Driving The Future With Software Defined Vehicles

Tom White:
Dipti, welcome to the IoT Podcast.

Dipti Vachani:
Thank you for having me.

Tom White:
Thank you for joining us. As I was saying backstage, really genuinely excited for this. And I’m not just saying that, I think it’s easy to get into podcasts and say that on every single episode. But I genuinely am.

Dipti Vachani:
This time it was just me. Just.

Tom White:
Just you, it’s all for you.

Tom White:
But really interested for the folks to hear about your background, what you’re doing at Arm, specifically around automotive, and we’re gonna cover a range of topics today. But for everyone that’s viewing and listening or however they’re consuming this podcast today, could you give us an introduction to who you are?

Dipti Vachani:
Oh, who I am. I’ve been in the industry, I don’t want to say how many years, because I’ll be aging myself at that point. But my background includes about 17 years at Texas Instruments and some time at Intel most recently. And now at Arm, and I run the automotive business Arm. When I first started, automotive and IoT were together. We had a combined business as we were starting to nurture that IoT business as well as nurturing the automotive business. But both of those businesses have become quite significant for the company and a significant investment as well as a future. And so now I’m taking the automotive on as to what the next direction is, what software to find vehicles, electrification, all the amazing things that’s happening in that industry.

Tom White:
Excellent.

Dipti Vachani:
And at the end of the day, it’s just a big thing in the middle of IoT.

Tom White:
Yeah, yeah, absolutely. I think that’s the real key part here, IoT being a concept and connecting various industries and ecosystems together. And automotive is one that really interests me personally. I’m a big car nut, I’ve always had been, right? So anything to do with the automotive industry, naturally I lean to. But we had an episode actually a few months back with a friend of mine who was at AWS talking about software-defined vehicles. And I think for me, that’d be really interesting to get into actually, just to understand a little bit about where we’re all going. I think certainly,

Dipti Vachani:
amazing.

Tom White:
yeah, it is, isn’t it? It’s, you know, from what was that kind of mechanical internal combustion engine, very messy actually.

Dipti Vachani:
Doesn’t change, big hunk of metal, but pretty cool hunk of metal, but basically that. Now it’s a connected device and kind of the same experience you have from your mobile phone from a software experience standpoint.

Tom White:
So on that note, for people to understand that might not understand software defined vehicles, you’ve just labeled it there. So it is a mobile phone effectively, right? That the car or an automotive vehicle is turning into that. Is that how you would best describe it?

Dipti Vachani:
Yeah, and often software-defined vehicles get synonymous with autonomous, right? So we think, okay, autonomous vehicles is software-defined. But honestly, it is much more and everything beyond that. So think about just back in the day, I was just remembering this, I had a car which had a mapping, a map in it and that was really cool back then. when we had a map in her car, right? And I had to go buy new CDs. I’m really aging myself to update that map, right? I had to go get a new
CD from the dealership. I mean, remember the pain that is?

Tom White:
The PNDs, right? Personal Navigation Devices.

Dipti Vachani:
yeah,

Tom White:
Yeah.

Dipti Vachani:
yeah. And then you had to update. So now that’s just done dynamically. And that in itself is software defined. It is defined by software. It can be updated by software. It’s on the air updates, I mean, on the go updates. And so it is so holistically changing the experience in the car beyond, in addition to all the kind of autonomous functions that we’re seeing in the car, a nice user experience we’re starting to see in the car, as well as the electrification. All of those things are kind of happening, but software defined as pervasive across all.

Tom White:
That’s a really nice way to explain it. And I think clears things up for people that automatically assume SWD is actually autonomous driving, which is a component, right? And as you say, it’s pervasive across all the aspects often those get mixed, right?

Dipti Vachani:
Like they think autonomous has everything to do with electrification. Well, no, not really. Those are actually separate too. It’s just all happening at the same time that it’s easy to put them all in one bucket. But those fundamental technologies are happening in parallel.

