Uber, Nissan, and Mercedes Chose This Self-Driving Startup | Alex Kendall, Wayve

"Every vehicle is capable of driverless operation. That's clearly the steady state of where we're going." Wayve started in a rented house in Cambridge with $1.5M, a car in the garage, and an aim to integrate end-to-end AI into driving. A decade later it's driven across 506 cities without a single HD map and is worth over $8.6 billion. In this episode, CEO Alex Kendall joins Lukas Biewald to talk about how he built the AI driver Uber, Nvidia, Mercedes, and Nissan all backed, and why putting self-driving AI into 100 million cars a year is a far bigger bet than 10,000 robotaxis. Waymo and Tesla both come up. He doesn't shy away. Connect with us here: Alex Kendall https://www.linkedin.com/in/alexgkendallLukas Biewald https://linkedin.com/in/lbiewaldWayve https://www.linkedin.com/company/wayve-technologiesWeights and Biases https://www.linkedin.com/company/wandb

Speaker names marked ? are inferred from speaking time and the show’s known hosts — they are not voice-verified.

• 0:00Speaker 00

  Autonomous driving is all about looking at the AV problem with an AI approach. There is an opportunity there to drive accidents down to near zero. So what we did is we raised one and a half million dollars, got some friends together. We rented a house and we put our car in the garage and started hacking away, taking it from expensive retrofit vehicles, which relied on compute HD maps infrastructure, to mass market vehicles that you can buy or manufacture for thirty, forty, fifty thousand dollars each. It had inbuilt hardware that's in global supply chains, doesn't need an HD map, so it can drive anywhere. We're now the first company to have driven zero shot in over five hundred cities throughout Europe, Asia, and North America. It's also driven in over ten different cars, from electric vehicles to vans to SUVs. We went north of the Arctic Circle, tested driving in twenty-two hour darkness and snow. We were even in Tokyo during a typhoon, where local trains and buses shut down.

• 0:45Speaker 00

  No disengagements, perfect autonomy the whole day. And it demonstrated that we truly can generalize to any vehicle, anywhere. I think now the question is, how do you bring it to global scale? And that's the shift that we want to take the industry through, where every vehicle is capable of driverless operation, which is clearly the steady state of where we're going to go to. You're listening to Gradient Dissent, a show about making machine learning work in the real world. And I'm your host, Lukas Biewald.

• 1:10Speaker 00

  All right, here we are at the CoreWeave house at Nvidia GTC. I just got done interviewing Alex Kendall, who's the CEO of Wave, a company that I've admired for a long time and has done phenomenally well in self-driving. And I think you could actually see him behind me messing with the CoreWeave F1 car. So this is a fun interview. I hope you enjoy it.

• 1:31Speaker 00

  Well, honored to be here with you, Alex, at GTC. This is our first podcast recording anywhere except my basement or the office. So this is pretty exciting. I have to start by asking you to tell me the story of Wave, because I think it's an especially interesting one. Yeah, actually, it's fun. I just got from giving a speech here at the event and they put my face up on the big screen in front of a fifty thousand person stadium before, like five minutes before Jensen came on. And it reminded me, because the last, I've only spoken in front of like fifty thousand people twice in my life. And the last time was at a conference called Wave.

• 2:10Speaker 00

  Web Summit back in two thousand and eighteen where we had this seed company pitch competition where I got to the the final and had to give it. I think it was like a three minute pitch in front of the crowd. And long story short, there was a crowd vote that I ended up coming last in, yet. The judges gave the competition to me because they, I don't know, they they they liked what I said, but just gives you the idea of. Ten years ago, when I started the company, what we were doing was completely contrarian, the idea that. So what Wayve started with a thesis that autonomous driving is all about looking at the. The AV problem with an AI approach. So it's a decision making, complex reasoning challenge. And we took the approach of the best way to go tackle that is with in two and deep learning.

• 2:53Speaker 00

  And a decade ago, we started working on this, building a single model that could reason and strive and generalize. In the largest scale possible without. The AV one point O, the first generation approach, which relied on retrofit sensors, compute, HD maps, infrastructure and really limited. Scaling and what's um and so we started that a decade ago when this was really laughed at and dismissed by the industry and. Having sort of quietly building it away and we're at the point now where. Not only is the market industry excited about AVs again, but we have an AI model that. We're now the first company to have driven zero shot in over five hundred cities. So this means this model is driven throughout Europe, Asia, North America. It's also driven in over ten different cars.

• 3:38Speaker 00

  From from electric vehicles to vans to SUVs. And so it's now emerging as an AI model capable of driving any vehicle anywhere. And we've got a sprint ahead of us over the next year to get this deployed in both robot taxis and consumer vehicles you can buy.

• 3:52Lukas Biewald?

  You know, I remember having a really amazing experience driving with you in London where you let me decide where I wanted the model to go. And I hadn't spent much time in London and I was astonished by how challenging it seems compared to San Francisco where. People, I think, obey. The traffic laws, as I understand it, much more thoroughly than in than in London. And there's just much more. Surprising things coming at you and I couldn't believe how well. Your model works at that time, and I think this was maybe two or three years ago. But but before we get to the the quality of your model, I. I kind of want to go back in time a little bit and ask you about. Your your history, right? You grew up in on a farm, I think, in New Zealand. Is that right?

