Toyota's "Tip Of The Spear" Is Choosing Rust

Drew: Most of our audience will know what Toyota is. Very few people haven't experienced Toyota in their lives. I think fewer people will know what Woven by Toyota is. Can you start by just describing what Woven is?

Pete: Yeah, sure. So I'll try to break this down a little bit. You can think of Woven by Toyota as a means to support Toyota's transformation into a mobility company, helping to improve and realize wellbeing for all. There's a few different business units within Woven: AD/ADAS, Arene, and Woven City. These are all innovating in various spaces to be like a tip of the spear to help with this transformation process. If you're interested in any of the particulars, I can go into more detail. AD/ADAS is automated driving and advanced driver assistant systems for folks that may not be familiar with the jargon.

Drew: Right. I guess you spend a lot of time in this space, so good to clarify for those who are in other industries. Let's talk a little bit about something that I discovered when I was doing research. This is something new that Woven's been working a lot on called Arene. What is Arene?

Pete: Yeah. So, it's a great question. It's basically a name, right? There's a bunch of things that fall under this. Arene is built to enable the safe, secure, and scalable development of software. And, it allows a few different things. One aspect is that it enables reusability for multiple vehicles and multiple different applications. Trying to shorten the time horizon for getting to market has become an important thing in the automotive industry. Also, supporting data collection for future development is a big piece. There's a number of different offerings that fall under these priorities. For example, there are software development kits or SDKs, there's tools, and then there's data. The idea is that Arene SDKs help you kind of unify the software development life cycle for automotive. Arene tooling allows testing, visualization, and a really important thing is that it helps to reduce the need for physical prototyping. Arene data is this kind of holistic and consent-based approach to shuttle data back and forth from say a vehicle to a backend. It supports the ability to update on board software. So, there's a bunch of different things. I'm kind of going through the layers here. So, there's the high level priorities, then there's these sort of products or offerings, and then what does that actually entail. To that end, there are things going on around communication and middleware, simulation, hardware and software co-design, as well as SILs and HILs. SIL is software in the loop and HIL is hardware in the loop. These all kind of come together into those products I mentioned earlier- Arene SDKs, Arene tools, and Arene data. They're really a means to an end to help keep the quality bar high for anyone who drives a Toyota while also enabling the ability to ship features quickly and efficiently.

Pete: There's also this industry backdrop. You might be aware that automotive is kind of in this space right now where things are shifting and there's this concept of software defined vehicles or SDVs. The idea there is that we should have the ability to be more flexible and be able to deploy software more easily to vehicles. Arene aims to enable software-defined vehicles while also keeping safety in mind and ensuring that the needs of our customers are met.

Drew: There's so much in what you just described. I imagine Toyota plans to implement this in all of its future vehicles. Is that the right way to understand this?

Pete: Yeah, exactly. The idea is that Woven by Toyota can work with the other partners within Toyota Group, like Toyota Motor Corporation in Japan or Toyota Motors North America, and be able to bring Arene to the various customer markets that we have.

Drew: I know that a lot of time vehicles software can be hard to update. Did I hear you say that’s something Arene addresses?

Pete: Yeah. That is something that Arene does enable and makes more seamless and consistent across all the different vehicle platforms. The idea is that Arene will enable folks that are working on software for vehicles to be better equipped to use similar tools and technologies whether they’re working on the newest RAV4 and gradually expand to other vehicles the software might be for.

Drew: What stage is Arene in? Have a lot of the components been shipped in vehicles that we see on the road today, or is this more of a new thing?

Pete: Actually, the 2026 RAV4 in North America has some of these innovations that have been happening over the last little bit within Arene. One thing in particular is that there's alignment around these SDKs from developers within Woven by Toyota and Toyota Motor Corporation.

Drew: I thought I saw something about the RAV4 .

Pete: Yeah, that’s relatively recent maybe in the last couple months or so. Yeah.

Drew: The RAV4 is a big deal too. Isn’t that Toyota's best-selling platform? It seems like it's a very popular car.

Pete: I think it's popular. People in North America like to sit a little higher. They like to have an SUV experience. I think it's very popular for that reason.