Tom White:
Yeah. Why do you think there is this coalition of growing advancements at the moment, certainly within automotive, you know, and it’s all happening at the same time? Is it because the tech wasn’t as robust before to be able to do both the software defined and the autonomous driving, the electrification? Is that your view on it?

Dipti Vachani:
You know, it’s a very interesting and again, this is a, it almost feels like the perfect storm, right? If you think about it, there’s multiple components coming in. One is the electric battery technology is ready, right? And we have our friends at Tesla to thank for that, with all that investment that’s been done to create. that battery technology now we’re ready for electrification. So that’s one thing that’s happening. And our awareness of green technologies and our sensitivity to that, right? And the next generation sensitivity to that. So that’s happening. And from a social construct as well as a technology construct. And then in addition, we figured out that we can put this technology in a car with lots of software to create a safer environment. So now more autonomous functions are capable, lane change, whether it’s that or crash detection. All of those technologies are becoming more mature and we’re ready to deploy them in a real situation. And so there’s maturity of that technology and also an acceptance of that technology. It’s, you know, you’d have to interview people to see who would be comfortable getting in an autonomous car. That’s a pretty interesting conversation we’ve had over at dinner table. There’ll be half the crowd who’ll say, no, I would never get in a car that would just drive itself. I don’t trust it. And

Tom White:
Hmm.

Dipti Vachani:
so there’s a trust element that’s still, we’re still navigating through that, but that’s starting to happen. So that’s the autonomous side of things. Technology is ready and social acceptance of that as well. And then the software to find element, it’s the cloud functionality or cloud technologies like hypervisors and… and the kind of software capabilities, AI capabilities, or perception technologies, or learning technologies that are done in the cloud are maturing as well, which allow for that really nice update cycles that we can have between the data center and the cloud. So those technologies maturing as well, the data center technologies, and now we’re more comfortable having that in a vehicle. those kind of technologies. And so it’s not one thing. I

Tom White:
Hmm.

Dipti Vachani:
guess that’s the point, right? It’s often we want to look for one answer, but it’s multiple elements happening all at the same time and that perfect storm of those elements coming together to mature the environment that we have in the car now.

Tom White:
Yeah, yeah, it really is remarkable, isn’t it, that everything comes together at the same time.

Dipti Vachani:
It does.

Tom White:
And that point that you made around the trust, I think it’s a really big one actually, isn’t it? Because I think for years that’s plagued the industry, certainly from self-driving, there’s certain levels of self-driving, I think level four and five, like the higher levels, do people actually trust that? And I think that’s a continual ongoing debate and something that’s gonna have to change as times change. But in terms of the benefits then for consumers and businesses, what are the tangible benefits of a software defined vehicle?

Dipti Vachani:
Oh, wow. It’s, they’re, you know, there’s just one of those, I’ll get to the probably the one that we experienced most. When I get a new software on my update on my car, it’s like Christmas morning, right? I get in there and I’m like, wow, what happened? Like, what’s gonna happen? Like, I got all these new functionalities. And so there’s just that experience of getting a new thing. It feels like a new thing when you get a new software update. Beyond that though, what’s happening is more safety functionality is being deployed, more security is being deployed. So as we learn of gaps in our security technology in the car, we’re deploying updates. And so our security is getting more robust. It’s not a stagnant unit. As we learn, we improve, right? We’re learning also from big data, meaning we’ve taken data across multiple vehicles and from that learn how to improve. the safety of that car and that learning then gets deployed to that individual car which is a really neat way to continue to improve our experience and our safety in the car. And so you see the updates and you see the new functionality. That’s what we experience. But there’s a lot of other stuff that’s happening behind the scenes that just makes that car more safe, more secure. and from all intents and purposes, just more robust, right? And it’s scalable

Tom White:
Hmm.

Dipti Vachani:
over time. It doesn’t become stagnant. The features of the car, most of us have more functionality to the car than we actually take advantage of. And over time, that functionality gets deployed as we get more software updates.

