Preview Mode Links will not work in preview mode

There are four waves of innovation sweeping through the automotive industry that will disrupt vehicles more in the next 10 years than they've changed in the last 100.

Each week, we explore connected cars, electrification, changing ownership models, and autonomous self-driving vehicles, as we seek to understand and prepare you for the future of transportation.

Apr 10, 2019

EP006 - Chief Technology Officer at Designated Driver, Walter Sullivan

http://www.vehicle2.getspiffy.com

Episode 6 is an interview with Walter Sullivan, Chief Technology Officer at Designated Driver; recorded on April 2nd, 2019. Walter and Scot discuss a variety of topics, including:

  • Walter’s career path from Microsoft to Designated Driver, which launched last October.
  • What Designated Driver offers to the autonomous vehicle space, as well as Walter’s thoughts on the implementation of AVs.
  • How the transition to 5G will positively impact companies and startups moving forward.
  • Realistic expectations for the current shift in car ownership, with reports showing up to 80% of new cars sold in 2030 being owned by fleets or shared services.
  • Regulatory hurdles for Designated Driver, as well as autonomous vehicles at large.
  • Defining the future tipping point for electric vehicles to outperform internal combustion engines.

Be sure to follow Walter on LinkedIn

If you enjoyed this episode, please write us a review on iTunes!

The four pillars of Vehicle 2.0 are electrification, connectivity, autonomy, and changing ownership models. In the Vehicle 2.0 Podcast, we will look at the future of the auto industry through guest expert interviews, deep dives into specific topics, news coverage, and hot takes with instant analysis on what the latest breaking news means for today and in time to come.

This episode was produced and sound engineered by Jackson Balling, and hosted by Scot Wingo.

 

Transcript:

 

Scot:

[00:52] Welcome to the vehicle 2.0 Podcast! This is episode 6 and it's being recorded. Tuesday, April 2nd, 2019. Welcome back, Vehicle 2.0 listeners. I am a serial entrepreneur and my first company, which I started way back in 1995, worked really closely with Microsoft. That was called stingray software. And I worked real closely with the Visual C++ team and that is where I met today's guest, Walter Sullivan. So doing the math on that, it's over 20 years. And then my career took me to eCommerce and Walter's took him to the automotive world. And now I am in the automotive world. Walter is now the CTO of Designated Driver and I really look forward to hearing more about what he's done since we last talked to probably 15 years ago. And I'm excited to have him on the Vehicle 2.0 Podcast!

Walter:

[01:49] Awesome. Thanks Scot. You're making me feel old.

Scot:

[01:52] Well, I didn't say we met when we were 12.

Walter:

[01:55] That's true. Okay, good. Good point.

Scot:

[01:56] So, I know your career path and I had a couple of highlights in there, but I'm sure there's a lot more on the journey. Tell listeners about your career path and how you ended up where you are today.

Walter:

[02:11] Yeah. Great. Yeah, so I, as you mentioned, I started my career at Microsoft. I spent really 25 years there in different capacities. Started up building development tools, which was where I was lucky enough to be able to work with you and a number of other really interesting people. So that was a lot of fun, that was sort of my first half of my career at Microsoft. Second half I moved into our embedded operating system group and started leading parts of a emerging team there that was building embedded technologies for vehicles primarily for, you know, navigation systems and infotainment or entertainment systems in those vehicles. And from there, I took an opportunity to move to a German automotive software company called Electro Pads. I opened up a research and development office for them down in California. So until that time I was living in Seattle. Microsoft is in a suburb of Seattle essentially and moved down to California to the San Francisco Bay area, open up a research office, spent a few years running that research office and last November, left that to start a new company, designated driver.

Scot:

[03:29] Very cool. So if anyone listening has Windows Mobility in their car, they can call you for tech support. Is that how it works?

Walter:

[03:40] Yeah, pretty much. Exactly. I'm sure you'll poke, give him my phone number at the end of that. At the end of the podcast here. But yeah, I built a number of systems for Ford or Kia, BMW. quite a number of car makers. That platform, actually, even today is still in quite a number of cars being, probably no longer that helpful from a support standpoint.

Scot:

[04:05] Okay. All right. We'll have to go online and figure it out.

Walter:

[04:11] Probably.

Scot:

[04:12] Cool. Or ask Clippy

Walter:

[04:13] Or ask Clippy. Exactly.

