uh wonderful to work with and um you know helping us advance the cause, like going in and and he's realizing the importance of that infrastructure.

SPEAKER_03 24:21

Just as he's willing to put his money where his mouth is, just like we are. One of the things I find, I don't care how big it could be Amazon, nobody wants to pay your way, or they always want everything paid for. And that is, you know, that's that's a constraint for a lot of companies. But the companies that really get it are willing to say, hey, you know what, we're gonna shoulder our responsibility. The the the uh outcome of what we're gonna invest here is gonna be bigger than the investment. Yep. So let's do it. And you know, looking for those great companies that have that can-do attitude, uh, that want to work and deliver the future. And that's what that's what we're doing. So it's exciting.

SPEAKER_00 24:56

It is another big thanks to Rithiker Vijay, again, the CEO and founder of Autonomy, one of our great partners.

SPEAKER_03 25:03

And nobody knocked on our door, nobody came in.

SPEAKER_00 25:05

Nobody did uh the episode last week. You might have uh if you missed it, you should go back and watch it because I didn't do it, but yeah.

SPEAKER_03 25:13

So that's the I don't think anybody will ever do that again.

SPEAKER_00 25:16

Hopefully not, right?

SPEAKER_03 25:17

Because we're gonna ridicule the heck out of them. So anyway, thanks for joining. Yep, bye guys. Thanks, Kylie. See ya.