Tom White:
Yeah, and I think that’s one of the major things, isn’t it? So sometimes you often see in the media, you get a bit of bad press actually, when you see things like microtransactions and having to pay for various elements of the car. But ultimately, the safety and the harnessing of thousands, if not millions of vehicles on the road and then being able to deploy that straight away, I think is really quite special.

Dipti Vachani:

And in the end, that is benefiting the consumer. And the consumer is getting the benefit of all of those, the value that’s coming from those software updates. And that’s how, you know, and the way ARM looks at it is, we have the assets in the data center, we have assets in the car, and they’re very similar but different. we have to think about the real-time nature and the safety elements of the car that are fundamentally different than anything else that’s out there.

Dipti Vachani:
And as we take into account the safety, the real-time nature, we then put that kind of knowing and that knowledge in our data center assets as well, so that there’s a cohesiveness between those two. deployment units, right? So what’s happening in the cloud and how it’s deployed in the thing is done holistically and viewed as from the safety and the real-time concerns of the car, right? That we’re not looking at it just another thing. It is a thing that has very, very important and assets that need to be considered when we’re deploying.

Tom White:
Yeah, absolutely. And I think you’ve touched upon that just a little bit there, dipti, about arms are old in all of this. And that would be really interesting to get into now. So, Arm are heavily involved in this area and for good reason, right?

Dipti Vachani:
You know, we’ve been in the automotive space for over 20 years, right? And that’s important. That’s important because longevity, experience, investment in the space, and the learning that comes from that is so important for this industry, because as you know, the dynamics of a car, and the longevity of what a vehicle is in the market for. And so there’s that longevity, but we’ve also been investing in safety for over 20 years as well. Those safety technologies are very unique and very important to hitting the criteria for putting a human being in a vehicle. And so our safety investments. And so for a while we’ve been deploying A.E. technologies, and that is automotive enhanced. That is technologies and IP that are ASLB and ASLD ready. That is that we’ve done all the work to ensure that you can get to ASLB and ASLD standards. And for those that are not in the automotive industry, those are standards that ensure there’s a safety that we understand what the safety and the testability of the vehicle is, right, or the technology is. And so that’s one area where we’re investing. But beyond that, what we’d like to see is the safety technology scale. And to scale. And this is where ARM adds its most value, right? We know how to scale technologies. And so one of the investments we’ve been making is our investment in SOAFEE. Because what we see is this vast amount of software that’s being deployed into a car, the hardware can be there, but the long pole in tech now is the software, the software investment. So just getting all of that software and then being able to reuse that software across multiple hardware technologies critical to the OEMs. And so we’ve created a standard called SOAFEE. It’s an open standard. And in that governing body, we allow for reuse of software across multiple hardware. We also create a technology that is synonymous with what’s in the cloud. So it’s built on the system-ready technologies that we already have in the cloud. So this is what I was referring to earlier. that we have technology in the cloud, but now we have an understanding of the safety and the real-time nature of the vehicle itself. And between those two standards, we’re able to cohesively look at this from cloud to edge. And then the other element of what we’ve enabled in SOAFEE is allowed to shift left to software development because we recognize that software is delaying the vehicle deployment. And

Tom White:
Hmm.

Dipti Vachani:
so if we shift left to software development, we enable cars to be software defined. and we enable reuse of software across multiple hardware deployments, that’s when scale happens. That’s when we’re really able to scale this kind of technology. And so that’s our position in the industry, and that’s why we’ve invested in what is SOAFEE. SOAFEE is an open standard, and we have the likes of AWS, Cariad, Red Hat, SUSE, all involved in the governing of SOFI. We’ve opened it up into the industry because… It has to be, it’s just not one of us, right? We have to come together across the industry to enable.

Tom White:
Yeah, and I think, you know, I think it’s just fantastic that you’re championing that as a business, right? I think if you look at IoT or any standard really in technology, it’s really, really important to have a standard, to have a benchmark, to be able to know where you are against something else, but also the software reuse to get it out to scale. So if you look at MATA, for instance, or some of the work that CatenaX are doing, right? You know, and, you know, the… The enablement of SOAFEE sounds like we’re on the right track.

