Embedded - 516: Voices From the Cataclysms of the Universe

Episode Date: December 12, 2025

You can find the embedded.fm survey here: https://forms.gle/xPp8YkX8MXvgUENM7 and please consider donating to Elevate Tutoring here: http://donorbox.org/embedded-fm-supports-elevate-tutoring-866250. The survey will be up until the Dec 25th 2025. Sophi Kravitz joined us to talk about art, science, and engineering.  You can see Messages from Space on Sophi's website /sophikravitz.com). A subset of the artwork had a short stay for a demo at Chabot Space & Science Center. The completed work will be shown in 2026. Sophi mentioned collaborating with two sonic environment artists Sofy Yuditskaya and Ria Rajan. Geiger–Müller tube is an ionizing radiation detector. Cosmic rays move through space at nearly the speed of light, generally originating far away and long ago. You can also see them in a cloud chamber, like the one at San Francisco's Exploratorium. We also talked about using sculpting in Blender (there are many online video introductions). Sophi does EE consulting and system quality checking at her company Greenlight. The show this week is sponsored by us. And you. Please consider supporting Embedded.fm on Ko-fi or Patreon. Or tell a friend about the show. Transcript  

Before we get started, past guest and friend of the show, Nathan Jones, has graciously put together another Embedded FM annual survey. We'd appreciate your feedback and will be collecting responses through December 25th. As part of the survey, we're raising funds for elevate tutoring by matching your contributions up to $5,000. Thank you for your support and thoughts, as always. Welcome to Embedded. I'm Elysia White, alongside Christopher White. Our guest this week is Sophie Kravitz. We're going to talk about art and engineering and consulting in Cosmic Rise.

Hi, Sophie. It's great to talk to you again. Yeah, it's great to be here again. Thanks. Could you tell us about yourself as if we met at the Chabot Observatory looking at various space-related art? Sure. Hi, I'm Sophie Kravitz. I'm an electrical engineer and artist and I work at the intersection of those two worlds and also sometimes separately. As an artist, I design and build projects that detect mostly invisible forces cosmic rays from distant stars, electromagnetic fields, environmental data. And I like to build sculptural experiences that explore. Various things that come from that data, but I've been working with emotion like hope, longing, and most recently, science fiction, human futures.

And as an engineer, I work as a consultant for companies or startups where many hats, wearing many hats as a necessity. I usually work in some form of hardware design, whether I'm designing products or testing, and sometimes I write content about the work. I took 2025 off from engineering consulting to focus on my art practice and I'm getting back into hardware work in 26. All right. And we will be asking about pretty much all of that, I think. But let's do lightning round first. Are you ready? I'm ready. Hardware or software? Hardware, definitely. Art or engineering? Both. Marketing or engineering? Engineering. What is your preferred way to share your projects? That's actually a good question because I've cycled through various platforms, and right

now I'm using GitHub quite a bit. I have quite a lot of stuff still on Hackaday. I'm going to leave it there. I also love that platform. And I put everything on my website, but I think that personal websites are, they're good when you direct someone to go there. Yeah. A favorite source of cosmic rays. My own art, of course. What superpower should they bestow upon you? The art or the cosmic rays are both. Either one. What superpower? The ability to feel the future. Unicorns or griffons? Unicorns. Do you complete one project or start a dozen? Complete one. Do you have a tip everyone should know? Hmm. I feel like my favorite thing to do right now is not.

going to be relevant in probably a year. But every time I'm learning a new software, I have an AI window up as a help chatbot. And I got fluent in Blender this year from doing that and I'm using it now for AutoCAD. And it's amazing. But I imagine that by next year, every software is going to have an AI chatbot in it, like inside, to help you do that. She's one of the few complete one projectors. I feel like we have a survey now that says something statistical about whether people start dozens of projects and don't finish them. I mean, both ways are very, very valid. Oh, sure, sure.

It's just it's rare for some of this. I think the people who complete one project say no to the podcast because that's yet another project. Okay. You're right. We have some sampling bias. Yes. Yeah, I would agree. Although sometimes I'll work on two projects at the same time.

I have two projects going at the same time, usually. Well, you talk about art and engineering, and those are separate projects. Kind of. I mean, I think that the artist mind of curiosity as a sort of baseline is the same as the engineering or the science mind. It all starts with some kind of question. or inquiry. This week's show is sponsored by us and you. If you'd like to become a member on Patreon or coffee,

$1 gets you access to the exclusive embedded FM Slack where you can find other like-minded, kind, and helpful listeners. You can ask questions, get feedback on ideas, make jokes, or just complain. And for $5 a month, you receive access to an always, always add-free stream of the show, and the occasional bonus episodes like The Real Genius episodes, Sleepy Science, and more coming soon. Thank you to everyone who supports us, monetarily, or by telling your friends, that Embedded FM is a good show to listen to.

