Building an Autonomous Delivery Experience with DoorDash Co-Founders Andy Fang and Stanley Tang

Yeah, we've actually been looking into robotics autonomy probably much longer than people thought, like since 2018 actually. Back when it wasn't obvious autonomy and robotics was going to be a thing. But we felt like this was going to be a technology that was going to be transformative to our space and potentially disruptive. And I think that's the nice thing about being a founder-led company is we get to think about much more future speculative things that are on the horizon and constantly think about how do we make sure we don't get disrupted by the next... I think like Andy said, the next DoorDash if it comes along is not going to be someone that builds the exact same version of DoorDash but maybe with a better UI. That would be dumb. It's going to be something like how do we incorporate AI agent to commerce, how do we incorporate autonomy, robotics, drone deliveries, etc. And I think fast forward seven, eight years later, you're seeing everything starting to play out in AI and robotics and autonomy. Waymo is happening. And I think we're glad we made that investment early on in 2018.

Yeah, I think it's probably the same of how we invest in a lot of things at DoorDash. Everything starts out as experiments. I mean in a way that was the founding story behind DoorDash. DoorDash was a Stanford college dorm room experiment. It started out as a website called politely.com with eight PDF menus and a Google Voice phone number. And it was only once we figured out, okay, there's something here. Let's turn this into a company. And we've kind of taken that philosophy throughout the past 13 years and applied it to autonomy as well, AI as well. I mean, when we first started in 2018, the intention wasn't to spin up this giant robotics program. Let's hire a roboticist, go build hardware. It was really me and half an engineer's time. It was a skunkworks project. It was an experimentation to go explore what's out there. We don't even know what autonomy looks like, how robotics is going to impact our space, but let's go explore. Let's go form partnerships. Let's go learn. Let's go experiment. And I think in the beginning, the intention wasn't to build our own robot. Actually, we didn't think we needed to build any of this technology.

We thought we can just partner up with a bunch of folks. Back then we didn't know anything about robotics. There's all these startups out there that have built robots and autonomy. Why don't we just work with them? We can essentially just be the platform. We'll build the APIs, we'll handle the distribution, etc. And we did that for several years. We worked with everyone in the space, from the sidewalk robot players all the way up to the robo-taxi players. I'll say there's three things we learned through that experience. One is it kind of validated or confirmed our belief that there's something here. Autonomy is a question of when it was going to happen, not if. And again, fast forward today, you see Waymo driving, right? It's happening. So we should keep investing. The second is I think it allowed us to learn what it takes to actually enable autonomy, because it turns out there's a lot of things you have to build around autonomy: the infrastructure, the ecosystem. How does autonomy integrate with DoorDash? What deliveries you take on? The operational aspect. It turns out a lot of things you have to build around autonomy in order to make autonomy possible.

It's not just you plop a robot in or even AI, just plop an LLM in and then things magically happen. There's a lot of things around it and you have to build a platform ecosystem. So one of the things we ended up building is this thing called the autonomous delivery platform. Essentially, it's what are all the products and technology, the APIs, the dispatch, you need to build in a post-autonomy world where autonomy and robotics and drones are everywhere. What are all the things you have to build? How do you integrate with merchants? What does the consumer experience look like? And I think the last thing, which is probably the most important thing we learned, which eventually led us to realize we had to build this technology ourselves, is really this idea of building towards a use case.

Yes. There's a lot of autonomy startups out there, but we always felt like these companies weren't really focused on a use case. They always felt like they build the technology first and then retroactively try to go find a problem to fit into. These things were all built in a vacuum. Which is kind of weird because in software world, when we went through YC, we're always taught to serve the customer, build something people want. That's kind of drilled into you and then you can iterate. But when it comes to hardware and hard tech and AI and robotics, people just kind of do the opposite. They try to build the tech first and not really think about the use case they're building towards. And whenever that happens, you just end up with something that just wasn't quite the right fit. We went through this process where a lot of these companies, it always felt like it wasn't exactly what DoorDash needed.