• 4:32Speaker 00

  Yeah, in Christchurch in the South Island.

• 4:34Lukas Biewald?

  And how do you think that's kind of affected your? Leadership or you're thinking about autonomous vehicles?

• 4:40Speaker 00

  Um, I think there were a few things that were really important to me growing up on reflection. I mean, I spent half my life building things, whether it was video games or different Lego or robotics or other contraptions. I spent a lot of time messing around with that kind of stuff. And the other half was in the mountains, climbing mountains, surfing, mountain biking, all this kind of stuff. And I could wave, wave because I want riding a car to feel as good as surfing a wave. But I really think, you know, when it comes to pushing forward frontier technology, it's all about an adventure. It's all about doing something where you have to plan, you have to risk manage, you have to be ambitious, push boundaries. And

• 5:23Speaker 00

  I think there's a bit of that that I got from those experiences and those values

• 5:30Speaker 00

  that we can take forward to today because what we've been doing, it's one of those ones where when we started, this wasn't possible, but I thought and hoped and it's turned out that the future, building this technology would become possible. And I think we've been lucky that so many things have come together that make that a reality. But it really did seem like optimizing for the vexed adventure. I was fortunate enough to get a scholarship to take me from New Zealand to Cambridge University. Turning up to that place was like, like a castle from the Middle Ages, full of the most amazing brilliant minds to just learn and explore. Again, like optimizing for adventure is somewhat led me to where we are today.

• 6:08Lukas Biewald?

  Okay, you gotta tell me about building the first end to end prototype because I would imagine that would be particularly hard with the way that you're approaching the problem. You can't really cheat, right? If you're trying to do end to end machine learning. Tell me about the experience of that and how you made it work.

• 6:23Speaker 00

  So in 2017, I just finished my PhD thesis and I'd be able to build the first deep learning model for different perception tasks, localization, stereo vision, semantic segmentation, uncertainty estimation, all of these kind of systems that could take high dimensional, multi-million dimensional images and turn them into compressed representations of the world, learning end to end. At the same time, friends down the road at DeepMind had just produced AlphaGo. That's amazing. That's a system that, well, there's more states of the game of Go than there are agents in the universe. Extraordinarily complex. And they showed through self-play, you could learn, and simulation, you could learn an agent that could beat the world champion. And so I saw these things, and I thought, okay, we can now understand the complexity

• 7:07Speaker 00

  of the real world through computer vision. If you have unlimited data through a simulator in a very low-dimensional state, you can solve the hardest game there is of Go. Now it's time to bring these together into the physical world, which I've always been excited about, about physical mobility and adventure and robotics. And so I thought we could go try it. So what we did is we raised one and a half million dollars, got some friends together, we rented a house, and we put our car in the, got a car in the garage and started hacking away. And the first thing we tried was on-policy reinforcement learning, where we put the car on a road and say, okay, go drive with a reward of, drive as far as you can without any human intervention, and your reward is distance traveled. And so the car starts driving randomly, and essentially, over a period

• 7:50Speaker 00

  of months, we developed an algorithm that, a Q-learning algorithm that essentially lets you intervene and grab the wheel every time it tries to drive off the road. And the big breakthrough was the moment where with just 10 bits of intervention, we got it to learn how to lane follow. And I got that breakthrough, I remember, it was on like a Sunday or Saturday, it was on a weekend where I was pushing really hard, nothing was working, and all of a sudden I tried it and it worked. It might have been the random seed. So these kind of algorithms tended to work one out of 10 times.

• 8:17Lukas Biewald?

  Wait, let me try to understand. So you only intervened, it had no prior information about the world. It's looking at a camera, and after 10 interventions, it understands that you wanted to follow lanes and it can follow lanes?

• 8:28Speaker 00

  That's right. So the input was a front camera, there was a GPU onboard doing on-policy optimization, sorry, onboard optimization. And it was essentially, yeah, trial and error, a few different, 10 trials, and then it was able to learn a policy. There's a video on YouTube that's got this, that we muted the audio, because the audio is me whooping and cheering as it starts to drive. But it was essentially just able to lane follow quite robustly after just 10 bits of feedback with no prior knowledge of the world.

• 8:54Lukas Biewald?

  Were you afraid when you got in the car the first time and it had no prior information? Like, were you actually on a road?

• 8:58Speaker 00

  This is going slowly, so you have time to intervene before it drives off the road. So, but that brings up a good question is that that method obviously doesn't scale. You can't do on-policy tests at scale around the world. That's pretty unsafe. And so, we quickly moved to an imitation and offline reinforcement learning paradigm, and started scaling out world models so we could learn through dreaming and understanding through their experience. And so, later in 2018, that's where we started to take things. And growth of there has brought us to this point today where we can drive any vehicle anywhere all around the world.

• 9:31Lukas Biewald?

  So what happened after you got the lane following? What was kind of the next step?