Scot:

[04:15] So let's talk about designated driver to, I know it, I know the name from the, context of, you know, obviously if you're out drinking you need a designated driver. but, but tell me more. That's not what you guys do. Tell me what you guys do, do.

Walter:

[04:33] Yeah, I mean, the name, it comes a little bit from that idea. Designated driver is really about providing, what we call tell operations for autonomous vehicles. And, and let me break that down just a little bit, is we're sort of moving into this world of autonomy. Vehicles that are carrying passengers are good. So, or other, other things, many of them will start to become more and more autonomous. And which is I think great from a shared mobility perspective and a usage perspective. And you know what to think. There's a lot of promise for the technology, but a is we're actually getting closer to the commercialization of that. The realization is that there's still some scenarios where we just haven't been able to train or develop autonomy systems to, to handle correctly. And this is where it designated driver comes in. So we provide the designated driving services. I'm a human foreign to autonomous vehicles that needs that human assistance essentially. and so the, the name is a play a little bit on the, on the concept. Have you been out drinking too long and you're really not safe for you to drive home because there are situations where maybe it isn't safe or just not feasible for an autonomous vehicle to drive itself. So that's the, that's the basic background.

Scot:

[05:58] Cool. Let me, let me give you a kind of a scenario just to see if this is, if I'm, if I understand it. So I saw the CEO of Waymo earlier this year said he doesn't think we'll ever get to 100%, and he cited weather. So for example, when it rains, the rain makes it very hard for the lidar to see the, you know, not only the surroundings but the, the lines in the road, for example. so is that a scenario where you guys would automatically kick in and a human driver somewhere would kind of take over or in some way augment what they view is doing?

Walter:

[06:29] Yeah, I think eventually we do get to something like that. I, I do think that a lot of weather conditions that we struggle with today in autonomy, we'll figure out how to solve. there's, there's new generations of sensors coming that can address some of that. There's better, you know, there's additional training for the, for the machine learning and artificial intelligence systems that we're working on to, to improve that. So, so maybe that specific scenario, we'll eventually get addressed. But conceptually that's exactly it. So things that I think quite a lot about art or when a autonomous vehicle pulls up onto a road that is closed because of construction, you know, maybe there's a path that the construction workers are guiding vehicles around the construction area, but that path may violate road rules that the autonomous vehicle has learned and, and hold sacred so to speak.

Walter:

[07:28] And we might need a human to actually step in and either tell the autonomous vehicle it's okay to break a certain road rule or in fact the human may actually just, step in and drive the vehicle as if they were sitting in the driver's seat of the vehicle and just remotely maneuver through, through the, through, in this case, the construction site. But in addition to that, there are maintenance facilities or, or other kinds of really kind of specialty purpose environments that have vehicle might be to operate that an autonomy system will just never be trained for because it takes quite a bit of effort and money to train an autonomous system to drive through a specific environment. And sometimes the return on that training investment just won't be worth. And then I think the last, the last scenario and the one that, we also picked quite a bit about is these things are still vehicles and they will still be operating on roads with human drivers.

Walter:

[08:28] And there's still the possibility that there will be collisions and accidents and, and, other kinds of failures other than vehicle, a flat tire or something like that. And then the autonomy system just won't really be able to cope with those kinds of circumstances very well and kind of human will, we'll be able to take over a vehicle and maneuver it into a safe location for a tow truck or for whatever the, you know, the action that needs to be taken for that vehicle. So there's quite a number of these scenarios where, where autonomy just would never even be able to, to handle it in any case.

Scot:

[09:03] Cool. And then where are you guys in the development of the solution? Is this kind of Napkin diagram stuff or do you have deployments or you're an in kind of in the middle there?

Walter:

[09:13] Yeah, we're, so we are beginning our first deployments. It just kind of to maybe give you a bit of the history in the company. We started the company October of last year. As I mentioned, I joined 1st of November, you know, we spent the last four or five months kind of building up the core technology and two weeks ago in San Jose at, invidious what they call their GTC, their GPU technology conference, we'd launched the company kind of formally. So maybe some press that you've seen recently is really sort of the result of that that company launched. But it, in that launch we drove, participants in a car around the convention center in San Jose from our office in Portland. So the driver of the vehicle is basically 700 miles away, remotely operating the car that people in San Jose we're riding. And so that was kind of the launch of the company. And that's, that's really, you know, that the state of our technology where we're pretty confident in it and it's Lisa is mature enough that we were, we were comfortable, you know, driving journalists and customers and other conference participants around, around a conference center from 700 miles away. So, as I said, we're, we're now starting our first deployments, some of the technology.