Drew: Your description expanded my understanding of Arene greatly. You were talking about how it kind of includes tooling and there's all these different pieces that fall under the umbrella of Arene. Does that mean that this is something where you're kind of adding pieces as you go and then more and more of these pieces will be in cars further down the road?

Pete: Yes, you can imagine it this way: Woven by Toyota works on Arene, works with Toyota Motor Corporation, and comes up with concepts that meet the needs of the OEM, right. So, Woven is kind of like a product company. We put products on the shelf for the OEMs. We listen and hear that an OEM like TMC might be facing a certain issue. We will propose a product that could solve this issue, and it could solve the issue for many OEMs. So, we kind of put that on the shelf once we’ve done some development. So, more and more of these things will start to come online and be used over time. Some pieces of Arene, a good number of them, are on the 2026 RAV4. There's also other things that will be coming that will help improve the developer experience as well.

Drew: So your customer is really the OEMs. Even though you're kind of all under that Toyota umbrella, you're specifically solving the problems that the people designing the vehicles face.

Pete: Yeah, it's a really amazing and interesting collaborative relationship. I think a lot of automotive suppliers would love to have the kind of relationship that we have with Toyota Motor Corporation, because we can work a lot more closely and understand their needs. And, we can usually build something that will fill the needs of many OEMs. Despite what many might think, a lot of the problems faced by the automakers are the same. That's why you see things like Eclipse S-Core getting announced. In that case you've got two major European OEMs, Mercedes and BMW, pitching in and saying let's do open source stuff. They want to build the common things that are not really value additive but are expected. So yeah, we work closely with TMC and other companies within the Toyota group. By doing that, we can understand the needs really well and try to build solutions that should suit any OEM.

Drew: Outside of Arene, are there other big wins or milestones recently for Woven by Toyota that you would point out?

Pete: Yeah, so Woven by Toyota has been involved with the AD/ADAS aspect of things. I don't work super closely in that domain, but I do know that we’ve been doing things like trying to make simulators and environments where things are easier to test so that you can have a faster iteration cycle while you're developing an algorithm.

Drew: Yeah, I'm sure that's a huge area of work. Well, I guess it's a huge push for any auto company right now. Everybody's trying to figure that out. I wanted to ask about you specifically. Could you provide a little background of what you do?

Pete: Yeah, sure. I work for a team called technologies and standards which exists within the Arene team, which itself exists within Woven by Toyota. The technologies and standards team has responsibility for two primary things- technologies and standards. So, it’s what it says on the tin. Standards might for example mean crafting coding guidelines that are suitable for the programming languages used both on vehicle and off vehicle. The technology piece exists because Woven by Toyota is trying to bring the future and as I said be the tip of that spear. Technologies and standards is trying to be even maybe a step or two ahead of that. We’re involved with some things that are more in the open or open source. These are things where we’re trying to be more collaborative and part of the community so that we can better understand what others are doing.

Pete: For myself in particular, I work primarily on how we can expand the ability to use Rust within the vehicle. So today, Rust is already a pretty nice language to use for non safety critical domains like infotainment. Infotainment is the little screen you have in the center of most cars. Rust is already a pretty good fit for that. But, there are some gaps when it comes to using it everywhere on the car. There's this standard called ISO 26262 that governs both the tools that are used and how we develop safety critical software for the automotive industry. There are still some gaps as far as how we get Rust-based software deployed in the right way for safety-critical applications. So that's a gap that I work pretty actively on trying to close. I'm lead of the Safety-Critical Rust Consortium where we have a bunch of people involved from different automakers, airplane manufacturers, and even a nuclear company. There's also a rail company in there. We’re all trying to solve similar problems trying to bring Rust software into safety-critical systems.

Drew: I've been talking to a bunch of people from this space. I think a lot of the people are actually members of the consortium. This is a very interesting moment in the maturity of the language and ecosystem because you can very clearly see why you would want Rust for these applications. But, there's some significant procedural elements that need to be filled in to make that happen. Are there specific technical problems that you've taken on as part of this work?