• 9:35Speaker 00

  Next step was starting to scale it up. So we tried so many different things. We tried injecting priors around computer vision. We tried building world models and doing learning and dreaming and learning in imagination states. We tried imitation learning, and we tried sim to real, learning in simulation, transferring to the real world. We just sort of tried everything. We wrote all these blog posts. It ended up being industry first in all those different areas. And then, settled on what a recipe that seemed to work seemed to scale, and then started really pushing, had a couple of years of pushing infrastructure, fully build out, and actually bringing up this technology to a point where it's hard to drive in London. And retrospectively, learning in London was great because it forced us to build something

• 10:19Speaker 00

  that could deal with the chaos and complexity of your average London street. You have like a hundred pedestrians and cyclists around you at any one time. And so, it forced us to build something that didn't rely on mapping infrastructure or anything like that, but could actually drive in the city that's got 2,000 years old.

• 10:38Lukas Biewald?

  I remember 10 years ago when you were getting started, the autonomous vehicle space was super crowded with lots of different approaches, lots of VC money flowing in. It seems like it's consolidated at this point, but also it seems like it's really working. I use autopilot in my test all the time. I ride Waymos every day. I feel like it's kind of quietly snuck up and become a technology that, at least there in San Francisco, we use all the time. I'm curious how you think about the competition at this point? Is it becoming a commodity? Is there something different about what Wave can do or how you approach things?

• 11:15Speaker 00

  Yeah, that's a good question because I think broadly speaking, when we started in 2017, that was probably the peak of the hype cycle where raising one and a half million dollars to go tackle, was it like 10 companies that just announced billion dollar ambitions and deals was a bit lopsided, but it's fair to say we saw a, you know, trough of disillusionment. And I think largely that was from the sheer expense, the 100 billion plus dollars that's required to build that AV 1.0 paradigm of mapping, rules, infrastructure, and to get that to maturity. And there was a lot of consolidation there where really only, you know, one or two companies had the appetite of capital to push through and make that, make

• 11:56Speaker 00

  that work. At the same time, Tesla FSD or ourselves matured an AV 2.0 approach, a extra generation stack that's more affordable, scalable, and is now going to the point from a demo, which we had five years ago, to a point where it's actually ready for global automotive product. And I think the most exciting thing today is we're also seeing huge validation of the product. You and many others are now buying FSD and paying for it and it's, it's, it's actually loved by a lot of customers. Well, you go to Shanghai and San Francisco, places like this, and people will pay, even pay more than a, a human taxi experience for a robot taxi, even though it has longer wait times, it's limited

• 12:39Speaker 00

  by a geo fence, and, and it's, it's a product that people really love. And so now the question is, how do you bring this to any vehicle anywhere? And I think that's the step change that, that we've built and we can offer. Taking it from expensive retrofit vehicles to mass market vehicles that you can buy or manufacture for, you know, 30, 40, $50,000 each, that have inbuilt hardware that's in global supply chains, that doesn't need an HD map, so it can drive anywhere. And it's got the flexibility to be in a fleet that you ride how or in a car that you own. And so I think, and by the way, this, this, this ambition, this business model is only available because we've built this generalizable AI driver. But I think now the question is, how do you bring it to global scale? And

• 13:22Speaker 00

  that's the, the shift that we want to take the industry through. But for me, I'm thrilled that we're now seeing, what are the conditions in the market? So I mentioned that people are now paying and there's real commercial validation of this as a product. Regulation is in place, not just in the US, but we work really closely with the UK, Europe and Japan. We co-chair the UN Committee on Decas Regulation for Autonomy Systems. That later this year is now going to come through and actually legalize these products.

• 13:56Speaker 00

  And then, of course, manufacturers are now building vehicles at millions of scale volume that have centralized compute, surround sensors and the ability to over their update and get data off them, which makes a fleet learning product possible. So all of these factors together give me a lot of optimism for going through a massive commercial inflection point in the coming year.

• 14:15Lukas Biewald?

  It's kind of interesting you didn't mention in that list, improving the algorithm. And it's very hard to judge the quality of an algorithm or the safety of a self-driving car when you're in it. But I have to say, when I was inside of your car, and this is now, I think, two years ago, it really felt ready for prime time. I think it felt safer than the cruise cars that were on the street in San Francisco, which I don't know if I should trust my own judgment or not. But are there still things to do with the algorithm to get it to the point where you feel good about using it globally? Or is that no longer the challenge and you've kind of moved on to logistics and regulation and things like that?

• 14:53Speaker 00

  Absolutely, there are still things to do. We don't yet have a driverless service today. But this is no longer the critical path. We are seeing that the quality of innovation coming from our frontier embodied AI science team, plus the data and compute growth that we have is just driving compounding scale. And then you layer on top of deeper integration into vehicles with our partners, better sensing and compute on the vehicles. These things compound that performance is no longer the critical path, but we still have work to do there.

• 15:24Speaker 00

  And so we'll keep pushing that and pushing it into a generalized state.

• 15:29Speaker 00

  I think. A large challenge is evaluation and validation of it. How do you prove the level of performance is sufficient to yourselves, to regulators, to consumers? And then, of course, commercially deploying and setting up. I mean, the last couple of years for us has been a big era of growth of the company to build strength of team to partner with regulators and automotive and fleets in major markets like Germany, Japan, and the US. It's to get the integration going. It's the multi-year sales cycle of automotive. And now it's going to be the deployment and bring up of these systems. And so I think that's where the critical path is. But we're not taking the foot off the accelerator in terms of performance, which clearly has orders of magnitude still to go in terms of opportunity ahead.

• 16:14Lukas Biewald?