Scot:

[10:30] Cool. What's, what's the passenger, for that demo, maybe we talk about the passenger experience. Am I, can I talk to you kind of like an Onstar type of scenario or can I see a little video of my operator? How, how, what's the cabin experience?

Walter:

[10:43] Yeah, that's, you know, it's pretty much exactly the passenger experience we developed entails screens and the rear seats of the car. So the passengers essentially would sit in the backseat. I was actually sitting in the front seat talking to them answering questions and the, the screens show, you know, a realtime position of the vehicle. They show the state of the vehicle whether the vehicle is driving autonomously or whether there was a remote operator controlling the vehicle. And, and when a remote operator takes control of the vehicle, there's a kind of series of introductory screen. So the passengers most likely, at least in the scenario we were talking about there, which is the autonomy system encounters some sort of failure, the passenger of the vehicle is most likely going to recognize that there was a failure. So we thought, you know, let's have the remote operator introduce themselves and establish a a video.

Walter:

[11:41] They have a two way video, a link into the car, and they can sort of make the passengers feel comfortable than actually someone is taking control of the vehicle. We're going to move new or the vehicle into a safe location. You know, everything is sort of being taken care of as a passenger. You don't need to worry about the fact that maybe that was a failure in the vehicle and sort of establishing this human connection, would help maybe ease the ease, the anxiety of people who might be in their vehicle. Ultimately, I think the passenger experience will be defined and determined by the, the company who is operating that vehicle. So, if you imagine, maybe it's a writing company, you know, the ones that many of us probably use every day, they might have a specific passenger experience, they want to have it in their cars and we would certainly help them implement that. But but yeah, for us it was a two way video link to the remote operator apartments so passengers could see them, could talk to them, could ask them questions.

Scot:

[12:44] Cool. And is your, is your business model where you could, you could have the remote drivers yourself or you could even just license the whole system and someone else could have their remote drivers?

Walter:

[12:55] Yeah, so we're building the technology and the business so that we're actually licensing the technology. So we do have our own trained a remote drivers that we make available to, some of our early customers really for their sort of trial or pilot fleets, not, not their mass scale commercial fleets. Because we expect the evolution of the, of this tell operations industry to be that larger fleet providers. You know, maybe like a ride hailing service. They would probably operate their own tell operation center. Yeah, I'll call it where they have a group of, of remote operators that are sort of monitoring and managing your speed. And I just want to, I want to live, I'm on a license on the technology that they're using to do and basic. So that's the US the business that we're planning to build.

Scot:

[13:48] Got It. Very cool. Well, congrats on the launch. I didn't, I didn't realize we were this close to when you launched. yeah, yeah, it was very timely actually. Like, yeah, I feel like we've got a scoop and I didn't even know it. so, so the whole idea on this podcast is to really look at, it seems like everyone has one of these frameworks, but ours is the vehicle to oh framework. And we think about conductivity, all these new ownership models, electrification and autonomy. and you know, you're kind of sitting squarely in, in all of those, which is great. So you've, you've thought a lot about autonomy. Obviously, if you guys are already kind of seeing some of these edge cases where you'll, you'll need to tell our operations to be involved and whatnot. yeah. What's your point of view on, on when, when we're going to have autonomy at mass scale and it seems like you guys are going to be, you know, kind of sit between that level three, four or five area. I would love to hear your thoughts around just autonomy in general.

Walter:

[14:44] Yeah. I think I always felt, I always tell friends of mine that I think this is the most exciting time to be working in the automotive industry, especially the technology side because of the really those four points they knew that you raise, I mean connectivity I think is really a foundation to, to enabling sort of the new business models and ownership models and autonomy and, and so I, I think that autonomy, what's clear to me is autonomy has coming. I think where the debate is, is how, how quickly do we start to see it? You mentioned mass scale. No. The question in my mind is what do you think is mass scale? Like we will, I will deploy my technology in on top some autonomous vehicles operating in the public offering rides to the public this year. So we'll will be on the road this summer.

