Enjoy the Journey: Ari Mahpour's Passion for DIY Engineering

 

Join us as we explore the incredible DIY engineering journey of Ari Mahpour, a designer at Rivian and frequent Octopart contributor, and discover his passion for home hobby projects. From his early days with RadioShack kits to his advanced DIY electronics and DevOps work, Ari shares insights and tips that will inspire any aspiring engineer.

Listen to the Episode:

Watch the Episode:

Episode Highlights:

• Raspberry Pi and Arduino tutorials
• Transitioning from government to private sector
• Balancing fast-paced startup work with high-quality output
• The importance of community in DIY projects

Links:

Connect with Ari here

Watch Ari's videos here

Transcript:

James: Hi everyone, this is James from the CTRL+Listen podcast, brought to you by Octopart. I'm here with my co-host, Joseph Passmore. Today we actually have a, a bit of a different guest for you taking the show in a unique direction. It's Ari Maur. He is a designer at Rivian and he's here to do in a different capacity, though he is the home hobbyist king as we like to refer to him as he's done some amazing content for Octopart over the, the time that he's been working with us. Go check that out on our YouTube channel. But Ari, thank you so much for coming on. Welcome to the show.

Ari: Great, thank you. Happy to be here.

James: Anytime.

Joseph: Just because we, we talk about the DIY side of things today. I, I wanted to kinda ask you, what was your engineering journey? How'd you get to where you are and where'd you start?

Ari: Yeah, good question. I, so I started, my dad actually is an electrical engineer and I started on the, you know, the very basic Radio Shack kits when I was a little, very, very little kid. My mom worked at Apple early on, so we got an Apple two and I played all day on the computer at age, I don't know, three. And I just, I just got into computers, loved it. I loved tearing apart computers and so I really wanted to do something that was both hardware and software related. So I went through college and initially started working at NASA, JPL, the Jet Propulsion Laboratory, where I spent about half my career working on power, power, electronics and radar systems. And then eventually moving around a lot doing electronics design and software and embedded systems at PGAs. And then as I progressed through my career, I started to do DevOps, things like that. And so I found a, found a happy medium where at Rivian and some of the previous companies I've been at, I do a little mix of DevOps and infrastructure and embedded systems and hardware and everything kind of all together. But my, my real love and true passion has been, has been to home hobby when I, when I have some extra time with my, my home lab over here just to play around with electronics and build stuff for fun.

James: That's awesome. It's an exciting journey. What was it like shifting from government sector to private sector in that space?

Ari: Dizzying and exhausting. I remember when I joined my first startup at Virgin Hyperloop, it was like very, very fast paced compared to working in government and military and aerospace and space. But I loved it. I love the dynamic aspect of it. I love moving really, really quickly. At NASA, we were very, very meticulous. 'Cause you know, once you send a spacecraft out, you can't just bring it back and, you know, do a couple fixes and then send it back. So I loved both, but I also, I, right now at Rivian, I also really love like the fast-paced and the dynamic aspect of it, but still, you know, we're delivering a car and so that's, you know, it's a, it's a mission-critical safety system. And so you still have to be really, really careful about the product that you're delivering has to be super high quality. And, and I love that kind of mix of both the rapid iterations, but also like really high fidelity, high level, high-quality products that you deliver.

Joseph: Could you tell us a little bit about the effectiveness of simplicity done right?

Ari: Ah, so simplicity done, right. So I, I think one of the things that I see, especially as I do like DIY projects is that one of the things when I try to take a topic, especially a really complex topic and distill it, is one of the biggest challenges is being able to like, articulate that well enough to your audience or to just other engineers. And especially when I mentor other engineers, I say, you know, if you don't understand, one of the most important principles I learned in school was that my UI UX teacher at USC when I was in my master's program, he said, if you don't understand how to use the interface, that's not a failure on you. That's a failure on the designer. And I feel like it's the same thing when you're trying to articulate a point. Richard Feynman is known, you know, in, in teaching in Caltech, he's known for being able to take complex topics in physics and being able to like distill that and teach that to his intro classes when they, they're undergrad students. So being able to take a project and make it as simple as possible, break it down into small pieces, being able to articulate that I think is really, really important. And so there are different ways of doing that. There are different ways of being able to distill that information. And so giving, for example, lots of metaphors or ideas around how you're gonna introduce that topic, I think is really, really important to explain that you have to take a moment and, and really understand, you know, this, who is the person I'm explaining this to? Are you a junior engineer? Are you very advanced? Do you understand this topic? Maybe you understand that topic and try to explain that in the language that they understand. And I think that's a really core component of being able to do things simply. You have to make sure you explain it in their language and explain it very, very, very clear and concise, especially when you're walking through like a DIY project.