  One of my favorite demos I've ever seen was someone on your team who showed me plugging your perception model into an LLM so that you could actually see what the car was thinking as it was driving, which seemed surprising and delightful. Was that a toy thing that you realized you didn't need? Or is that something that you've continued to invest in?

• 16:36Speaker 00

  Yeah, you know what? In 2021 or 22, like way before ChatGPT, one of my team, Vijay, came to me and said, "Hey, Alex, I want to start working on language for driving." And my first reaction was, "No way, man." This is, "Go stay focused. Whenever product ships, come on, stay focused on driving." And he laid out the argument that actually the knowledge you can get from text data is enormous and it might produce new interesting products and interaction modalities. And actually, by bringing in language, we can improve the representation, the performance, and the product experience. And so we actually said, "Okay, let's start exploring this." And so very quickly, we got up a prototype that it actually wasn't plugging together

• 17:19Speaker 00

  an LLM because LLMs didn't really exist in that large scale yet. It was more, we trained a vision language action model. We trained a single model that could see the world, drive a car, and understand language. And so it started off as a gimmick. You know, you could drive along and have it explain its driving. You know, "I'm going around a double-parked bus. I'm stopping for a red light," and we'd just explain it. And, you know, you could cherry pick some really delightful things. But then again, there were also some things where it was just hallucinating and wrong. But from that point, we started to, you know, train it up. We actually got it to a point where it would become very robust. We saw that There's actually one interesting thing I saw, which is if you drive and you have an oncoming car and you want to make a turn across that oncoming car to go into a side

• 18:04Speaker 00

  street, if that oncoming car stops and flashes its lights at you, it's a signal for you to go in front, right? Even though you don't have right of way. Our AI wouldn't do that. It would stop and yield and just wait for that car because it had right of way. When we started bringing language pre-training into the VLA model, it actually demonstrated that behavior. When it started flashing its lights, it would now turn. I saw that for the first time. And so that was an example of we could actually improve the representation and reasoning capabilities by bringing in language into the system. And so today, it's a core part of our training. It gives us boost in performance. It's opening up new interpretability capabilities and of course opens up product experiences of personalization

• 18:45Speaker 00

  and interaction with the vehicle. You can now get in our car and prompt it to drive in different driving styles. And I think that'll be a neat experience because when you get into an ADAS or a robot taxi, driver assistance or a robot taxi car, if that car is driving too conservatively, you're going to get really frustrated. Or if it's driving too fast for you, you're going to get freaked out. And different people have quite wide variety and different expectations there, or different brands. You can go out some brands that want to be sporty, or some brands that want to be safe and reliable. And not to say that sporty won't be safe and reliable, but you know what I mean. And so providing that personalization is quite a nice outcome of this work.

• 19:23Lukas Biewald?

  It's a funny moment for self-driving, at least in San Francisco, where we suddenly have Waymos everywhere. And I love it. I think it's an amazing experience. But we're also witnessing all kinds of different crazy traffic issues that are being caused. And also, people, maybe for fun, or maybe maliciously, kind of messing with them in ways that you wouldn't have expected. So it does seem like that capability might actually be completely critical to making a self-driving car assisting people trust.

• 19:50Speaker 00

  Yes, although a couple of things. Firstly, our cars, you know, over time, you won't notice anything different because they're just normal cars with built-in sensors. So I hope that people don't behave differently around them compared to other traffic.

• 20:04Speaker 00

  And then secondly, our system is designed to drive in a very human-like way. The way you see it picking up and dropping off as a robotaxi experience in busy cities, or the way you see it nudging through crowds of pedestrians and doing things that are very human-like and not just staying stuck, I think these kind of things are really going to improve public acceptance. We designed the system to operate in the messy cities we live in and not require infrastructure or behavior change. I think that's really important for adoption of embodied AI. You need to use mass market hardware. You need to be able to leverage availability of off-policy data, and you need to be able to operate as a drop-in to existing infrastructure, not requiring massive cap exchanges. I think if you get these things right, then embodied AI is a delightful deployment experience.

• 20:48Lukas Biewald?

  How did you make the decision to not build your own car and to license your technology to other car companies?

• 20:57Speaker 00

  That was very clear to us from the beginning. I think the mindset I've always and we've always had as a company is that self-driving is an AI problem. We want to work on the hardest problem first. So we started working on AI and we started aggressively partnering on cloud, on car, on infrastructure, on insurance, and working with the best companies around us so that we can focus on the critical path problem. And by always focusing on the hardest problem first, it's kind of, you know, not let us stumble into some glass ceiling that we hit down the road. So I think it was that mindset. But then retrospectively, it's become important because I think it's a mistake to focus on one vehicle form factor because it's

• 21:39Speaker 00

  really unclear what will have product-market fit. Is it a normal SUV like today? Is it a bidirectional vehicle where you sit facing each other? Is it a two-seater electric vehicle? Or are there other applications that really take off? And I think that's unclear. But at the end of the day, if you focus on one form factor, you're also going to have a small part of the market. And so we've tried to focus on how do we build a generalizable driver that can address everything and then follow where the market is most advanced at any one stage. We started off with retrofit when there wasn't really anything for us to integrate into. Then during COVID, grocery and fleets started really skyrocketing in profits since we started partnering with that sector.