Walter:

[15:36] Other companies, you know, there's a number of companies operating sort of limited service shuttle programs or other kinds of, there's a company here in the bay area called neuro. It's doing kind of a grocery or package delivery, autonomous shuttle and they, they've began testing. There's an Arizona and you know, the, I think we're at the stage where the very first sort of significant test fleets are getting deployed. You know, those tests. Fleets are there to gather data together, training data to improve their autonomy systems, to gather operational data, to understand what are the, you know, the costs and the, and the operations needs of these vehicles. And this is a multi year process probably to gather all of that. I think you are, I will be riding in and autonomous Uber or Lyft or whatever, ride hailing vehicle, you know, by the middle of the next decade by, you know, by mid 2020s. I think it's quite a bit longer before you or I, and we'll buy our own autonomous vehicle if we ever do this. Sort of maybe gets to your other point about the, the ownership models it may be then that private vehicle ownership of autonomous vehicles just never really happens because the business model for them is really better. you know, operated in a ride hailing kind of service, but, but I think it's, it's you know, Mass Dale, it's probably measured in decades rather than years. Cool.

Scot:

[17:00] dude, so it seems like Waymo is kind of out front and then you have Uber and Lyft to all working on things who've got an apple doing something as a long term fan of Microsoft. I've always been surprised they're not really active. Do you think you think they kind of step into this in some way or do you think this isn't really their scene now and they're, they're more into like cloud computing and other stuff?

Walter:

[17:21] Yeah, it's a good question, which probably no longer really know too much information about. I mean, I think just based on what I see, which is the same stuff you see that, that what they're really looking at is how do we enable all kinds of interesting automotive scenarios with their cloud technologies. So, and so it's clear. I mean it seems, it seems every week they're announcing a new partnership with a carmaker or some other automotive related business using their Azure platform in the cloud. Did that. That to me seems clearly what they're, or at least primary focuses, whether they get into the vehicle side of things again or not. It's a good, I really don't have any insight on that. You know, apples a little bit, I would say pretty secretive about what they're doing in this space as well. I'm not really sure what apple intends to do. Waymo is the one who cause Ben most public. Right. And they, they, they really have, they had the most, the most experience, the month, the most mature technology I think out there from.

Scot:

[18:26] Cool. oh, pivoting from ab to conductivity, one kind of question about designated driver. you know, we're in a 4G world heading to 5G is that data connection and the coverage is good enough for you guys to do the, the remote driving solution?

Walter:

[18:42] It, it sort of is I of course connectivity, throughput, latency, these are, you know, these are all things that can always be improved. But let me just maybe describe a little bit one what our technology, what we use for technology today to remotely operate a vehicle. We have to equip the vehicle with some technology that allows us to safely operate it remotely and that rarely includes cameras that give us essentially 360 degree view of the vehicle. We need to be able to get the position of the vehicle so that we can always know where the vehicle is and we can, we use that to identify certain environments that the vehicle might be struggling with. We need a computing module in the vehicle that has a pretty significant amount of computing horsepower that, that we use for processing that video from those cameras. And it handles the communication with the driver's station, essentially both the up and down communication and then, and then the interface with the vehicle control system.

Walter:

[19:43] But when we talk about the connectivities specifically, we use a multi radio cellular modem in the car. So it's, you know, it's traditional 4G LTE cellular radios, but there's four of them that we use in the car to provide the connectivity to the back end. And the four is not necessarily because we need that bandwidth. What we actually use multiple radios for is to try and ensure the robustness and reliability of that connection. So we actually communicate over those four radio simultaneously because each of those radios is provisioned independently. So it could be on each, could be on a different cellular network. we actually, in our development vehicles, we use two on one carrier and two on a second carrier and essentially, but, but, but each could be on, on its own carrier. And that's really to try and ensure robustness of that connection. And then we have some proprietary algorithm about that divide the communication across four radios and send some redundant communication across multiple radios to try and just to ensure that we're successfully communicating with the backend. But we can do that all with, with LTE or 4G today. It's not really too big a problem in our view. Life just gets better when 5G gets here.

Scot:

[21:07] Yeah you could probably go down to two antennas. You still want to have double coverage so you don't have it single point.

Walter:

[21:12] Exactly. Yeah. Yeah. We still need some of the robustness and redundancy, but I mean 5G has the promise of reducing some of the latency that we see. So, so we we try and stay, we have a magic number. We try and stay under about a hundred milliseconds with a hundred and 150. We actually stop remote controlling a vehicle, but we try and stay under a hundred milliseconds for that communication latency it feel. So that's video up and the commands back down into the vehicle. That's pretty fast. But you know, it can always be faster. I think latency is the thing that is really the, the critical aspect of, of remotely operating a vehicle, a vehicle going 30 miles an hour in a hundred milliseconds travels about four and a half feet and in you, so, you know, you can't, you don't want to have significantly more latency than that just for the safety of, of operating the vehicle. And, and in fact, if you could cut that latency in half, you wouldn't, you would certainly love to do. Yeah.