Joseph: Could you give us an example?

Ari: Yeah, actually, so something I'm, I'm, I'm to, I'm toying with the idea as, I don't know if you've seen some of the videos I've been doing recently. I've been introducing a lot of AI into the, the engineering process. Two of the really exciting videos and tutorials that I've been working on recently that they're already published. One is having generative AI control all of my lab equipment, which sounds really dangerous, but it, there's a lot of, there's a lot of security measures around that. Controlling my lab equipment and doing a lot of testing for me. And another one is writing embedded firmware and then running that on the actual target. One thing that I'm toying with right now is looking at how to build neural networks like from scratch. There's like a bazillion tutorials online, but like all of them are so, so complicated. And so the first thing I did is I went to Chachi PT and I said, you know, ex I'm an engineer, you know, ex I understand the manufacturing process or I understand this process. Explain to me in my language that I understand how a neural network works. 'Cause I wanna, you know, implement this on an FPGA, for example. And so it, it basically tells me, okay, you're walking through the manufacturing process, you have a piece of sheet metal, for example, and first in the sheet metal you have to like stamp it or you have to, you know, clean it or whatever it is. And then in the next stage of the manufacturing process, you have to then shape it or you have to color it, things like that. And then the next stage of the process, you then put it together or you weld it. So at each stage there's, there's a different process. And then finally at the very end, and as you go through that process, you have pieces that maybe didn't pass, muster, didn't pass quality. So eventually at the end of the line you have a finished product that you know exactly matches the criteria that you're looking for. Same type of thing with like a neural network. If you, you start with a certain, like let's say a handwritten number for example, and you say, does this, you know, does this look like a nine? Does it have the shape? Does it have this? And you go through neurons or in your network, a neural network, and you go step by step and then you've kind of filter down or go through that process. And then eventually you get to the point where the output is the product that you're looking for. And all the things that don't match kinda get filtered out through that process. I'm still kind of working with metaphors, but like, this is where I start and then, you know, maybe I didn't get that a hundred percent right. So I go talk to my friends who are AI experts and I try to try to distill that and figure out, you know, how can we make that simpler? How can we explain that to people in a, in a much more, you know, palatable way. And so I think that's, that's really, really important. Otherwise you can just keep talking and it'll go way over their head and like, what's the point you weren't able to communicate or articulate in their language. It's like, if I speak one language and you, you only understand a different one, and I could talk till the cows come home, but if you don't understand me, what's the point? I think that's really, really important when you're trying to explain a, especially a complex topic.

James: Or have you ever seen the series Silicon Valley?

Ari: Yeah.

James: You just remind me exactly of the Pied Piper issue that they had on that.

Ari: Yeah, yeah, yeah,

James: That's great.

Ari: It's the, the whole, you need different parts of a team because you have engineers that are really good at the technical side, but they may not be able to explain, like you just said, to in layman's terms enough for an actual user to be able to actually make the product functional on market, so.

Joseph: Right, right. Exactly.

James: So, okay, this kind of brings us into another topic that we were talking about before we started recording, which is you said you had a number of ways that you view found to enhance the DIY process.