For example, in the autonomous world, there's basically two buckets of companies. You have these sidewalk robot companies which are kind of these two or three mile per hour water cooler on wheels, super effective simple technology. But we quickly realized the speed and distance was a huge limitation. The average delivery at DoorDash is about 3 to 5 miles, and the typical delivery time is 15 minutes if you exclude the time to make the food. So if you put a 2 mph sidewalk robot, it's just never going to work. And on the other end of the spectrum, you have the robo-taxi players which are designed for carrying people around. It's a 4,000-pound vehicle, goes super fast, you're transporting people. But the problem for carrying people and carrying goods is actually a little bit different. If you're only carrying a couple burritos, do you really need a 4,000-pound car with chairs and AC? The pickup and drop-off problem is also very different in robo-taxis. You can walk to a Waymo. How often have you taken a Waymo where it drops you off half a block or a block away from where you need to be, which is totally fine because you can walk, but packages can't do that. How do you solve that? I call it the first and last 100 feet problem. How does the food get picked up at the merchant? What does that integration look like? And on the customer side, how do you drop off the food? How do you find the driveway? People expect their food to be dropped off or the vehicle to be pulled up straight to the front of their driveway or their porch.

So when we looked around and asked ourselves, if we were to start from first principles, and again this has always been our philosophy at DoorDash, if you start from the business, the customer use case, work your way back, first principles, and build exactly what we need to solve our use case, what would that look like? We looked around and no one's really building that. It's not a sidewalk robot, it's not a robo-taxi. We felt like the right metaphor for us was something in between. If you're trying to solve that 3 to 5 mile delivery in dense suburb, which is where most deliveries happen, the right metaphor is probably an autonomous motorcycle or scooter or bike profile vehicle. It doesn't need to be 4,000 pounds. It's probably 300 pounds, but also has to be a lot faster than a sidewalk robot, let's go 20-25 mph. And when we saw no one's building that, we decided, well, if no one's going to do that, instead of waiting around and waiting for this to happen, we're going to control our own destiny. Let's invest in this and see what we can build. And it took many iterations. We started testing this with real DoorDash deliveries, looking at our 10 billion deliveries we've done, extracting the insights and operational learnings, and that eventually led us to launch and ship Dot, which is our in-house autonomous delivery robot. So it's been quite a journey, but again this is something we look to bring to every aspect of the business. Whether it's autonomy, robotics, AI, it always starts out as experiments. It always starts out as what is the customer problem you're solving for? What's the use case? Work your way backwards, then iterate and validate your hypothesis and slowly build the product over time.

I Yeah, I agree with you there. I mean that's basically your methodology to building the dot form factor. I think maybe that approach works in software land, but for a business like ours, DoorDash is a physical world business. You bring technology into the physical world, and the physical world is always a lot messier, a lot more complicated, a lot more nuanced. I think one of the things people don't realize is just how complicated DoorDash is. I mean, we do over 3 billion deliveries a year. There are no two deliveries that look the same. All three billion deliveries look different. And they all come in all sorts of shapes and sizes and different geographies. A delivery in downtown San Francisco is completely different than a delivery in Dallas or even in Europe or in Helsinki where it's snowing. A pizza is very different than ice cream, your dinner is very different than your grocery order, which is very different now that we're expanding to retail and pharmacy and parcels as well. The diversity of deliveries that happen at DoorDash is so complex that I think people sometimes don't realize just how nuanced the problem is. And that's what we have to solve for at DoorDash. And that's been the learning process, especially when it comes to building autonomy or even AI: how do you manage through all that complexity? And again it always comes down to understanding the use case. I think we just have such a huge advantage over everyone else because we have something that everyone else doesn't have: it's called DoorDash. We have 10 billion deliveries of data to extract from. We have over 40 million consumers ordering every single month. We understand the complexities of how to handle when things go wrong, how to integrate across all different types of merchants. The way you work with a McDonald's or Starbucks is very different than working with a mom and pop sandwich shop. A drive-thru restaurant is very different than a restaurant at a strip mall or downtown Main Street. And how do you handle those different use cases? Different interaction, different pickup points. I don't know if there's anything you want to add on the AI side. I mean, for me, the analogy you brought up, I think about it in terms of the autonomy thing, but I also think about how the humanoid robotics space is starting to play out potentially. We also launched a product called Tasks a couple months ago where we're having people in the Dash fleet help collect data points to help train some of these world models. And I think we're so early there, and there's so many different form factors that you can use, and there's different opinions on what type of model is going to work versus not. But I think unlike something like ChatGPT, there's a lot of expense needed to invest in just the V1 of this. I guess ChatGPT could cost a lot of money too, but I think there's a lot of pressure to figure out how to actually provide value. I have to be better than what people can do today. And whether it's Dot and delivering something end to end, or investing in a bunch of different players in the space, there's real pressure to be better than the alternative from either a quality or a cost perspective. So yeah.