Pete: Yeah, so one key thing that needs to exist based on ISO 26262 is a coding standard for whatever language you're using. We realized early on in the consortium that this was an important piece, and it was not there yet. What we found when we started talking to members that were joining the consortium is that people were building their own standards within their companies. That’s okay, and it gets the job done, but now that we have this consortium we can do this in a collaborative manner so that we can all benefit. So, in addition to leading the consortium, I also am chair of the coding guidelines subcommittee. This is a group that is focused entirely on shipping an artifact that can be used by an automotive company or an aerospace company on their own software. They can bring their software together in a bundle with these coding guidelines and say, hey we followed these coding guidelines with this software, therefore you safety assessor should closely look at these guidelines. If they think it's okay, then you get safety-certified. We have a repo out there. You can take a look at it if you're interested in our progress and how we're coming along.

Drew: I'm hearing you talk about Rust and all this effort to make it work for safety-critical domains. It's making me think about a question that I wanted to ask anyway. Where does Rust fit in the stack for Woven right now? I imagine there are places where it can be put today and places where there's still work that needs to be done for it to be used.

Pete: Rust has historically been an off-vehicle technology for us. We've leveraged it in areas which require what Rust is great at security, stability, and reliability. A good example of this is a simulation engine built by a team within Woven by Toyota. It’s written entirely in Rust for stability and extensibility reasons. That was written in Rust a few years back almost as a proof of concept to see what might be the shortcomings of Rust. But, it took off. It now looks like quite a viable path for doing this kind of work that involves a lot of number crunching and also must be very reliable.

Pete: There’s another thing I’ll point out, and I think this is the announcement of this actually. Rust is today shipping in the 2026 RAV4 in North America. There's a group called TCNA or Toyota Connected North America that is largely responsible for this. Their UI/UX team was working on delivering customer experiences in infotainment, and they found Rust to be a really strong choice for what I might call a kind of vehicle interface layer. That’s where there’s software managing vehicle resources and activities. They found this really allowed their designers to focus more on the UI and the UX and building a customer experience that was engaging. They found that Rust served their needs quite nicely here because it's sort of like this rock solid, stable layer that lets these UI UX designers really focus on building a good experience.

Drew: That's awesome! I guess people heard it here first. There's something satisfying about hearing that Rust shipped on a product that we all see rolling around.

Drew: Something that I think about when I think about Rust for automotive applications is the fact that Rust is kind of all about safety and reliability. These are things that are obviously very important in the automotive space in terms of safety in the physical world. Are there ways that you see the safety guarantees that you get from Rust translating to kind of real world safety with these vehicles, or am I inventing that connection?

Pete: You're exactly right. I think it's a good point to bring up. I touched on this earlier, but there are certain industries like automotive, aerospace, and nuclear that are safety critical. They have, for good reason, certain regulatory burdens in order to keep people safe and prevent destruction of property. In particular, I mentioned the ISO 26262 standard for automotive. It prescribes increasing levels of safety criticality as you go down the list from ASIL QM, which is quality managed all the way through ASIL A, B, C, and D. So D is the highest level of safety criticality. When I look at Rust and what it provides- memory safety, thread safety, and type safety, these all really dovetail nicely into the needs of safety critical systems. And, these are all things that were built into it with the idea of providing a secure experience for people that are using a web browser. That was its initial application. If you look at the ecosystem now, you'll see that others are starting to feel similarly. Tool vendors and compiler vendors are starting to step up. There are now tools available and compilers available for Rust that target specifically the safety-critical segment. If you look, there’s Ferrocene, GNAT Pro for Rust, and the HighTec Compiler. These are good signs along with seeing that, for example, Volvo Cars has a low power ECU that has Rust on it.

Pete: There's a lot of activity that's happening here. There're a lot of people that are thinking in the same way that Rust’s qualities make it a natural fit for safety-critical systems. That's where you see the thing that got started last year, The Safety Critical Rust Consortium, come in. I think we have around 170 members or something now. It’s very diverse across different industries. Certain industries may take longer than others. But, I have already heard, for example, that some medical device manufacturers are starting to use Rust rather than other languages. It's definitely coming along.

Drew: 170 members is a lot bigger than I thought it was! That's exciting to hear. If we come out of the safety realm for a minute and just talk about rust in automotive generally, what do you see as the outlook for Rust in the space? What are you seeing and hearing about adoption?