Ari: Yeah, so when I first start with the DIY process, first of all, again, especially if it's something complicated, like some of the, some of the tutorials are very, very simple, like how to get started with Arduinos, you know, but some of them could be a little bit more complicated, especially when I bring in some of the more sophisticated FPGA stuff or some of the AI stuff that we require. Even sometimes I have multiple DIY projects that kind of build on each other. And so I think that process a starting to really do your investigation. So before you get started looking at it, trying to get an overview, again, AI, you know, of course does hallucinate, but it can be a really great starting point. A chat GPTA claw, a Gemini or whatever, you know, or one of these like large language models can really help you kind of break parts down because you can ask those questions and then you don't feel so dumb to go ask more intelligent questions to your AI experts. Like I plan on doing, and again, like the metaphor I gave, maybe it's not a hundred percent right, but at least it's a talking point. Now you can start tweaking. So you start with that process, you say, okay, you know, I have these pieces and I really want to do this. And then you start to investigate, you know, back before generative AI, we would read articles or we would get books from the library, I don't know if the audience knows what books are, but books we would get from the libraries, not eBooks. And we would pick them up and we would read and we would try to study on this topic and then start to prototype. Now we have the ability to really grab information, ask intelligent questions, and then start to piece things together and then start to play around with it. So using evaluation boards, using DIY components like raspberry pies, Arduinos, these things to like rapidly prototype concepts and prove out these ideas, I think is, is also really really key to enhance that do it yourself process because you can, you can come up with the end result without all of the effort. At least you can show, show kind of an, an idea. One of the things, if you're familiar with the design institute from Stanford, IDO is a company that came out of that is the idea of the, the industrial design, the UI UX, they, they're, they're obsessed with rapid prototyping and being able to come up with a minimum viable product that Eric Reese, a minimum viable product to be able to show that you can do that. And I think with the D-I-D-I-Y process, you wanna just iterate really quickly with these components like raspberry pies and Arduinos and evaluation boards to get that minimum viable product and then you can go back and then make it, you know, even fancier and bigger and things like that.

James: There was something I actually thought of while you're talking about that I, I I wanna ask you is from a DIY perspective, is it practical in any sense to employ 3D printing into that process to help you get past some of those limitations?

Ari: Yeah, absolutely. I've got my, I've actually got my 3D printer right up there, as you can see. Yeah. And it's, it's really, really helpful to, to use for prototyping. Yeah. Especially when you're trying to build enclosures and things like that or holders, you know, it's, it's a little bit of a learning curve. If you're trying to learn how to do 3D CAD yourself. You could actually have generative AI build 3D models for you. Like in open s I've had it, I've had it design like models that I can 3D print writing open SCA for me. But being able to rapidly prototype that rather than, you know, getting like a wood box and trying to, you know, cut it up or saw it out or plastic box or, you know, trying to purchase like a sheet metal box or something at what happens if it doesn't fit. So being able to get all those fittings and everything with iterating many, many, many times just right here, like in your, you know, in your office, in your home office or in your living room, whatever it is, that helps a lot. I found that helps a lot. And the cost of 3D printers have really come down in price. You don't need the $10,000, a hundred thousand dollars 3D printer, like a couple hundred bucks is probably good enough for what you're doing.

Joseph: You mentioned Raspberry Pi and Arduinos. These are, these are key DIY components 'cause you tell us a little more about them and how you use them?

Ari: Yeah. Yeah. So, so Raspberry Pi, so Arduinos are basically microcontrollers that have a lot of different peripherals on them. A to D converters, which basically take analog signals, you know, voltages or things that would come out of sensors and then it can read that directly on, on your Arduino or you can control motors, for example with an Arduino. Raspberry PIs have a full like operating system on top of it. So it's a lot easier, much more accessible to write code in many different languages versus an Arduino where you just write it in like C or C plus plus. And so being able to rapidly prototype a concept with these little development boards, these development kits, you can very quickly hook up to different external sensors. You can build motor controllers, you can build robots, you know, even with these devices without having to spin up a whole PCB or being able to include together a whole bunch of eval eval boards in order to do that. And the, the, I mean really the, the best part about using these type of boards, there are other ones out there too. There's the orange pie nowadays, there's a whole bunch of beagle bone, but using these kind of development, not really development kits, but whole ecosystems like the Arduinos and raspberry pies is that the community out there is just absolutely phenomenal. You can find any project under the sun using one of these boards and like that is so cool because you can, you can get started so quickly. And again, like I said, going back to how to enhance that DIY process, if you can look and see what other people are doing that are related or even tangentially related to what you're doing, that's a big jumpstart than like trying to do it all the way from scratch. So these, these kind of boards are like paramount to doing a DIY project.

James: I think that's, that's another topic we could maybe touch on is the community for DIY for anyone looking to get into it. I mean, from what you've said and what I've seen online myself is, is is pretty open and welcoming. People are pretty happy to help out if people have questions.