Yeah. Exactly.

Yeah. So DoorDash Dot is an autonomous delivery robot. It's built entirely in-house at DoorDash. It weighs 300 pounds, travels up to 20 mph. It's about one-tenth the size of a car. It's the only delivery robot out there that's designed to travel not just on sidewalks, but also go on bike lanes and on the road as well. It's live in Phoenix. We've been live doing deliveries for almost two years now. It's fully autonomous L4. So if you come to Phoenix in Tempe, it really feels like Waymo in San Francisco.

Yeah. No, exactly. And I think that's again where DoorDash gets to shine with our advantage. We don't necessarily have to solve for 100% of our use cases. That was part of the learning from our early years when we did the partnerships for how we built our autonomous delivery platform. It was understanding what kind of deliveries fit into what modality. And I think the vision was always let's not design something to solve for everything, but instead come up with a multimodal strategy where perhaps you have DoorDash Dot do the 3 to 5 mile suburban deliveries from a strip mall. Right now we're live in Phoenix, that's our starting point with Dot. That's the perfect market for Dot: dense suburbs yet things are still far apart enough. Maybe if it's a rural area with poor road infrastructure, you send a drone delivery for a lightweight order. If it's a complicated multi-step grocery order where we have to climb stairs and pick and pack orders, you're still going to have a dasher for that. And I think that's the nice thing about DoorDash: you don't have an all or nothing approach. You can phase in these modalities over time and pick and choose what the right use cases are to solve, what modalities fit into each use case, like are there certain deliveries we can carve out that make a lot of sense for robotics versus humans.

Exactly. And then from the consumer side and the merchant side it's like the exact same experience. Still the same app for the customer that you can access everything. And then for the merchant, it's just one integration. Uh you already integrated Door Dash. All of a sudden, you get not just Dashers, but you get drones, you get autonomy, you know, you had access to all the, you know, AI tools and products every ship. And I think again it's like that is what ultimately Door Dash is building is really that ecosystem for local commerce and I think that is again something that is really hard to replicate right and I think it's again trying to do that in the real world across you know 40, 50 plus countries and all these different jobs, all these different merchants, that's the hard part about the business.

There is an insufficient supply of researchers and people who know how to work on robotics or applied AI in the ecosystem for the recognition of all the different cool use cases you go after. And a lot of people gravitate toward the general case.

Yeah, my pitch is really simple. It's basically like, do you want to go work on prototypes and demos and be at a PhD lab or do you want to work on something where you can actually ship something in the real world? And I think that's kind of really been the culture we set up, both at Door Dash Labs and all the AI efforts, is that we're not just here to do pure research. At the end of the day, we get to ship something where we have real impact. And I think people in the autonomy world for the past 10 years were just fed up working on something for 10 years and never actually getting to a point where they saw their products being used in the real world. And for us, because we've always been much more focused on taking this much more pragmatic, practical approach, we're not here to go work on a crazy moonshot idea. It's like let's get something out that can be shipped in the real world and actually start learning how these technologies interact with the physical world and start iterating because again, these things aren't built in a vacuum. You have to put something out in the real world, make contact with the real world, and actually learn from that.

I think we did that pretty early on for Door Dash. Actually, I don't think a lot of people know, we've actually been doing autonomous deliveries in Phoenix for over two years now. We publicly announced last year that we've been doing it for over 2 years, but really in the beginning it was just learning. Okay, like I think you mentioned earlier, it's one thing to just do a fancy demo or have something that works in a one-off environment. It's entirely different to now figure out how do you turn this into an actual scaled fleet, a scaled service, a scaled business. I mean, the thing I always talk about a lot is building an autonomy business takes more than just autonomy. It's like how do you actually scale something in the real world? Scale fleets. All of a sudden, you're running into all these edge cases, right? Like you just don't see it. When you have to do something seven days a week, 10 hours a day, 7 days a week at scale, things start breaking.

It could be something as simple as a dirt covering one of your camera sensors. How robust is your autonomy stack able to handle that? There's some leaves on the ground. But it only covers, because again our dot drives on the road but it tries to act like a bike, so it'll take the right side of the road or the bike lane. And if there's leaves located along where the sidewalks are, maybe half your wheels, the right two wheels are on the leaves, the left two wheels are still on the asphalt.