Thank you for listening. Which brings me to one of the questions I was looking forward to asking on this show, which is, tell me about the sounds you hear from space. Are you asking specifically about my new art project? The one called Messages from Space? Yes. Yes. Okay.

So maybe I'll tell the audience here what that project is in a brief way. That's okay. Please. So Messages from Space is a project where cosmic rays, which are real, not science fiction, particles from space that originate during space violence, like gravitational waves or black holes or any kind of things crashing around in deep space. Those things travel at the speed of light, but from a far distance. So they originate millions of years ago before humanity ever existed.

And that's really what I think is so cool about cosmic rays. they come and hit a sensor on the board that I designed, which goes to an SD card of pre-recorded stories and sounds about 100 years in humanity's future. So these cosmic rays not only come from far away, they come from long ago. Yeah, sort of a galaxy far, far away. Some of them.

Some of them come from our galaxy. too, but so. But, yes, many from the galaxy far far away. There's Star Wars particles. Exactly. That is where I was headed, yes. Yep. And then how do you sense them?

So the circuitry is similar to a Geiger counter, which is a sensor. The sensor I'm using is called a Geiger-Muller tube. And that tube is filled with three gases. two of them are noble and one of them is not and what happens is you charge that gas up inside of the tube the vacuum pack tube up to a very high voltage in my case i have a 375 volt supply on my board when an ionized radiation particle hits that tube it causes a short across the tube, which is basically makes a one or a zero or an on or an off. And so I sense that binary. In a Geiger counter, typically the hits or strikes are counted. And so you might not say it's an alarm

until several thousand strikes have been counted. But instead of that for yours, you trigger. A trigger sounds and light. One of the sounds you trigger is it a meow? It is a meow, yes, a meow and purr. So visually, the circuitry sits inside of a sculpture. And this project is to be realized in 2026. So I've play tested it a couple of times, but it's very much in progress. And the sculptures will probably look something different.

But right now the sculptures look like planets, and one of them has a bunch of cats hanging off of it, and that planet has cat sounds. You have about five, six, a dozen of these planets so far? Yeah, I have six planets, and I have made 15 boards, or I've had the boards made, the first versions I made myself on my very own hot plate, but the yield. was terrible. And you took this smaller version to Chabot, which is an observatory in Berkeley, California. Science Museum, yeah. Science Museum. I brought it to a private event that was held at Chabot Space and Science Center recently.

So it was a large party, and I was able to show this art piece there. A preview. Yeah. Okay. How do you know the cosmic rays are going to come when you're having a demo? Oh, they're always coming. Yeah, go ahead, Chris, yeah. You remember cloud chambers, right?

Do you have any cloud chamber in a museum? They have one at the exploratorium. Say that I don't... This is not the same as the blue radiation. Ternikov? Turenkoff radiation. No, that's from nuclear reactors. This is the...

You get a... Sophie help me out. You get a chamber, and you reduce... it to vacuum and then you get a vapor in it, so it makes like kind of a cloud in it in glass. And then you can actually see the particles as they interact with, they make little trails through the cloud as they come through. And they're coming through every second, every three or four a second, but random directions, yeah. The amount of time, well, let me just back up a little bit. So the way they come, they do

originate in deep space, they come through our atmosphere, and then they split into showers of secondary particles. So what we receive here on space, here on space, here on in Earth, what we get here on Earth is the secondary particles of the first particles splitting up. Not that that makes it any less cool, just being more accurate. And Chris, I actually didn't know that's how Cloud Chambers work, but that sounds about right. I may be making up half of it. So if I'm wrong, somebody let me know. The chamber and the Geiger Moller Moller tube are about the same thing just at scale. Maybe.

All right. Well, I mean, this is what Wikipedia is for. We don't need the exact technology. Yeah, exactly. Yes, exactly. I did just see a cloud chamber at the Exploratorium in San Francisco. Yeah, I love those. They're just, it's, yeah.