Ari: Yeah, yeah, absolutely. Actually there was this DIY project that I worked on, which was a, a kiln series where we did a, we did a, basically my wife, she's a ceramicist, she's a potter and she has a few kilns and it was really, they're a little bit old and so it's, it's very annoying to sit there and babysit the kiln if anybody's not familiar with it. It's basically a hot oven that goes to like 2200 degrees Fahrenheit and or sometimes even hotter. And you basically have to sit there and babysit it towards the end to make sure it hits that temperature and then you turn it off or you have things like different cones, like different pieces that, that can sit in there that it'll eventually turn it off on its own, but still you wanna monitor it, you wanna, if you're not sure or if you didn't do everything right, it's good to kind of like observe that. And so the community, this is a great example. Somebody had put together a repository using a raspberry pie on GitHub and they built a kiln controller and what they did the software for kiln to control the, the flow, the oven controller for that. And they based that off of a community project, which was for a reflow oven that people used for a toaster. And then they were able to reflow solder in order to paste parts on their boards for a, a printed circuit board. And so that guy took actually his name, Jason Bruce from, from GitHub, he took that and then redid it with a whole bunch of people from the community who com contributed to that project. He took that and then built a kiln controller like algorithm and a web interface on top of that. And then I was able to then leverage that, make some tweaks obviously, and then work around all the electronics around that to get the whole series going. But like without the community, it would've taken me much, much, much longer to put that whole interface together. So being able to like springboard off of that, again, I did my investigation, I told my wife, oh, let's automate it. And then I looked online and I was like, oh, somebody has already done this project, I just have to worry about all the electronics and how to wire it and tear down my kiln and put it back together. People on YouTube also did that too. So I was able to like leverage some of that, leverage some of the code, and then I was able to put my whole project together. I'm like, that saved me maybe hundreds of hours of research and lots of possibly melting stuff or blowing stuff up because I was able to observe what, what went wrong and what didn't with other people's projects. So leveraging the community was like so, so, so critical for that project.

James: But now you've become a reference for the community yourself with that, the kiln guide.

Ari: I mean, I'm just piggybacking off of everybody else's successes, so hopefully somebody else will learn from this and be able to enhance it even more than, than what I've done.

Joseph: Yeah, if anyone wants to watch that series, the entire, I think the whole thing's out now on YouTube, right? Every episode.

Ari: Yeah. Yeah. All, all installments are on the Octopart YouTube channel. So you can see Ari's journey from start to finish there if you wanna check that out.

James: And you also done some raspberry pie and Arduino stuff too. Some really good walkthroughs and tutorials in that space as well.

Ari: Yeah, yeah, I've, I've done for both raspberry pies and Arduinos, I've also done for some more like complex boards, FPGA boards that are capable of like machine learning, machine vision, stuff like that. And I think it's really, really important to be able to, for a lot of people that look at those videos, they say, oh well, you know, of course I could just like find an on a tutorial or I can just like fumble through it. But I think what's really important, especially one, the one for example with the CREA two 60 psalm, which is basically an FPGA with a processor together, it's, it's a more complex board. The, the getting started guide with it was not terribly complicated, but if you were to go look at the docs, at least at the time of writing the tutorial and doing the video, if you were to look at the docs, you have to like go here for this piece and here for that piece and then follow this link which follows then that link and then that link being able to kind of put it together and just like a, a flow have a flow from A to ZI think is really, really important. Even if it's simple, you know, you might be a really sophisticated engineer, but to not have to spend those brain cycles to that, oh, I have to go find this here and that here, having it in one place with a nice flow going A to ZI think is, can be really beneficial for, for a lot of engineers, beginner or expert. And just to having that getting started, I think, I think could really enhance, like we said, the DIY project.

Joseph: Could you tell us a bit about setting up your lab and how it's changed over time and what improvements you've made?