Tom White:
you’ve got some really good partners in there. And as a sort of governing body, what’s the aims behind SOAFEE then, and where’s it likely to go in the future?

Dipti Vachani:
So like I mentioned, we have three different pillars that we’re focused on.

Dipti Vachani:
One is just the technologies that we’ve been able to use in the cloud. We now can deploy those kind of technologies, technologies like hypervisors, technologies of the real time nature of software deployment. We have the standard base such that we have now cloud native assets that we can deploy into the cloud, into the edge. while keeping in mind the safety and the real-time nature of the car. So that is the software defined and the synergy between what’s in the Edge and what’s in the cloud. Then it’s the software reuse. So multiple hardware technologies. Right now, it’s amazing if you think about it, every new hardware deployment, whether you get it from one Silicon Ven or another, you have to do a brand new software deployment. How is that going to happen when you have that many vehicles and that many different technologies available and the… the vast amount of software that has to be deployed for each one of these vehicles. If you had to do it bespoke for every single car, every single model, can you imagine that there’s no way to scale? The problem becomes exponential. So it’s software reuse. And then it’s starting software early. So being able to simulate the car enough so that you can start your software design and deployment early. And so those are the three premises of what we created SOAFEE around. But even, and, As we get into SOAFEE, what we’ve been really focused on this year is seeing some deployments with that SOAFEE standards. So we have lots of different players that are already investing in platforms or have demos or having, being able to demonstrate that technology. But what we’re really focused on now is to see it on the road. And that’s going to be our number one focus.

Tom White:
Yeah, I can imagine because you need the real data, don’t you, right? In order to then collect it, harness it across everyone else that’s involved from a safety first perspective, one would imagine. Being able to understand various elements and make the right decisions. And is that what we can expect for SOAFEE in the future? And is this how…

Dipti Vachani:
What you can expect from SOAFEE in the future is that you’re going to see all of… So in SOAFEE, you have OEMs, we have tier ones, we have silicon vendors, we have software vendors. It’s over 80 participants in SOAFEE. And what’s beautiful about it is that all are participating in the software development of and the creation of SOAFEE. So what you’re going to see in the near term is these vendors all leaning in, all of these players leaning in, AWS leaning in, and Cariad and others, and start to see that develop into real code that can be reused in a vehicle, right? So we’re starting to see everyone participate in contributions. That’s what’s happening now. And then what we’re going to see in the near future is how we can see it demonstrated in a vehicle in production. and then see that software to find capability being reused in the cloud and in the edge. And so I think the stuff across industry, it’s easy when you go alone, you go further when you go together, but it takes a bit longer, right, as we see it all come together this next year.

Tom White:
Yeah. And I think, you know, if history teaches us anything, it’s this collaborative, open source nature gets us to that end result actually, quite quickly. And I think when it comes to things like automobiles and the automotive industry, it’s really, really important. You know, we can’t gloss over the fact that the safety of this and the coalition of various… electrification and autonomous driving, etc. It’s critical that we learn from this and get the real world data, as you mentioned, and actually deploying SOFIA out there so that we can learn, understand and make the necessary changes. So I completely agree with you. So just on to my next point then. So what sort of lessons have already been learned so far in this process from implementation of software defined vehicles?

Dipti Vachani:
Well, you know, it’s interesting because we started with software-defined vehicles, and it’s been around for quite some time. But what we’re starting to see with the onset of electrification, autonomous functionality, and software-defined vehicles, you know, I talked about that perfect storm element of things,

Tom White:
Yes.