Scot:

[22:12] Interesting. I never thought of it in those terms. That's pretty interesting. How about now that you live in San Francisco? I know you're, you're a car guy. did you have all of your cars? Cause it's, it's unhip to have a car in San Francisco.

Walter:

[22:27] I guess I'm just uncool cause I, I have two cars and two motorcycles down here, which really just means like, I, I have a pretty hefty parking bill every month to park my vehicles. Yeah. It's, it's funny whenever I was just going say, I mean as you said, I'm a car guy so it's hard for me to imagine not owning a car. But that said, when I'm in this, when I'm traveling in the city, I pretty much never drive. So I always either take a ride sharing service or, or public transit because parking in San Francisco and it's kind of just ridiculous both from an availability standpoint and an expense standpoint.

Scot:

[23:06] Yeah. And if I remember right, and this is really stressing my memory, your, your guilty pleasure is Aston Martins. Sorry, Alfa Romeo.

Walter:

[23:15] No, I used to have us, I no longer any Alfa Romeos actually, but yeah, I have, I have two, nine 11 actually. So it's a different kind of guilty pleasure. Those are fun on this San Francisco Hills. Then they, yes, this, the hills don't bother me so much, but the quality of our street pavement here and needs a little bit to be desired. So then the cars are not too far. That perspective being a lot of rims. Yes, exactly. Exactly.

Scot:

[23:50] Cool. Well, where do you see car ownership going? You're, you're kind of living in the heart. There were, you know, most people don't own cars anymore and, and they're using the ride shares and it seems like we get contacted by a new company trying a different type of car sharing pool and there's seems to be like 80 different models under test right now. do you see car ownership kind of leaving from individuals to kind of more of a fleet model, over the, over that same kind of timeline we talked about with AVs?

Walter:

[24:18] I think that there will be, there are certain environments and certain people who, I think car ownership, remains for four decades. You know, you're, you're, you're in a city like San Francisco or Chicago or New York or you know, a densely populated city. The benefit of owning a car is that, I mean, I even asked myself this question and you know, so a little bit questionable, but when you, you know, go to start a family, you move out in the suburbs, suddenly a card becomes much more a utilitary and then then maybe in the city. and certainly if you're even in more rural environments, having access to or or ownership of your own vehicle I think makes a lot more a sense for those people. I see car ownership surviving for actually quite a wild. However, what I think is clear is that people younger than, than you or I are, are much more happy with different kinds of mobility services and, and they're delaying the purchase of their first car or even they're getting a driver's license can. I certainly would expect that to continue. Well, younger people who who at least earlier in their lives or are living in more urban environments and they just, you know, they're probably making a wise decision just not to bother with a car. Especially when you have such convenient, you know, point to point mobility services of animal. When do we reach a tipping point of car ownership? I know, I have no idea. That's a really good question, which I'm, I don't know how to speculate on,

Scot:

[25:53] well a lot of these fleet solutions, so we had a guest on that was talking about, you know, you have kind of station based, which is like the zip car model where you go between station a and B then yeah, more fluff free flowing ones. one of their challenges is, you know, so if I, if I get in a car and an apartment building and then kind of take it to maybe an, an area out in the suburbs, they have to send these valets out to kind of rebalance the fleet. Is that a possible solution for your technology where you could equip these cars with, with the, the tell operations and at night, you know, I remote driver could be driving these things kind of even outside of autonomy?

Walter:

[26:29] Yeah. Actually I think, I think that's a great observation and it's definitely one that we've, we've talking to a couple of companies about the possibility of doing that. They send a fair amount of money repositioning and optimizing the, the location their fleet for, for the next, renter. Essentially that business survives on utilization of those vehicles. You're trying to amortize the cost of that vehicle over as many people as you can. So it doesn't do much good for that thing to be out in the middle of, of a neighborhood where very few people are gonna walk by and rent it. And I think that's definitely a possibility for us to, to offer a service to help reposition those fleets. And there's a couple of car share companies that are interested in and exploring that. So yeah, I think it's a definite possibility. and the benefit for me as you pointed out is I don't necessarily like, you know, and for, for my businesses, I don't necessarily have to wait before autonomy to, to scale my business. I can, I can develop businesses before maybe autonomy. Is that a mass scale?