• 22:23Speaker 00

  Then things changed, but then automotive, who would never speak to us, they said this is unsafe, it would never work. All of a sudden started talking to us and building software to find vehicles so we could integrate with them. And now today, we've got the excitement of robot taxis. So we've kind of followed the market, but been open to integrate into anything. And the end state, of course, is consumer vehicles, delivery, robot taxi, trucking, non-automotive forms of robotics. We want to license our embodied AI and make all of them intelligent, safe and possible.

• 22:53Lukas Biewald?

  Where you sit right now as CEO, do you have like a stack ranking of those different modalities and which ones most top of mind for you?

• 23:01Speaker 00

  Yeah, today it's all about consumer vehicles and robot taxis. It's about point-to-point driving in urban and highway situations with mass market consumer vehicle hardware. So this is an aligned product that we can now commercialize with great partners. So this is the real focus for us today that can generate really extraordinary business. I mean, you look at the scale of, you know, whether it's data, revenue, number of vehicles, exposure, about 100 billion cars produced each year. And people forget that often they only think of robot taxis, but there's less than 10,000 robot taxis in the world. And so getting this natively integrated into consumer vehicles opens up tens of millions of vehicle opportunity. That's going to be far outscale robot taxis in the short and medium term.

• 23:45Speaker 00

  But for long term, of course, every vehicle is capable of driverless operation, which is clearly the steady state of where we're going to go to, given the safety and convenience and product benefits it can bring.

• 23:56Lukas Biewald?

  What geographies do you plan to launch in first?

• 24:00Speaker 00

  I mean, we're really looking to make sure the system is global, but we're focused today on Europe, including the UK, as well as Japan and North America. So we have fleets and offices in London, Stuttgart, Tokyo and the Bay Area. We've driven around most cities throughout Europe and North America and growing in Japan. And those are the markets we're starting with, given the customers we work with.

• 24:25Lukas Biewald?

  Do you have a sense to where you will go mass market first?

• 24:29Speaker 00

  It's those markets. I think in general, we're seeing faster movement in North America and Japan, closely followed by Europe. But of course, China is, unfortunately, it's very hard for us to work in China and vice versa. I won't get into geopolitics, but it's a bit of a shame. That's where the world's at. But in China, this technology is moving so quickly and the rate of innovation is very impressive. And so I think it's worth looking at if you haven't from an automotive technology perspective. Another proof point that when this technology is deployed, consumers really love it. And it's now alongside price and safety, it's now a top three reason why people buy a car in China now.

• 25:08Lukas Biewald?

  Wow.

• 25:09Speaker 00

  Is autonomy.

• 25:10Lukas Biewald?

  So you had amazing results at the AI 500 roadshow last year. I don't know if you want to talk about it, but you took one single AI model, deployed all over the place with, I think many of those geographies actually not having any prior training data on it. Were you surprised that it worked so well?

• 25:26Speaker 00

  No, because that was the core thesis, that by building a general surface AI model, we could drive anywhere. But to actually see it was huge validation. So yeah, we drove around the world, some very cool anecdotes. I mean, we drove in places like around the Arc de Triomphe in Paris. Have you driven around there?

• 25:41Lukas Biewald?

  I have, yeah. Well, not driven, but I've gone there, yeah.

• 25:44Speaker 00

  It's chaos, right? We're there. We went north of the Arctic Circle in Finland and Sweden and tested driving in 22-hour darkness and snow and things like that. We were even in Tokyo during a typhoon, where local trains and buses shut down. And actually, that day, we were giving a bunch of media drives. We had a day long of journalist drives, no disengagements, perfect autonomy the whole day in literally the heaviest rain I've ever seen. And our cameras are situated just behind the windshield, the front camera. Windscreen wiper was just constantly wiping water off it. But it drove reliably around central Tokyo, which was quite remarkable. Less pedestrians on the road that day, but still busy, busy traffic. But yeah, we've seen all of these things in that. And it demonstrated that the first company to do something of that scale.

• 26:30Speaker 00

  And it demonstrated that we truly can generalize to any vehicle, anywhere. Given that over half of these cities, we had no prior training data in. And so it was true zero shot generalization test at scale.

• 26:44Lukas Biewald?

  That almost sounds like superhuman level performance.

• 26:48Speaker 00

  Yeah, I'd be, I'd be careful with that. I mean, you and I, if we jump in a new city, you can get a rental car and go drive there. But to claim superhuman, I mean, there's some great reports that I read one from Swiss Re, for example, the insurer, where they measured the impact of an autonomy system of Waymo operating of their operational history and found that they had much fewer accidents and safety events than the equivalent human drivers. So I think there's some robust evidence that's really showing autonomy can be superhuman.

• 27:22Speaker 00

  We have work to do to demonstrate that, but we're making good progress there. But then you also look at sadly, over 99% of accidents are caused due to human error. And so even just making a self-driving experience that can never be distracted, drunk, compared, that can see 360 degrees or once make decisions over 10 times a second. I mean, there is an opportunity there to drive accidents down to near zero. And that's what we want to see. But I think you've got to see scale deployment to really show that. And we want to go do that in the coming years.

• 27:53Lukas Biewald?

  Are there situations that still make you nervous at this moment? What would be the kind of configuration that would put you on the edge of what the model can do?