Pete: I touched on this a bit earlier, but there's the Eclipse S-Core project, which is hosted under the Eclipse Foundation and the Eclipse Software Defined Vehicle Working Group. There was a memorandum of understanding signed a few months back between a couple of big auto players and a bunch of suppliers saying that they would work together in the open. So, they’re going to build these shared pieces that we need.

Pete: In fact, in the Eclipse SDV Working Group, there's a bunch of other projects that are written in Rust as well. There's Eclipse KUKSA, which is like a data broker based off of COVESA VSS. So, it's kind of like a way of shipping data around within the vehicle and out of the vehicle in a specified format. There's Eclipse Ankaios, which is kind of like a workload orchestrator for embedded devices. When I say embedded, I don’t mean microcontrollers but embedded Linux environments. There is also the Eclipse uProtocol, which you can think of as kind of a service mesh. It's sort of in the name of uProtocol, but the idea is that you can bring your own protocol to put underneath it while getting a consistent developer experience for anyone using it. There's also Eclipse Iceoryx. That is a zero copy shared memory transport that seems very promising for use in the vehicle to be able to shuttle data around without making a bunch of copies. That can be very important when you're running on automotive grade hardware that is more expensive than consumer grade hardware. There’s always a question of bill of materials, so using something like Iceoryx can save you some copies…

Pete: There's also Eclipse Zenoh. Zenoh is another form of transport. It's basically a PubSub request-response mechanism. You can use a bunch of other protocols underneath it as well. At OxidizeConf, Ansgar of the Eclipse Foundation shared that these days around 35% of the code in the Eclipse SDV repos is actually written in Rust. That was really cool to hear about. I’m not sure if the talks have been released yet, but it was really cool to hear that.

Pete: There's also a Rust Special Interest Group or Rust SIG that's in the Eclipse SDV Working Group. So, we actually have this space where people in the automotive domain that are using Rust or are curious about Rust can show up and talk about things. We also hold some office hours where people can stop by if they're curious. We've had a couple automakers and a supplier come by and just basically ask where to start. That’s kind of a good place for sharing that information and finding that base of people that understand how to use Rust in automotive. For me personally, it's more a question of when and not if Rust comes to automotive.

Drew: I feel like I'm going to say this a lot of times in this interview, but it feels like an exciting moment. There's enough that's happened where we can see that this is a real trend that's beginning, but you can also see that it's just the beginning.

Drew: We talked a little bit about how Woven occupies this interesting space within all of Toyota where it's kind of this research and development lab. Does that make the culture differ in any significant ways from a more typical corporate environment?

Pete: Sure. So, the first thing I'll say is that there also is a company called Toyota Research Institute, and they are the folks that are truly doing pure R&D. So, their R&D is a little bit further out. It’s basically fundamental R&D. But, when it comes to whether Woven feels different, I think the answer is yes and no. It's a little bit from column A and a little bit from column B.

Pete: In certain ways, Woven by Toyota is like many automotive companies. We develop cyber physical systems. We've got hardware and we've got software. The environment is sensed. Things are actuating. Actual breaking and steering is happening. On the other hand, I find that the spirit at Woven by Toyota is uniquely empowering in a way. There’s a lot of space for individuals to try to figure new things out and explore new avenues to accomplish things.

Pete: A good example of that comes from when I first started at Woven. There was this discussion about which technology to use to implement a tool that was needed. I proposed this parallel path that would allow us to learn some things. So, one team would implement the tool in C++ and I would implement the tool in Rust. By going through that process, we learned some valuable lessons about how efficiently we can develop and ship things in Rust. And, we were still able to capture the needs of our customers and get them software delivered on time.

Drew: Well, you dangled that in front of me. I have to ask, what did you learn about efficiency and Rust? I know this is a big topic for the community as well.