Ari: Yeah, yeah. So that was quite a journey, not quite the journey ahead from getting into engineering, but, okay, let's rewind back to March of 2020. I was working at, I was working a, a few companies ago and the pandemic hit and I was working on hardware and I didn't have a lab set up and everything shut down. I don't know what to do. And, you know, I'm on kind of a tight budget. I still wanna do home, you know, at, at like work, you know, sometimes if I needed to like tweak something that I had at a home home project, I was able to like get into the lab after work and solder something up that was gone. I couldn't get into the lab at work. And so now I'm stuck at home and I've got no equipment, but I also don't have lots of cash. You know, I can't, can't build a 10, 30, 40, $50,000 lab at home, nor do I have the space. So I kind of went on this journey where I built a home lab on a shoestring and I looked for components, parts. As you can see over here, I, I have a lot of equipment, but I tried to find kind of the best bang for your buck in some areas, like my soldering iron, I, I cheaped out and bought junky soldering irons and learned the hard way that like, that's the place where you need to like spend the money. But then other things like different components or different wires or things like that, like some of them you can buy kind of cheap, especially if you're buying like on eBay or Amazon or things like that. Especially for, for prototyping and hobbyists. So over time I kind of like collected things. I built different shelving, I built a lot of custom shelving over here. Didn't have much going on during the beginning of the pandemic, so, so I did a lot of retrofitting in my lab over here and, and I was able to really build up like a full-scale lab with, with not a high investment. Same thing with the equipment. You know, I was able to get equipment either used or I was able to get equipment that wasn't so high cost, but still gave me the biggest bang for my buck. And again, like do I need a hundred thousand dollars oscilloscope for a home lab project? No, most of you don't need that. If you do like, I don't know, you should be charging a lot of money for those home projects. I, I'm not sure what to tell you, but, but you really don't need to be spending that kind of money on such crazy equipment unless you're using for your business of course. So, you know, you should be able to get away with, with doing a lot with a little, and that's kind of like what I learned as I, as I went through this process.

Joseph: You said that the, the soldering iron were a place where you, you tried to, to skimp out the money and you realized that you, you shouldn't have, was there something where, which you initially put a lot of money into which you discovered later you didn't need to?

Ari: I, so I'm pretty cost-conscious, so I would say that most of the stuff I bought, I was able to maximize the most out of it. Maybe, maybe I kind of splurged on my oscilloscope. I spent maybe over a thousand bucks on my oscilloscope, which actually in the scheme of things for what it can do is, is not that much. Do I use it like every day and, and use the maximum amount that I could, you know, squeeze out of it? Probably not, you know, I probably could have gotten away with my analog discovery too, which is a couple hundred bucks and connects directly to my computer. So that was kind of my gift to me. So sometimes, you know, the equipment, especially when it comes to instruments, it's easy to overspend on those and so you have to like really think about it. But I did graduate from like ridiculously cheap old school power supplies to nicer power supplies. And, but you know, you, you just have to, you have to do assess that trade-off, what is it that you really need? You know, do you really need that or can you get away with something a little bit cheaper to, to make sure, you know, you can pay the bills that month.

James: So to kind of wrap things up, there's a topic we definitely wanted to cover, which was empowering beginners that you mentioned before we started here, set, you know, setting attainable goals as someone who's starting in DIY. Do you wanna expand on that a little?

Ari: Yeah, so I think when you take a really large complicated topic, like I talked about neural networks, which I'm still learning, I'm gaga over AI, but there's so much to learn and it's changing so fast. But when you take a really complex topic, it's, it can feel incredibly or an in or a really complicated DIY project, it feel incredibly overwhelming. Like, whoa, this is totally out of my depth. That's really hard, especially if you're a beginner. Like where do you start? And so I think it's, it's super, super important to just read and learn and talk to people and engage people, engage in the community post on Reddit, post on, I remember when I first started, I mean now there's so many resources, but like Ada Fruit where I would buy a lot of my prototyping they have and Spark Fund too. Like you could just post on the forums and say, Hey, I have this project, you know, can somebody help me think about how to do this? And like people would respond, you know, within a couple hours or within the day. And that was like super, super, super helpful when I first started. And people just scour the forums, just they, 'cause they wanna help, like the engineering community are a lot of, there are a lot of benevolent engineers that are really seasoned that love to just teach and love to help other people, which is super nice, especially in the world we live in today to see so many benevolent people just want to share. So absolutely, positively take advantage of that, especially if you're a beginner. Just, just ask, ask, ask, ask. The only dumb question is the one you never ask. It's really important to just ask people, talk to people. I've had people reach out to me via email, via LinkedIn about my tutorials online and I I always love answering them. I love to get those questions. Number two is you have to just start somewhere. So even though it's incredibly daunting and you don't understand all the topics or you don't understand how to build that full, you know, DIY project, just start with something small, prove out a little piece of it. If you're trying to build like a full-scale walking robot that can like get in a car and drive autonomously and like fly to the moon, maybe start with like an Arduino based, you know, project that can control motors and maybe move some wheels, you know, like that's a baby step. You know, there are plenty of tutorials with that. And then you graduate and then you iterate the agile process. Just keep delivering something every time, but iterate into better and higher, more sophisticated products. So asking, starting somewhere and then just learning again, part of that agile process, the DevOps process is like learning just from every iteration. Just learn as much as you can. What went wrong, what went, went right, and what you need to do in order to like get to that next stage. And again, just keep getting as much as feedback as possible. I think being able to do that, leveraging technologies that are given to you with open-source communities like the Arduinos, the raspberry pies, the eval kits, things like that. Using generative AI as your pair so you can ask even more intelligent questions later if you're too embarrassed to ask them. All these different pieces really have moved the needle for DIY, you know, than it did like 15, 20 years ago. So it's, it's a, it's a totally world and, and engineering is so accessible with, with the YouTubes and the, and the recordings like these and the projects online, it's become so accessible to so many people. And like, again, people so benevolent that they wanna share is just, it's such a, an incredible thing and it's just such an incredible time to be in engineering and, and just play around with that. So I think it's a, it's a really exciting time and, and if you don't feel like you know where to start, just start. Because, because it's, it's just so accessible these days.