• 28:04Speaker 00

  I think it's very hard to give general statements to that because where the models today tend to struggle is when you have a culmination of many failure modes. If it's, you know, dark, you have high-speed traffic, you have adversarial behavior for someone else driving on the wrong side of the road or not yielding. Maybe it's bad weather, you know, it's when you have multiple confounding things, then maybe you have a sense of failure. When you get multiple confounding factors, that's when you tend to find challenges and failure modes. And so we need to keep addressing that with better system resiliency or redundancy. We're building that with some next generation vehicle architectures.

• 28:45Speaker 00

  We need more data to improve the generalization of risk assessment and yeah, making sure we deploy responsibly in areas that we're confident meet, you know, meet the required bar for safety. But we'll do that responsibly with our partners as we grow. And I think the other nice thing about being deployed in consumer vehicles and robot taxis, starting with supervised trials, is that we can incrementally grow this with communities and regulators to the end state where every vehicle is autonomous at a global scale.

• 29:14Lukas Biewald?

  I remember there was a moment in sort of ethics and AI, maybe when AI was like, when self-driving was hot, that people were asking questions, maybe toy questions about should a car prioritize the safety of the passenger versus the pedestrian? And you know, and end to end, you're kind of sidestepping those questions or maybe you're not inserting yourself. Is that true or do you need to push the date in certain ways to get certain types of ethical behavior that you want?

• 29:42Speaker 00

  So this is the trolley problem where you have to, where you've got two bad decisions and you have to choose bad decision A or bad decision B. There's many ways of framing it. The practical reality is, as an industry, and I think I speak on behalf of most AV companies for this, is that

• 29:58Speaker 00

  We need to design a system so there's always a third good option. It's called a minimal risk maneuver, which is if you're in a bad scenario, you try to, and you see risk, there's always a third option you can take, which is the minimized risk. And that typically involves reducing kinetic energy, maintaining a predictable course of action, and stopping or pulling over the vehicle or something like this. And we always need to be prepared to do that. If any systems fail, there needs to be a fail operational ability to do that. And so we design these systems to always have a third good choice, not to be stuck between two bad choices. The second point is that to get into that two bad choice state, so many things have to go wrong to get to that point. And the probability of those kind of scenarios is so small. And even if you are in that position,

• 30:42Speaker 00

  the probability that you can actually accurately sense in those scenarios, that those two things, it makes this more of a theoretical thought exercise, an actual practical decision the industry needs to take.

• 30:55Speaker 00

  So we focus on, and the industry focuses on building these minimal risk maneuvers that take us away from those situations. Having said that, if you do get into that unlikely situation where you're in there, then of course, the behavior will depend on how you bias it from your training data and the system that you design. And so there, of course, we need to be thoughtful with the principles and the knowledge that we distill into the system and the behavior that we generate. I mean, there are some good principles there around managing risk, making the system predictable, and of course, maximizing safety.

• 31:31Lukas Biewald?

  But I would imagine that in your end-to-end training, you're not seeing a lot of crashes. So if there are crashes, you're probably generating them in the training data somehow. So I think that the types of crashes that you're feeding into your system would give it a certain amount of point of view.

• 31:48Speaker 00

  I'm really proud of our safety record, actually. You know, we've had no incidents since we started operating in 2018. So really fantastic safety record around the world. And essentially...

• 32:04Speaker 00

  There's a few things that are important here. Firstly, we don't just learn from the data of our vehicles driving. We have enormous data partnerships that give us quite general purpose knowledge. Unfortunately, there are still a lot of road accidents today and we partner with dash cam providers, car companies that get data from their consumer vehicles. And sadly, there is still a lot of quite frightening scenarios that they experience. The good thing is, is we can use that data to learn from it. Then secondly, we have our world model Gaia that can simulate these scenarios that are too rare or unsafe to see in the real world and make sure that we're robust to them through synthetic data and simulation too.

• 32:40Speaker 00

  And so whether it's those strategies, ultimately, we need to build robustness to those events. And we still see some pretty bizarre stuff on the roads around the world. Unfortunately, some pretty unsafe behavior and we can resimulate or augment or modify these scenes to, of course, to build resiliency to them. And it's a constant game of fleet learning and data iteration to make sure we're grinding up performance over time.

• 33:08Lukas Biewald?

  So I remember when Tesla launched, I believe what they said was end to end train, full self-driving. You really felt it, I think, when that version came out, at least in my Tesla. And one of the things that was really notable about it was it really stopped every stop sign in San Francisco, which actually San Francisco drivers don't really do. So it's kind of annoying if you're really like following that rule to a T.

• 33:33Lukas Biewald?

  And it made me think, you know, actually, the average of everybody's driving is probably, you know, not what you want. How do you deal with that in your end to end training?

• 33:42Speaker 00

  It's a great question. I was just in Japan last week and there, everyone drives about 20% above the speed limit. And it opens up an interesting question around policy, regulation, liability. You know, if you're a eyes off or driverless system where the system's taking liability for driving, it should meet the road rules. But if you're a consumer, a lot of products today in a driver assistance setting allow you to set the set speed above the speed limit. And so it's a really interesting question. I hope that self-driving improves road safety to the point where you can actually increase speed limits because it's safe enough to do so. But we're probably some, you know, some distance away from being able to have those kind of arguments. Yeah, so today.