Yeah. Well, all of a sudden the torque you have to send to the wheels is very different and your autonomy stack and your middleware and your low-level controls has to handle that differently. That's something I would have never thought of if it was just driving in a nice little demo environment. Things just start breaking. How do you handle operations? People don't think about actually in order to scale autonomy, there's a lot of non-autonomy operations. You have to set up depots. Again, it's a physical world business. You have to set up depots, maintenance. What if your battery? How do you recharge your battery? What if one of your braking system over? Here was an issue we ran into. There are certain situations where the vehicle has to brake so hard that the regen braking system overpowers the battery because it causes this electric shock. Again, it only happens in extreme edge cases, but there are certain situations where you have to do that because it's something in the real world. Safety is super important. So if it can't handle that, you got to figure that out. Another example we didn't think about is booting up the robots. When this was still a demo project, no one thought about boot up time. The original version of the robot boot up was a simple Jenkins script that one of our engineers hacked together in a couple hours, which worked fine. But now you're doing hundreds of robots a day, every morning needs to get booted up and the script crashes half the time. It takes 30, 45 minutes, but multiply across 500 robots, all of a sudden it's a huge productivity issue.

And then of course it's like how do you think through reliability? Now you have to start thinking about manufacturing, supply chain, and the operational aspect of how does this thing interface with merchants? How do you handle the pickup drop-off problem? How do you educate the merchant? How do you even find the location of a customer's home? Which again sounds kind of silly, but when you punch in someone's address on Google Maps, the GPS pin, especially if you're going to an apartment complex, it's never exactly the same spot. But if you're a human, you kind of figure it out. You don't think about it. A human dasher shows up, they can find where the restaurant is, this is the building, this is the front door.

Exactly. Right. And again, that's something you have to figure out. But the nice thing is again Door Dash has that data. We can see where people are actually dropping off the package.

You can only simulate so much. You can only pretend and imagine various demo situations. I think one thing that makes us very confident is pairing that world-class operational expertise that we have with world-class technology. And I think a lot of AI researchers are very hesitant to do a lot of the operational stuff or they think it's easy to handle. But one thing that's really powerful about what we have here at Door Dash is we have a world-class operations team that you can partner with, whether it's to collect or annotate data, whether it's to figure out how to deploy robots and get the fleet operations to work. For a lot of people we talked to, that's very compelling because it's like, hey, we're not just talking hypothetical here.

I mean, we've been doing deliveries in Phoenix for over two years now. We went fully autonomous L4 last year. That was a super exciting milestone. And really it's just a matter of how do you take this from originally just a couple robots, 10 robots, to 100. It's just like we got to make that hill climb of how do you scale this. I think it's really three components. Can we get the autonomy to scale? 5 years ago the question was, was autonomy even possible? Was this just a research project? Is this science fiction? Now with AI, Waymo's made that breakthrough. I think Tesla's starting to make that breakthrough. We made that breakthrough last year. Our entire autonomy stack is built in-house but purpose-built for delivery, which is a little bit different. You can't just copy and paste what Waymo's done and plop it into the Door Dash dot and everything works. The use case is a little bit different. This is a bike lane profile vehicle that's constantly navigating between the road and the sidewalks. As far as I know, there's nothing else like this in the world besides that even behaves like Door Dash dot. We built it uniquely to our use case.

So autonomy is definitely one piece. How do you keep scaling across not just Phoenix but want to bring to Bay Area, more cities? I'm sure we're going to run into more and more edge cases. But the funny thing is, the autonomy is increasingly becoming less and less of a constraint, less of a blocker. It's really more the next two. The second is operational. How do you scale operations? Restaurants behave in Phoenix look different than restaurants in San Francisco versus London versus Helsinki. How do you adapt to all these different integrations?

Interface and fleet management. And then the last piece is hardware. It's kind of funny. When we first started 5 years ago, everyone thought hardware was a commodity. Now it's starting to look like hardware is becoming a bottleneck. We hand-built the first 100 robots ourselves, which is not an issue. But then the next thousand or 10,000, we're going to have to start thinking about supply chain and component reliability. These things have to last for a really long time.

Exactly. Right. Manufacturing, learning all that, and that turns out to be a pretty hard problem at scale. So one of the things we did was we partnered up with a company called Also, which is a micromobility company spun out of Rivian. RJ is actually the board founder and chairman of the company.