Dipti Vachani:
is that we’re starting to look at the vehicle differently, in that can we create centralized compute and zonal controllers, right? So some level of… merger of the IVI and the ADAS unit coming together. So most often, the IVI was one unit, and the ADAS is another unit. And within the vehicle, we’re starting to see the merger of those, because it’s easier to deploy software, because that functionality is kind of blurry between the two. And so what does that create as an opportunity for mixed criticality? It’s an opportunity to take some of the assets that ARM has when it comes to. when it comes to our safety and the security technology that’s embedded in our architecture. We have the ability to do mixed criticality. We have confidential compute. All of those features that we’ve been thinking about for the last 10 years and deployed in our V9 architecture are now starting to see the light of day as the merger of this architecture happens in the car, IVI and ADAS. And then the other element that we’re seeing is across the vehicle, we’re starting to see zonal controllers. And these zonal controllers are also software defined. And so that way you have the ability to play software all the way down to heated seats or whether it be your power locker or things of that sort. So it’s pretty amazing the merger of what is software and then the actual change and how hardware is done.

Tom White:
Yeah, I mean, it’s it is mind boggling how quickly this has all happened as well, right?

Dipti Vachani:
Well, you know, like the electrification of a car, the powertrain, that hasn’t been touched in over 20 years. Right? We don’t, this just worked. Why do we need to mess with it? And now we’re starting to redefine that. It’s pretty amazing how quickly that’s happening.

Tom White:
Yeah, and I think public interest in this is quite high as well now, isn’t it? You know, there’s various tax breaks and incentives for people to get more involved in electrified vehicles. And of course, we’re all trying to work towards a particular turn-off date, you know. So I think, you know, it’s definitely the perfect storm as you’ve mentioned a couple of times today. And really, really interesting. What does it look like in the future then in terms of software-defined vehicles? How would you envision? it impacting the automotive industry in a kind of broader sense.

Dipti Vachani:
I think we’re going to continue to see ADAS functionality improve. So the safety elements of a car, you know, we talked a while back about a fully autonomous car. Let’s put that aside for just a moment. But what you’ll start to see is just more and more of that safety being deployed in the car. And we’ll get more and more comfortable with it because we have the ability to update it as we learn, right? And so more ADAS functionality in a car, because we have the ability to learn faster, and we have the ability to deploy software faster. I think you’ll see the, as more ADAS functionality gets deployed in the car, you’ll see the user experience of a car start to change. More entertainment in the car, more ability to have some independence, right? Then there’s this panacea of solutions where, the software defined nature of the car and the software defined nature of the city, smart cities, you start to see that come together where you have an awareness of the city around you, the red lights around you, the streetlights, the cameras, the traffic monitoring that can happen within the infrastructure that also being used in the car. Now we have the technology to do that. It’s a collaboration that’s necessary to bring all that together. But that’s a, that… That is when we really start to take advantage of the software defined nature of the infrastructure and the software defined nature of the car. Is when we can bring that.

Tom White:
Yeah, yeah, absolutely. And in terms of arms involvement, automotive is gonna continue to play a major part of arms business unit moving forward in terms of the growth and various.

Dipti Vachani:
Absolutely. It gets even better for us in such that we continue to push the limits of how much performance you can get within a power window. Power efficiency has always been our forte. That is our value. And the ability to get more and more performance within a certain power budget is our secret sauce. And that’s what we used in the mobile phone. is also being

Tom White:
Yes.

Dipti Vachani:
used in the car. Because you can’t have these power hungry data centers sitting in a car. How are you going to power that? How are you going to go the distance with that? And so we’ll continue to invest in that. But more so we’re going to continue to invest in pulling all of these solutions together, because it’s no longer just about the CPU. It’s about the CPU, the graphics unit, the camera unit, the ML, the AI. and all of those technologies coming together. It’s a heterogeneous compute unit. And so we’re looking at this problem more holistically, not just being power efficient in the CPU, which we know how to do very well, but we also have a safety GPU, a safety ISP, and looking at the ML and the AI problem and how we integrate that into the GPU and the CPU. And so we’ll continue to look at this problem more holistically and then bring it all together with Sophie in such that you can deploy this at scale.