Scot:

[27:33] Yeah. And then, in the ecommerce world, drone delivery, you know, everyone talks about it, but the regulatory hurdle is so high for autonomous drones, that I think deaf a is doing more, allowing kind of, you know, remote driven drones and I could see the tell operations having a little bit easier kind of, you know, getting through some of the regulatory hurdles, especially you guys seem to have it all dialed in. The of, you know, what happens if there's not conductivity and all in and all that kind of stuff. So it seems like maybe that would be an easier on ramp into things through the regulatory hurdles too.

Walter:

[28:05] Yeah. I mean there, there are some challenges particularly we have to the first, pretty much, there's not a single state that has contemplate at the fact that they driver of a vehicle might be not in the vehicle. Yeah. so they make all kinds of assumptions like law enforcement and can identify who the driver is. They can get their driver's license and insurance information that the operator of the vehicle or has certain control over the vehicle. And, and you know, we can guarantee a lot of that, a lot of the same things to law enforcement. For instance, in fact, that's when we were doing our company launched down in, in San Jose, one of the, I actually contacted the San Jose police department as well as the California highway patrol just to make sure I wasn't crossing any lines I shouldn't cross doing, doing this remote driving essentially.

Walter:

[29:01] And they, they basically told me that, you know, as long as I wasn't doing this for hire, because there's a set of regulations that would apply if I were doing it for hire. But as long as I wasn't doing it for higher, which in this case I wasn't, and I would be able to provide law enforcement with the driving license of whoever was in control of the vehicle when, you know, at the time the police maybe became interested in the vehicle, I'll say then. Then they were okay with, with what I was doing actually. And that was the, that was the first, my first interaction with law enforcement just to see what the, you know, what the feasibility have of doing this might be. And it was, I, I would say, surprisingly positive actually. So, so I am hopeful that we'll be able to cross some of those regulatory hurdles like you said, but okay. But today I would say it's really a vague spot in the loss. It's not contemplated really at all.

Scot:

[29:55] Yeah. It's gonna be interesting to see how all these rules and regulations keep up with, yeah, they've struggled with scooters, let alone, you know, cars that are being driven remittent

Walter:

[30:04] exactly, exactly. But we'll get there. I'm that I'm pretty sure.

Scot:

[30:09] Do, do you have a prediction of which of these different ownership models ends up winning? You know, is it, is it going to be ride shares, subscriptions or, or more kind of shared pools of cars?

Walter:

[30:19] I think there's a viability for all, to be honest. I'm not sure it's one, I don't know that one wins. I think that each, each actually has some interesting applicability under specific scenario. So, so I could see a world in which, you know, they all continue to exist in. Maybe it'll vary a little bit by geography, but I think they, I think, I think we're, you know, all of them, I would say all of them are sort of mature enough now that I think they understand the business model and the operational cost and efficiencies that they need to achieve. And you know, I could easily see them all surviving, you know, sort of in their current form. The question is whether somebody comes up with a new one that's going to kind of displace all of them. I don't let that, I don't know.

Scot:

[31:04] Yeah, you could kind of see the moat multimodal thing that both Lyft and Uber are doing around, you know, car bike, scooter. You can almost see different use cases, which would be, do you need a car for a trip a day, a weekend, a week, a month, and know exactly. Like choose one and there's all, yeah,

Walter:

[31:22] yeah, we're or, or do you, yeah. Or do you just need you know, to get from the bus stop to your office or the train station to your office? Right. I mean, that's, I think what Uber and Lyft are doing with their bike and in scooter acquisitions is really pretty interesting that that's certainly the, there is, you can imagine they're very popular here in San Francisco because we, we have a fair amount of, of public transit here and more or less all forms, whether that's a train and our assembly or or a street car or or a traditional boss. And, and there's a lot of people that use these bikes and scooters as sort of that, you know, the proverbial last mile sort of solution. And I think it works. I think it works pretty well. But I think, you know, a company like such as Lyft or Uber, I mean these, these are, these are mobility providers and I think they're getting, they're gonna look at every model of mobility, you know, that they can offer to their customers. So a multimodal solution makes a lot of sense.

Scot:

[32:20] Yeah. How about that? So the last leg of the stool we haven't talked about is electrification. it seems like your solution works. You, you don't care if it's an internal combustion or electric vehicle. But do you have a point of view on, on when we get to some kind of a tipping point with evs?