James: Yeah. That, that's our first step is just taking the first step every time.

Ari: Exactly.

James: Well, is there anything else, like any message you wanna give to DIY engineers starting out before we wrap up? Anything on your mind?

Ari: Like I said, I mean, I, I've pretty much summed it up, but I think, I think the most important thing is just reaching out to your peers, whether it's people in your company. I had so many mentors, especially when I started at NASA, JPL, like I had, I was so, so, so fortunate to have just absolutely incredible mentors. If I went to name all of them, I'm sure I would miss some of them, but like just amazing people who guided me through my engineering journey. I should have started with that throughout my whole, you know, engineering career and, and being able to leverage them. So people who are more advanced than you are in your workplace, or again, being able to leverage the community or just sending emails to people or LinkedIn messages to people that you like, their videos, you know, their DIY projects. Hint, hint. But it, you know, just, you gotta get started somewhere and you just gotta engage people because like I said, people are, you'd be surprised how willing they are to share and help. And it's important once you figure that out. My, my mentor from JPL, he would always, he'd spend countless hours, countless hours working with me after I had already left on my, all my hobby projects, especially my early ones. And I said, you know, you, why are you spending so much time? Like, you have a job, you have a lot of things to do. Like, I feel really, really bad. And he says, no, you shouldn't feel bad because I know when you become, as, you know, advanced as an engineer as I am one day, you're gonna pay it forward and share it with everyone else. And his name was David Hawkins. He was my, he was really one of my, my biggest mentors at JPL. And, and he, he would just spend countless hours and being able to, to share that with me and now, now I feel compelled to share that with everyone else. And that's part of the impetus of wanting to share all these DIY and really empower engineers out there.

James: That's awesome. Yeah. And mentorship's fantastic. I've been lucky enough in past roles where I've had mentors as well and it makes a huge difference to your career.

Ari: Yeah, absolutely.

James: Well, that, that kind of brings us to the end of the discussion, but just for anyone who does want to contact you with questions or anything like that, is, is LinkedIn the best place for that?

Ari: Yeah, yeah, it's great. Absolutely. Please feel free to connect bombarding with messages, especially engineering technical questions. Like I love, I love to talk shop any day of the week.

James: Awesome. And for anyone who wants to see more of, of Aria's content, check out the Octopart YouTube channel. There's a very regular stream of stuff that he, he submits to us and it's all really useful stuff people seem to really enjoy and get a lot out of it. So check that out. And again, Ari, thank you so much for coming on the show. Really appreciate

Ari: Taking time. Thank you so much for having me.

James: Yeah, yeah. Thank you. Anytime. Have a great time.

Ari: Thank you.

James: You're always, always welcome back on, I'm sure there's DIY is always gonna be happening, so there's more to talk about.

Ari: Thanks.

James: But for anyone listening, thanks for tuning in and we'll have another guest for you next week.