Tom White:
And I think that’s the critical part, isn’t it? You know, deploying it at scale. Because if you didn’t have that, it’s all very kind of esoteric, isn’t it? It’s all quite difficult. You’d have to start from scratch each and every time someone wanted to try and do something.

Dipti Vachani:
And you and I have both been on the streets and seen those test vehicles running around. When are we going to get out of those test vehicles and actually see something in production? That would be cool to see. And that’s what it’s going to take. It’s going to take this power efficient solution and the ability to have mature software to be able to do that in. And time under tension. And the amount of data collected in these vehicles that are running around is going to help us be able to deploy this thing.

Tom White:
Yeah, I think that’s a really interesting point that you’ve just mentioned, the amount of data, because it really is colossal. If you look at the cameras, lidar systems, ADAS systems, everything that’s within a vehicle now, it’s huge, isn’t it? So that in itself is probably quite a challenge to actually interpret that data and make good use of it, because the amount of vehicles on the road.

Dipti Vachani:
Well, and what’s so beautiful is we’re having the same conversations in automotive. This is an IoT podcast. We’ve had this conversation forever, which is what data do we need at the edge? What data sits in the data center? What needs to be private? What is open? All those conversations, we’ve had in IoT for a while now. We’re having those same conversations in the car.

Tom White:
Yeah.

Dipti Vachani:
How much data do we need at the edge? How much data is in the cloud? Where does perception happen? Where does learning happen, right? And it’s still a conversation up for debate because we’re trying to figure out that. And then line’s gonna continue to change. It looks like it’s changing in IoT, right? It’s gonna change in the automotive as well.

Tom White:
And, you know, when you look at companies like Vertical Aerospace or Zoots or other people in the automotive field, this kind of problem of data and harvesting and understanding it magnifies multiple amounts when you look into actually the prospect of flying vehicles, right? Because we’re not, a million miles away.

Dipti Vachani:
I still haven’t got my head around that one yet, but yes, flying vehicles or I don’t know, mind controlled vehicles. Let’s go with it.

Tom White:
Yeah, yeah. Well, I mean, people are running tests, aren’t they? You know, we’ve had, you know, UAVs, delivery, etc. At the pace at which this is going, it can’t be 30, 40, 50 years, right? It probably has to be sooner. Obviously, there’s a safety to trial aspect, so on and so forth to governments. But, you know, I think… It, you know, everything comes back to SOAFEE, doesn’t it, actually about why that’s really, really important to do at scale to learn from the data that’s out there and being collaborative.

Dipti Vachani:
Well, and SOAFEE used in industrial. So one of the things you mentioned is delivery, whether it be transportation, whether it be a factory, whether it’s just moving inventory around in factory floors, all of those problems are a robotics problem, right? It, at the end of the day, is a robotics problem, and, or a robotics question. And that’s fundamentally what Sophie is looking to do. It is to create an environment such that you can deploy software safely and reuse software in a safe, real-time environment at the edge.

Tom White:
Yes.

Dipti Vachani:
And most of us think of a car, but it’s no different in a factory, which is moving heavy machine or things of that sort.

Dipti Vachani:
And so we’re seeing it more pervasively across the industry.

Tom White:
The embedded engineer of me from years ago is screaming out the cost saving and time saving benefits of code reuse as well, right? Without having to do it from scratch and refactoring big chunks of code. So I think that’s really, really critical for this. It’s not only time collaborative, but it’s cost, right? So it’s all those aspects.

Dipti Vachani:
I often say software defined vehicles is survival for the aliens. And I fundamentally believe it. The exponential problem of software deployment and unique software for every single vehicle or not being able to update that software is not solved unless you have a software defined vehicle. They have to, there isn’t a choice. This isn’t a choice anymore. This is a survival element of the only way we can reliably deploy software is to have a vehicle that’s software deployed defined in the future because of the vast amount of software that’s required.

Tom White:
Yeah. Dipti, what a way to come and wrap up the podcast today. Absolutely

Dipti Vachani:
Are we ready to wrap up already? We can keep going. Hahaha. I’m having fun. I just got started.