Walter:

[32:38] I'm pretty of the power, the propulsion system, so to speak. but you know, electric vehicles, there's a lot of advantages to an electric vehicle in enabling autonomy, both from the power available. So, you know, the, there's a lot of power to run very powerful computing systems that we need an autonomous vehicle. the architectures of electrical vehicles, electric vehicles are more or less, you know, developed new from the, you know, there's very few electric vehicles that carry over historic vehicle architecture. And so that allows us to really build systems that, that an autonomy system can remotely interact with much easier. you know that, I was reading the news the other day actually about the Tesla launching the model three in Norway. And this, I can't remember, it was last the last month or the last or the first quarter of this year here, but Norway now, 58% of the vehicles sold in that month or quarter, whichever it was, our zero emission vehicles, which is, this is really pretty interesting because that's a lot of those Scandinavian countries are pushing for, for these, you know, reusable, recyclable, zero emission technologies.

Walter:

[33:57] And to see that they at least one of those countries has, has, has crossed the majority, you know, market share and to electric vehicles I think is very interesting. I think for, for a country like the U s were probably 30 years away from seeing the majority of vehicles sold in this country being electric. But I think we're on our way. I think, I think it's, it's definitely the future of, of powertrains in my opinion. So if Porsche comes out with an electric nine 11, are you switching over? You just be like that. You're like the home of the, the engine. Well, I, what I do like the hum of the engine. I wish I would not, I would not shy away from an electric non 11 by any means. And they do have an electric sportscar coming my, I'll say my, my personal business model doesn't really support me driving brand new Porsches does. So my, my forces are little older, so until those electric courses become, a few years old, probably I will be switching.

Walter:

[35:05] All right, we'll have you back on in five years and we'll, we'll do a check in. There we go. That sounds good. That sounds good. one thing I would, I mean, electric motors, I was just going to say, I mean the Torque of an electric motor is addictive. Oh yeah. I think, you know, someone, someone who's, who loves internal combustion engines, which I love. I love the smell. I love the sound, I love the feeling of them. But you get in an electric car and you stomp on the gas and you get pin back into your driver's seat and that's that's a feeling that gets, gets addictive really fast.

Scot:

[35:39] Yeah. I'm I'm a tussle guy, so I've lived that many times.

Walter:

[35:42] Yeah. There. You're there. Yeah, exactly.

Scot:

[35:47] And then so as a guy that's been around the mobility space as an ecommerce person, it's interesting because the OEMs would, their dealer network, they feel like they're kind of stuck in their current model. A lot. Like some traditional retailers were once ecommerce came around and they kind of went through these different phases of denial and then like, oh crap. And then kind of like, you know, existential threat. And at that point some of them made the decisions and have survived and others didn't. What do you think happens to kind of the, as we kind of play through these different trends, like maybe out 15, 20 years, what do you think happens to that traditional dealer network and that traditional OEM,

Walter:

[36:26] The traditional OEMs? I think most of these guys really, their product portfolio changes obviously. So when we move into autonomy and electric power train, these guys, you know, the traditional OEMs will definitely be producing those kinds of vehicles. I don't see electrification or autonomy is any kind of doomsday scenario. Okay. That might not, there may be some smaller manufacturer or some manufacturer of it. It just fails to make the transition. But as far as the general industry grows, how it goes, I think the number of vehicles sold in 50 years is not going to have to look that much different than the number of vehicles sold today. Maybe it's even a bit more. And so I think the industry from that perspective, we'll remain pretty healthy for them. From the dealership perspective. That's a good question. I think dealerships, I think, I suspect that their businesses will just evolve. Electric vehicles still have maintenance requirements and they still have, you know, servicing is cleaning and other kinds of, of services. And I can see how traditional dealerships could, I mean they still have to sell the vehicles, you know, from as new to begin with. but I think that their, their businesses evolves a bit whether they are there as many dealers in 30 years as there are now. And I'm not really sure. I think that's a good question.

Scot:

[37:55] Yeah, I think the ownership one's tricky. So if we do move to kind of like, you know, half personal ownership, half more of these shared services, or do the dealers kind of become suppliers into there and service centers or are they kind of left out of the picture? We'll have to see how that goes.