• 34:25Speaker 00

  The first thing to realize is when you are training a world model is you want to see as diverse data as possible. You want to see good driving, bad driving. You want to really understand the full spectrum of dynamic events in the world. Then when it comes to actual policy learning of how you drive, of course, there a data distribution that matches whichever customer of ours preferences, the OEM driving style behavior, the rules and policy of the country they're operating in becomes really important. And so we can separate those two data distributions. The first one, you want to be as diverse as possible, as large as possible. The second one can be a highly curated, small, whether it's RLHF style feedback, whether it's expert demonstrations.

• 35:08Speaker 00

  There's many ways to do it to make sure that you drive in a very safe and considered way for that application. So you can separate these data distributions.

• 35:18Lukas Biewald?

  Actually, one thing I wanted to ask is someone as a kind of fellow entrepreneurs. You started, I think, as CTO with another co-founder who then left and he became CEO. I wonder what that experience was like. I guess it was earlier in the journey, but you wanted to share about that.

• 35:33Speaker 00

  Yeah, it was early on where kind of stage, I'm sure you remember where everyone does a bit of everything.

• 35:40Speaker 00

  And so as we're getting started, there was a bit of that sense. But yeah, for a number of reasons, it unfortunately made sense for my co-founder, you know, to leave the company and pursue other things.

• 35:54Speaker 00

  And we're still in good touch today. But that transition was a my first, you know, real experience of like change management communications of the company. I think we're about 25 or 30 people at the time.

• 36:09Speaker 00

  You know, just raised Series A. But also it's, I guess, matched my experience over the last decade, where every year has been a completely new challenge. Okay, we started. Um, you know, I started the code repo and was coding a lot of the first demos. Then I started to work on like hiring the team, finding commercial partners, growing investments. You know, now managing customers and managing through an executive team where I'm so thrilled to be working with some people that are truly at the top of their game, they're industry legends. You know, whether it's in science, engineering, product, uh, you know, people and finance, uh, commercial across the board. We've got truly world-class leaders.

• 36:53Speaker 00

  And so it's taught me a few things is the importance of building complementary teams, working with people that you can learn from. Uh, and also getting used to as soon as I get good at a skill, uh, pretty quickly, uh, you know, our growing it and having a new challenge as soon as I get good at something, it's like no longer the relevant skill for me to be working on because, um, it's usually me plugging a gap for the organization and then growing the organization to absorb it. And then there's a new challenge. I enjoy that though. I think that's part of the original theme is part of the adventure of building a company. Um, and, uh, I think the next, the next year for me is now I'm like, this, this phase is new one where I'm now starting to live on a plane, work with global customers.

• 37:37Speaker 00

  Uh, it's going to be our first 10 years in, it's going to be our first, uh, product launch. Um, we're going to start to have major public and regular exposure through that. Uh, while keeping up strong and consistent execution velocity. Uh, so this is a completely new challenge that I'm, I'm really quite excited for. Um, and let's see what comes after that.

• 37:57Lukas Biewald?

  If you could send a 30 second message of advice back to yourself when you're just starting out, what would you put in there?

• 38:03Speaker 00

  Oh man, I think, I feel like if I built wave again, I don't know. If, if you build things again, how much faster do you think?

• 38:13Speaker 00

  You could build second time over.

• 38:14Lukas Biewald?

  Oh, totally.

• 38:15Speaker 00

  It's like two or three X, right?

• 38:17Lukas Biewald?

  Yeah.

• 38:20Speaker 00

  There's so many shortcuts or things that I was hesitant to doing.

• 38:25Speaker 00

  Or mistakes that we made that I think I could just give the tactical playbook.

• 38:30Lukas Biewald?

  What if you had to distill it at a higher level because you only have that 30 second window to?

• 38:34Speaker 00

  Not just give the loss function recipe and relationships and partnerships, like just really tactical. I always thought the interesting question is, I feel like we'd be a bit, you know, useless if, if, if we went back in time like 200 years with today's knowledge, what would you do? Because I don't know what you've learned through weights and biases, but you're built on top of stuff that just, we're standing on the shoulders of giants that just did not exist two hundred years ago. And so there's a bit about timing, you know, in many ways, our product, if I've developed three times as fast, we may have been ahead of the industry where automotive did not have the right infrastructure or products to actually accept our product. So I think timing is everything.

• 39:14Speaker 00

  And in many ways, self-driving has been a question of like staying alive and keeping the company capitalized while making progress on our next generation approach. But yeah, if I was to distill it down.

• 39:28Speaker 00

  I've always had conviction in this approach. So it's not like I'd say, you know, keep going, Alex, because I've been kind of resilient towards that one.

• 39:39Speaker 00

  So I think, yeah, I think I've been fine with how bold and brave I've been in, in, in, in building this.

• 39:48Speaker 00

  Honestly, the biggest thing has been, I think.

• 39:54Speaker 00

  Yeah, I think.

• 39:57Speaker 00

  I think two things actually genuinely come to mind. One is about.

• 40:03Speaker 00

  Talent growth at Wave. And I think I've seen so many great things about great people, leading to fantastic people, leading to brilliant people and growth of the expertise we have in the company. That I think moving, you know, faster and harder and, you know, I guess, starting.