Tom White:
We could go on for hours. We really could.

Tom White:
But we haven’t finished just yet. I’ve got some more questions for you.

Tom White:
we’ve got the wrap-up questions.

Dipti Vachani:
Okay.

Tom White:
But honestly, your passion, knowledge and the way that you articulate this is really impressive, right? So I really appreciate the time. Deepthi, as we come to the end of the podcast, the near the end of the podcast, not quite there yet, we always ask guest questions about predictions for IoT in 2023, things that they would either personally love to see actually come into fruition, or perhaps what they’re hearing at the moment that’s gonna come in by the end of the year. Do you have anything specifically to IoT, either with an automotive salon or not, that you would love to happen in 2020?

Dipti Vachani:
You know, industry and government collaboration, that’s what it’s gonna take to truly deploy IoT. And we’ve had it, but I would like to see in 2023 us to make a significant progress and deploying some standards, deploying easy ways for us to scale in IoT, because we still have a scale problem. We still have a scale problem. We’ve been trying to solve it for a long time, but it’s going to take the industry coming together to truly solve that scale problem.

Tom White:
couldn’t agree anymore. It comes up in 50%, if not more, the podcast that we do. And it’s the reason why IOT, it’s the reason why we’re still talking about it. And it’s not actually just doing its thing, if that makes sense.

Dipti Vachani:
Yeah, no, absolutely. And what we’ve seen quite often is we call it wild gardens.

Tom White:
Hmm.

Dipti Vachani:
It’s a deployment, but it’s really not at scale because it’s really done bespoke. And what we need to see is that to open up so that all participants can be involved in it.

Tom White:
Yeah, absolutely. And we have a question from one of our viewers. So we always ask our viewers in advance if they would like to submit any questions to guests. And this is one we have for you, Dipti. And I’ll read it verbatim. What role can software-defying vehicles play in advancing the concept of smart cities? And how can they help cities become more sustainable, connected, and efficient?

Dipti Vachani:
Well, of course, there is, and that’s one of the things I talked about as the next thing for software-defined vehicles is this collaboration with smart cities, right? The ability to know through cameras and streetlights where traffic is or be able to predict where deadlock is or be able to predict. to make things safer if we know that this area isn’t safer, something else is going on. It’s just this information sharing, data sharing between the city and the car is critical to bringing smart cities to light. And I think that’s another area where we’ll start to see collaboration here in 2023, and hopefully that continues into the future.

Tom White:
Yeah, it’s that cross pollination, isn’t it, of devices, data sharing, standards, as you’ve mentioned, and absolutely on the money completely. Completely agree with you. A couple of quickfire questions, if I can.

Dipti Vachani:
Okay.

Tom White:
A quote you live by.

Dipti Vachani:
Uh If you believe the world is good, or you believe the world is evil, you’re right. I’d like to choose the world is good. Ha ha ha. Ha

Tom White:
That’s good.

Dipti Vachani:
ha.

Tom White:
It’s quite striking that actually isn’t it? It’s got me thinking about it. I wonder which way it is. I hope it’s the first one.

Dipti Vachani:
Yeah.

Tom White:
If you could learn a new skill, any skill, what would it be and why?

Dipti Vachani:
I want to learn how to fly. I want to learn how to fly a plane.

Tom White:
Okay.

Dipti Vachani:
It’s one of those things I’ve been trying to make enough time to go do is get my pilot’s license. I want to get out of cars and get into a plane. I don’t know.

Tom White:
Nice.

Dipti Vachani:
It’s the next thing I’m going to do.

Tom White:
Do you know what, I can relate to that because I’m not the keenest flyer in the world. And I said to myself one day, I’ll learn to fly, right? Because if I can learn to fly, then perhaps I’ll be more comfortable with flying regularly, right? So

Dipti Vachani:
I fly quite a bit. I don’t I fall asleep as I get on the plane. So I don’t worry.

Tom White:
You’re one of those people.