If you know, why don't we work with someone who knows how to actually scale vehicles? So that's one of the partnerships we struck up. But it's kind of funny. The problem 5 years ago was autonomy. Now it's increasingly becoming more about operations, commercialization, hardware, manufacturing. And again, this is where Door Dash gets to shine with our scale advantage and operation advantage. How do we take this thing from not just 0 to 1 but like 100 to 1,000?

Yeah. 1 to three billion. Right. And I feel like Door Dash is just so well positioned to take on this. We have just such a unique advantage here. That's where we want to play, play to our strengths.

I feel like in the past couple years, what was required to really operate at a high level in the technology industry has changed a lot. And I think one of the reasons why we were so excited to acquire a company called Metis last year was really to infuse some of that AI native thinking into the company. For a company of our size, it's been really, and I think every large company is facing, and a lot of startups, you see this better than anyone else, the way they operate is so different. A lot of people at our company have struggled to see what's possible because they're so used to how things have worked historically. So figuring out how do we bring in people who have seen what is possible on the frontier and incorporating that into how we do our work. Coding is obviously the most obvious place to do transformation and we've seen a lot of gains there. But there's also work we're doing in terms of AI enablement across the entire organization. Figuring out how to benchmark various parts of the company. We announced a benchmark called Dashbench a couple weeks ago that was mainly focused on our ability to figure out how well various models and harness performed on coding tasks. That was a really good initial exercise for us to figure out how do we calculate the ROI on all this money we're spending. I was looking at it a week ago. I think our spend in June went up like 20x versus what the spend was in January.

Yeah. And so I think it's like, okay, clearly this has got to get some sort of return. And obviously I think we're seeing a lot of...

We're seeing it flatline. Okay. And I think a lot of it is through some of these intentional efforts. When people were experimenting, especially at the beginning of the year or December last year, there was a step function change in what was possible. A lot of it was just experimenting and letting people run with it. But it's gone to a point where there are easy things we can do to make sure we're not doing wasteful stuff. But also, as it relates to the benchmark we released, we need to start calculating the ROI. If there's a way for us to maximize the intelligence but delegate to open weight models for some of the cheaper tasks, we can get the same level of intelligence but pay less than if we were using closed weight models. Coding is where we think there's a lot of opportunity because the vast majority of that spend is still within engineering related tasks. But we're actually seeing the highest amount of growth in our organization in terms of seats in the non-technical organizations. Analysts are finding a lot of value, operators, account managers who are trying to figure out how to do their QBR with strategic merchants, how to automate a lot of that. So there's work we're doing to benchmark some of the work in these other areas. Another thing that is interesting for us is we work with some of these frontier labs on accounting tasks or analytics tasks. How well do the latest models perform? A challenge we've run into is we'll ask our teams how well the models perform on their task, they're like, yeah, it works okay. But then when we do...

Yeah, exactly. It's the cost. But then when we send somebody's data to the labs, we have to do data scrubbing, put it in an RL environment, and then the models crush it. But then we're like, it's kind of like what you're saying with the Sunday robotics example. If you dumb down the problem, maybe the models do well, but for some reason when we actually have it with the enterprise data with all the real stuff, it's not performing as well. So for us, it's a question of, is it because there are things we need to do with the harness to get the model to perform, or are there inherently things that the models just don't have in their data distribution or capability set that is not allowing that step function change enablement in accounting, analytics, or finance functions? So that's the next step for us beyond the coding stuff.

Yeah. Well, my take, my prediction actually is in a world where robotics, drones, AI is everywhere, my guess is that in 10 years time, we're actually going to have more Dashers doing deliveries, not less. Simply because the pace at which Door Dash is growing and the scale at which we're operating is pretty insane. We have over 9 million dashers doing deliveries and the business is growing 25% year-over-year. Fast forward 10 years, if we want to 5x from here, 10x from here, where is the supply going to come from? Are you going to have half America doing deliveries for us every month? That's probably not going to be the case. We're going to have to find other areas of opportunity to bring in new modalities as well as improve efficiencies within our business.

I think dot robotics, drones, Waymo, sidewalk robots, we're going to see a world where we have this multimodal fleet. We need to get our hands on every single modality we can get. So I think you're not only going to see more autonomy and more robotics, but I think you're going to see even more humans as well. With the introduction of autonomy and robotics and efficiency gains, you're going to see an even stronger surge in demand as delivery becomes even more affordable.

Yeah, it's a pleasure to be here.