Pete: Yeah. I'll be honest and say that I have been learning Rust and using Rust for quite a number of years now. I think it’s been since 2017 or so. So, I knew a bit about Rust, and I was able to use it somewhat effectively. But, what I found is that I as a single individual was able to ship and deliver a piece of software before a team of other individuals were able to ship that same software with similar features in C++. That's not to say that C++ is a bad language or that it shouldn't be used or anything like that. But, some of the tooling and things that are available in the two languages are different. With Rust for example, you can reach out to crates.io. There are many, many library crates that already exist to do many things that this team that was using C++ ended up having to build themselves or pull together from resources that were not as easily available. That was definitely an aspect that I found was helpful for being very efficient and cutting code very quickly.

Drew: That's cool! People are gonna go crazy for that story. People will really love to hear that. I think it's interesting that you point out crates.io and kind of the tooling around the language. I think that actually matters a lot for productivity.

Pete: I totally agree. I was giving some training about rust over the summer at Woven by Toyota in the Ann Arbor office. I was like, hey Rust is this great language, show up and I'll teach you all about it. I was using the comprehensive Rust set of tutorials from the Google Android team. So, I was going through the things and it almost felt like I could mint a saying from this experience. I think it would be something like “lure them in with memory safety and they'll stay for the tooling and ecosystem.” When I showed them the ability to use documentation tests or doc tests where you can have examples that are guaranteed to compile and guaranteed to work, it blew people's minds. They thought it was amazing.

Pete: People hear about some of the other aspects of Rust that are really appealing for domains like automotive and safety-critical, but then they kind of take a step into the ecosystem and they're like, wait, all these tools are super helpful. It all just helps them move very quickly.

Drew: I wouldn’t say that there's nothing to the claims that people make about there being a bit of a learning curve with Rust. You have to learn how to work with the borrow checker. That is real. But, I think your case is a great example of the fact that once you've kind of done that climbing, you start to kind of speed up from there. All of the other aspects of the language like the tooling and the fact that you can use the compiler to help you fix things are all very helpful. I think that's a very cool, underappreciated aspect of working in Rust.

Pete: Yeah, definitely. I had a chance to join the Rust Vision Doc team that's in the Rust project. As part of that, I might've gone overboard, but I interviewed like 20 to 25 people that are using Rust in safety-critical areas or would like to be using Rust in those areas. It was just so fascinating learning from them. And, like I said, what got them in the door was often not what kept them. They got all these ecosystem benefits and they were able to deliver software that had less defects. And, they could deliver it faster than the equivalent software they would have written using other languages. This is something I heard almost universally. If you're interested in learning a little more about that, Jack and Niko gave a talk at RustConf 2025 about this. They shared some details.

Drew: Picking back up with the thread of your work at Woven, you mentioned that part of the work is actually actuating things and stuff. How much hardware interaction is there? Or, is it very software heavy? What does that look like?

Pete: The reality is that in the automotive space you very much need to think about hardware-software co-design. I think the idea that you can be a pure software company in the automotive industry is a bit of a folly. So, we really have embraced that. From the beginning, we're thinking about how what we're writing will work on the target hardware. We get to prototype things to make sure that it works when we ship it to customers. So, there's plenty of hardware involved. In my day to day, there's not a lot of hardware. But, I did spend my years at other companies doing things like actively writing code while I'm on the test track. So, I've put my time in there. These days I'm not interacting with a ton of hardware. But, I do see those things happening in the offices, and it is really cool..

Drew: I always like to ask about that. For a lot of software engineers that spend a lot of time in a cubicle with their hands on the keyboard, it's kind of fun to think about what it might be like to have that different aspect to work.

Drew: Is there anything else unusual about the culture at Woven that you would point out?

Pete: Yeah. I've worked at a few other automotive companies, and I would say that Woven by Toyota is one of the most deeply technical places I've worked. It’s unique when it comes to the way we sort of draft out designs while keeping in mind production constraints and future evolution. We also try to gather together those that will be users of the software and hardware eventually and get feedback.

Pete: So, this has been an interesting experience. It's very unique amongst the automotive industry. It's much more typical to have fairly rigid hardware and software designs created by a systems engineer who is often very far removed from the requirements they write. I've seen situations in other companies where the systems engineering team finishes the work and then it might be months or years before a single line of software is actually written. That kind of stuff is not uncommon to see in the automotive industry. But, at Woven by Toyota there seems to be tighter collaboration between the different teams. And, I've seen it working out well. We've been shipping really well, so I think it's a great cultural aspect of Woven by Toyota.