Walter:

[38:09] Yeah, I mean, I would assume that they would become, you know, service centers for fleet operators and they will compete, you know, a fleet operator doesn't necessarily need to to build up their own expertise in cleaning and servicing and maintaining the vehicles of their fleet that can be contracted on and that can be contracted out to a dealership as you know, as well as any other company. So, you know, I could certainly see their business evolving, but I think that's where some guys will evolve and they will, you know, thrive and in sort of an, business climate like that, maybe some guys will, they're still there. There are still places where people board their horses and ride their horses and you know, people will take care of your horses for you. And maybe that's the future of some dealerships. If you, if you believe the analogy then maybe our, our old gas driven cars or you, they're the horses of the future. Where are we? We drive them occasionally and in special environments and whatever. And so they might get dealerships that specialize in that. But yeah, I dunno, I, I wouldn't be if I were a dean, if I owned a dealership, I wouldn't be that concerned about the future of my business. I would just be thinking about how do I evolve my business to, to serve as the, you know, to service this new industry in this new market.

Scot:

[39:32] Yeah. One, a couple last questions here. I know we're, we're going to run up against time in a second. The, so you, you've spent a lot of time kind of at your time at Microsoft thinking about the in dash experience. and no, so Microsoft was kind of early. They're within bedded windows and then now we have kind of apple, Google and Amazon battling it out for that experience. And then the Em's kind of all have their own experience too. How do you think about what's going on in there and is there a winter or do you end up with a lot of different kind of experiences?

Walter:

[40:01] I haven't been thinking about that space in quite a while now actually. I don't know that there's a winner. And then you have companies like Tesla for instance, that have really defined the Tesla experience in the car. Right. And, and I think a lot of the traditional manufacturers, they still prefer to present their own experience in, in the car to, to the passengers and drivers of their car. I don't know that I see that changing that much. I think a lot of companies, whether it's Microsoft or Google or whomever are, are today focused on helping those OEMs present a bespoke experience. But at the same time we see carplay and android auto becoming more and more prevalent for, for basic services and car. But I don't know, I don't know if I have an opinion on whether there's a winner or not. Actually, I haven't thought too much about in recent years.

Scot:

[41:00] Fair enough. You had to, you had your fill of worrying about that back in the day?

Walter:

[41:03] I had my fill of worrying about. Exactly. Indeed.

Scot:

[41:08] Now you're just trying to drive the cars remotely and that's a much beefier problem. I think. I'm just trying to provide your designated driver. Exactly. Cool. Well this has been really awesome to hear. you know, from someone that's been in the industry for awhile where you think we're going. Any last thoughts on the future of vehicles you want to share with listeners?

Walter:

[41:28] Well, I think I would just go back to one of the things that I said for people in the industry, like myself working on technology, it's, it's the most, I think it's the most exciting time to be working in this industry because we're, we're sort of on the brink of, of some pretty interesting a revel evolution or, and maybe even revolution in transportation. I'm, I'm excited for her personally and I think, hopefully your lessons listeners are as well because I think we're going to, the next 10 years is going to be a pretty interesting time for those of us interested in vehicles.

Scot:

[42:04] Awesome. Last question. for listeners that want to learn more a and follow you online, your website is designated driver.ai. do you publish, are you a prolific tweeter or linkedin writer? Where can people find you?

Walter:

[42:18] No, not really. I am on Linkedin so people might, people can certainly look up my profile on linkedin. I'm not much of a Twitter. I do write or contribute to some of the blogs that we've put up on our website, designated driver dot. Ai. That's probably the, the easiest place to see what my thinking is in the industry.

Scot:

[42:36] Awesome. Well we really appreciate you coming on today and excited to keep an eye on what you guys are doing. We'll have to have you back on after you kind of have some deployments out there and hear what you've learned and see if, see if your perspective on the future vehicles has changed.

Walter:

[42:49] Yeah, that'd be great, Scot. Appreciate it. It's been fun to talk to you again.

Scot:

[42:51] Yeah, we have to, you know, catch up more than every 15 years.

Walter:

[42:55] I think so, yeah. That would probably be a good idea. We're on opposite sides of the country though, so it's a little bit challenging.

Scot:

[43:03] Yeah. Yeah. Well next time I'm in San Francisco, I'll swing by and we'll, we'll grab a beer.

Walter:

[43:07] That would be awesome. I look forward to it.

Scot:

[43:09] Thanks for coming on the podcast. We really appreciate it.

Walter:

[43:11] Thanks a lot, Scot. Take care.