• 40:20Speaker 00

  Out of a PhD, you know, growing from London compared to the scale of the amazing people we can work with today. I think there's a growth journey there that that that I could have leant more into. And then the second one is.

• 40:33Lukas Biewald?

  So you mean like try to get the experts earlier?

• 40:36Speaker 00

  Yeah, yeah, I think so. Um, you know, we started off with general like with a generalist mindset, really trying to figure things out and had to build a lot ourselves. But I think.

• 40:47Speaker 00

  I think maybe there's a mixed mixed feeling on that one, but um. I think the the better, the the the more bigger thing that comes to mind for me is. I think I over indexed on algorithmic innovation compared to infrastructure. And like the big learning I have with deep learning is that it's one percent algorithms, ninety nine percent infra. Whether it's cleanliness of your data, reliability and iteration speed of your of your learning loop. We're very happy adopters of weights and biases over the years, the acceleration you get with it.

• 41:23Speaker 00

  Whether it's introspection, measurement, evaluation tools. I think we've been on a journey there and under invested on it. I think going harder and faster there, rather than the shiny sexy algorithmic innovation, is something that I wish we'd done earlier. The thing is, is we were held back by a robot platform for many years. And as soon as we had a robust robot platform, it's like giving our AI. Being blind and giving it glasses, all of a sudden it could see and drive in a remarkable way. So focusing on that earlier and harder and building that maturity, I think. It was quite, quite key to bringing up performance. And we were too late in that. In simulation and evaluation, I think both large language models and embodied AI has a real challenge in evaluating the performance of models. It's a.

• 42:08Speaker 00

  Opening a problem. It's not like um you could have a a clear metric around that today. Uh, and so developing these techniques, we've been on a journey. We've consistently tried different things from generative models to procedural game engines. To Nerf and Gaussian spitting to now um generative world models. And throughout those those times, we haven't been afraid to. Forgo sunk costs and you know, we spent millions of dollars, months of time building something and then just delete it, move on because we found something better. And so that, that, that culture and that journey, I think has been important. And today, I think we have something on the cusp of a major breakthrough there that will really transform the industry. But that's another example where getting that right earlier, it's chicken and egg.

• 42:51Speaker 00

  If you figure out how to measure it, then you can just optimize against that target. But getting that right is something that we should have moved faster on.

• 42:59Lukas Biewald?

  Okay, some quick questions to end with. When you took Bill Gates for fish and chips in your self-driving car, were you nervous that something would happen? And what did you talk about?

• 43:09Speaker 00

  Yeah, that was one of the first times we got the system reliable enough. And actually, there were a few interventions on that drive. So Bill was still excited at the time. We talked about, it's interesting how when he was commercializing Microsoft and Windows, he had to get a bunch of design wins with PC OEMs. And in a similar way, we've had to get design wins with the biggest car companies in the world. So there's some interesting similar dynamics there that he was interested in. And then he was in London for a day, wanted to get fish and chips. One of his team went in to get the fish and chips, handed it through the window to us in the car and he just put down and sat there and wasn't going to eat it. I said, come on, Bill, we've got to, got to heat it. So we've got to, come on, we can't, we can't not eat the fish and chips.

• 43:52Speaker 00

  So he goes, all right then, grabs a piece of fish in his hand, brings it up, breaks in half and just starts eating this, this battered piece of fish in the backseat of the car with me, which, which I love. He, he clearly has a soft spot for some greasy fish and chips.

• 44:04Lukas Biewald?

  Oh, nice, nice. All right. What about when you built the first prototype in your rented house and you said, you threw on the biggest parties of your life? What was that like?

• 44:15Speaker 00

  We invited everyone we knew were in Cambridge to the house. We, you know, the living room was where we worked with a bunch of computers. The small bedroom was our boardroom. We had another bedroom full of servers and it was like constantly 40 degrees Celsius because of the 60 GPUs that were running there. And that we just sent the house into a bit of a party house. We've got a few musicians in the team. We were jamming and that was, that was a really, really fun vibe. Yeah, those days are very special and you know, I, I really enjoy the lasting memories of a team that we were really in the trenches together during that time. But that was, that was quite a big one.

• 44:50Lukas Biewald?

  I don't know about you, but I feel like I'm always chasing that feeling of the early days when everyone's cranking together. Such a good feeling.

• 44:56Speaker 00

  I just finished out, I just visited our Japanese office last week, which is about, you know, 20, 30 people. Really feels like that. That's super exciting. And then, um, I think, yeah, I think there are pros and cons to operating at a different level of scale. You know, today we're about a thousand people globally. Um, but the culture you set at that scale really drives through where we are today. And, um, I think it does require different skills to operate at this level of scale, but the impact you can generate is just tremendous. Uh, but holding on to disruptive innovation, not falling for the innovator's dilemma, you know, these kind of, uh, principles are still really things we really fight for and I think still hold true today at Wave.

• 45:36Lukas Biewald?

  Well, congratulations on your success. It's been a joy watching you guys grow.

• 45:39Speaker 00

  Thanks, Lucas.

• 45:40Lukas Biewald?

  Appreciate it.

• 45:41Speaker 00

  Chris, yesterday.

• 45:42Lukas Biewald?

  Thanks so much for listening to this episode of Grading Descent. Please stay tuned for future episodes.