Dipti Vachani:
I’m one of those. It’s survival.

Tom White:
It was going really well as well on the podcast, you know. And now I’m just jealous that you can do that.

Dipti Vachani:
Yeah, no, I am. I, I’m, I’m actually scared of flying a plane. So I tend to I tend to take on something that scares the bezebus out of me. Like you know, just to just to push myself a little bit. So it’ll challenge all aspects.

Tom White:
Do you know what, you’ve got to face your fears though, right? Because it’s the only way that you can grow. When I started this podcast as well, I was always at the start, I was like, oh wow, like you’re on camera now. Whereas if you just have a normal conversation with someone, like it’s fine, but then you put a camera in front of someone’s face and all of a sudden it’s like, ah, da da, you know, and you’re shaking.

Tom White:
My final question, if you could have a conversation with any renowned inventor or innovator living all past, who would that be and why?

Dipti Vachani:
You know, I’m going to give a typical answer, and I feel like I almost don’t want to give it because it seems like it’s obvious, but it wasn’t one everybody would give. But it is Steve Jobs. And you know, I find that the perfection in which he operated while still innovating, right? It’s very hard to find that perfect balance, or even a balance. When you’re innovating something new, when you’re doing something brand new, it’s rare that you can do it perfect at the same time as you’re doing it new, right? It’s kind of like, just get it to work usually. But if you think about some of the first technologies that were deployed, the beauty in which it was deployed and how perfect it was, that intercept between beauty perfection and new. I’ve never seen that happen the way that he was able to pull that together. So I find that very interesting.

Dipti Vachani:
It’s easy to innovate, easier. It’s okay to innovate. It’s okay to make it beautiful. It’s cool when you make it perfect, but how do you do all three? That’s pretty damn cool.

Tom White:
Yeah, you know, a few people have said Steve Jobs in these questions. Yeah,

Dipti Vachani:
That’s what I was afraid of and I hate giving the same answer to everyone.

Tom White:
But you explained it in a very different way, actually. Um, and it was that that’s quite striking to think about it that way. Cause I never, I never remembered. Not to say it’s not true, but I never remembered Steve in that way, right? You know, um, in being able to do all of those things at that. in that way and you know you’re right and it is quite a gift isn’t it?

Dipti Vachani:
Well, if you look at this thing, it’s a beauty, the phone, the iPhone, the

Tom White:
Hmm.

Dipti Vachani:
first ones, it’s just, it’s beautiful, and

Tom White:
Hmm.

Dipti Vachani:
just how it appears to the eye. And then I was, I’m a chip head, and I was working on some of the chip technologies, the silicon technologies that was going on, that went into the iPhone, and he expected perfection, even in the packaging of the silicon that came into the phone. which is that you had to make sure there were no scratches, even physical scratches that had nothing to do with the functionality, because it had to be beautiful, which is not anything anyone else has ever done

Tom White:
No,

Dipti Vachani:
that I know.

Tom White:
well, standards are important

Dipti Vachani:
Yes, and

Tom White:
in

Dipti Vachani:
he had very high ones.

Tom White:
all ways. Yeah, in all ways, yeah. Tipti, it’s been absolutely fantastic having you on the show. I

Dipti Vachani:
Thank

Tom White:
genuinely

Dipti Vachani:
you for having

Tom White:
mean

Dipti Vachani:
me.

Tom White:
that. You speak with such confidence and charm about the industry. It’s been a real pleasure.

Dipti Vachani:
Thank you.

Tom White:
You’re very welcome. Where can people find out more about you and Arm and SOAFEE, et cetera, perhaps somewhere online?

Dipti Vachani:
arm.com and there’s all the information’s out there.

Tom White:
Excellent.

Dipti Vachani:
Thank

Tom White:
Dipti,

Dipti Vachani:
you very

Tom White:
thank

Dipti Vachani:
much.

Tom White:
you. Thank you for coming on the IoT Podcast.

Dipti Vachani:
Thank you for having me.