Drew: To ask more specifically about hiring, what are the key things that Woven tends to look for?

Pete: So, Woven by Toyota has an internship program and hires successful interns out of that program. That is definitely an avenue to get hired. Woven also hires experienced professionals.

Pete: For interns, I think generally Woven by Toyota is looking for a lot of curiosity, kind of some spark, and having touched multiple parts of the system. So, not only software but also hardware. For example, with Arene, I mentioned a bunch of things that we do. So, in that case, having touched a bit of middleware and understanding what middleware is would be good. Maybe you’re someone who’s toyed with middleware or submitted a patch to some open source project dealing with communications protocols or applications development. What I’m describing is kind of someone that is sort of the classical T-shaped individual. They may be very deep in one aspect, but they've also touched and been involved with other aspects of the software and hardware development process.

Pete: When it comes to experienced individuals, I think what I said still holds. You want a T-shaped individual, but maybe their T is a little bit thicker on top and a little bit deeper on the part that they know more about. So maybe the proportions of the T get a little stranger or more interesting. Ideally, you've touched multiple different parts of the software stack, hardware stack, and you kind of maintain a certain level of curiosity. I think that curiosity is really beneficial at Woven by Toyota. Given how we develop things and what we develop, it's very important to have that spark and curiosity. We work on so many different things. For example, depending on the time of year or which year it is you may be working on something completely different from what you were working on previously. So, the ability to pivot and learn is definitely a plus in my book.

Drew: The work at Woven involves cars, automated driving, and just a ton of very cool stuff. I'm sure there are tons of people who are interested in this field and getting into this type of work. If you were to speak right to those individuals, what advice would you have for them?

Pete: So, Woven by Toyota is one of the most technical companies I've ever worked for. There are actually world renowned experts that call Woven by Toyota their home. That is really cool. You can ask people that are true experts questions.

Pete: So, if you’re earlier on in your career, I would again say to keep a wide lens. Try to keep an open mind and try to maintain that curiosity. Ask why things are the way they are, and try to touch a bunch of different things. I had the good fortune earlier in my career to work on R&D teams and advanced engineering teams. That let me go deep on something for six months to a year at a time and then kind of pivot and do other things. So, I got to touch a bunch of different things and learn various things. I think if you were to join a team and just work on that team for six years, you may learn something deeply but have no idea about the rest of the development process. So, I would definitely say keep an open mind.

Pete: If you are interested in getting involved with this type of work, one thing that you could do is look into open source software and things that are going on in the Eclipse SDV Working Group. Woven by Toyota has a very strong interest in being involved in open source communities and initiatives. It makes sense because if we collaboratively implement things upstream this has the greatest potential to improve Toyota Group’s entire supply base who may be using that software. For example, we have compiler engineers at Woven by Toyota that are active contributors to the LLVM and Clang projects. We also have the Safety-Critical Rust Consortium where Woven by Toyota was a founding member. That's a place where we work broadly with those that want to see Rust succeed. So yeah, if you're interested in sort of finding your way to Woven by Toyota, explore broadly, look into interesting open source projects that might be related to automotive, and just keep investigating and playing around.

Drew: That seems like sound advice from the outside.

Drew: Well, I feel pretty happy with everything we've been able to discuss. Is there anything else that you would want to mention?

Pete: I just want to really thank you for having me out here to discuss. And, I also want to say that it feels like there's been a bit of a thought process that Rust is a certain thing. But, what I've learned from the Rust Vision Doc and doing a bunch of interviews is that Rust is a lot more broad. It’s a lot more broadly applicable to writing foundational software. If you're someone interested in using Rust where you're at, or you're interested in trying it out, Yuri of Rivian gave a nice talk at RustConf 2025 all about bringing new technologies into a company and how to adopt them and scale them. I found the talk really interesting and insightful. So take a look at that if you're interested in using Rust at work trying to figure out how you might go about it.

Drew: Thank you so much Pete, and thanks for all you do for the community as well.

Pete: Thanks for having me.