The whole idea of being able to detect these things is exciting to me. I don't know. It's just, yeah. Yeah, it is super exciting. And you're right, Chris, they do. arrive constantly and there is a value for it. They come randomly. So they were at one time used to generate random numbers, but they come to us at the rate of about one particle per second per human head volume. Wow, that's a very specific statistic. I've read it in a few different

places. And per human head volume, let me just write that down. So the three of us, Right now, even though we are on different sides of the country, we are each being bombarded at approximately the same rate, but at random times. And so to come back to your question of how do I know it's going to work, that's how I know, because it's happening all the time. Real question is if it stops working, has something gone horribly wrong with the universe, or is it broken? That's right. I mean, did a dome just, like, come over the whole of Earth? Have the aliens arrived or has my battery died? Exactly. Yes. I have to say a very long time ago, where this project originated from was I did an internship about 20 years ago, more than 20 years ago, where I worked for a company who made KVP. meters, which are what measure the effectiveness or the quality of the image in x-ray machines

used for dentists and other medical procedures. I think, I believe these devices are, you can still buy them, but I think they're less relevant now because of the digital technology, but I'm not really sure. However, I worked there, and my first product as an intern was we designed a radiation detector, right after a personal radiation detector. It was called PRD. And it was right after September 11th, 2001. So I tested the radiation detector everywhere, and I also was the person who was writing the firmware

so I could adjust the sensitivity. And so I learned that that thing worked everywhere. It worked in the plane. You get more when you're at a higher altitude. It worked in the subway. It worked. Basically, it worked everywhere. It was kind of like, what a way to show an intern that there's invisible rays bombarding us all the time.

This is why airline pilots get way more radiation than normal people. Yeah. Yeah, that's right. Back to the demo, the smaller demo. So now I accept that not only will it work because there are particles coming all the time, but that they come in showers. So I can expect the whole thing, a number of the units to react at similar times because of the shower nature. Because of the shower nature, really?

It's like I don't speak for a living. shower nature? Yes, because of the shower effect, it is likely that two or three or four may go off at the same time. And in fact, I have a project from 2011 that's called Radiation Orchestra, where multiple sounds play at the same time. And so what would I, what would I hear? Okay, I would get a meow and a purr and light up on the cat's planet. what other planets do you have do they have names no not yet but i imagine that they will can i help with that yes absolutely the the way i had originally conceptualized this for the stories would be that the planets would be in clusters and so we would have a biological cluster an emotional

cluster, a scientific cluster, a mechanical cluster, and a health care cluster. And so in those clusters, because the original idea for the piece is that there will be 37 of these sculptures, each one holding a cosmic ray detector. And you will walk through a path and experience the the music and the stories that these cosmic rays trigger. How do you make music that is approachable, appealing, and random? I'm working with two sonic environment developers whose names are Ria Rajan and Sophie Yuditskaya. They are making this whole thing sound palatable. And I think we've landed on making it sound like a sound bath.

Okay, okay, like ambient sounds. And then, I mean, I started thinking about games where you have a soundtrack, but then as you do things, it needs to be pleasant inside the soundtrack. And so that's about right. Okay. Yeah. Okay. Yeah.

I would say that's a really good analogy. I put one of those game sounds, those ching-ching sounds, in one of the planets. And after listening to it for a couple of days at Chabot, I was like, that sound has to go. I can, there are a lot of things that are fun for about five minutes. And then after an hour, you're just like, please know. That is the trick of these generated procedural thing. I mean, it's random, but it's also somewhat. procedural. And the trick is how do you make something that doesn't just wear a groove into your head?

Right. And also is meaningful. Yeah. That's why I have the whole next year to work this out. For me, the cool part is the randomness and the making something invisible become experiential. I guess experiencing the invisible, experiencing the ineffable. Sorry. As an engineer, I find that sort of making things approachable. Okay, I get it from an engineering perspective. Tell me about the art.

So why is this art as opposed to engineering? Well, I mean, as, as, if it was just engineering, this would just be a sensor, and it would just give me ones and zeros, or maybe it would be the Geiger counter. Would it output a JSON file you looked at every afternoon? It would autograph, thank you. Yes. All right.

So, what's the question? Yes, I agree. The question is what makes it art? Oh, I mean, isn't that always the question? What is art even? What is art even? Yes. I think it's art because this installation messages from space is more than a sensor.

As you said, if it were a sensor, it would just be a Geiger counter or a JSON file or an auto-generating graph. This is giving you the randomness as though it's telling us something about our future as humanity. And so I think this inquiry-based installation is the place where it crosses over into art. I can see that. Because when I think about if we generated cosmic rays as humans, which I understand we don't, unless something's very wrong, our cosmic rays could travel through the galaxies, through the galaxies. and in a hundred million years might hit somebody else's art project. Yes.

The voices from the cataclysms of the universe is just kind of magical. You don't know that that's always around you, that it's always here. And part of me wants to be able to touch it, but if I can't touch it, then hearing and seeing it is a reasonable second. place. I'm going to ask a slightly more specific, is it art question? So thinking about a cloud chamber, I'm thinking about my experience, looking at just a cloud chamber, which is a completely anodyne, not art, not intended to be art. It's a box that you look through.

I think the museum curators would have some issue with this, but go ahead. Let's just stipulate it was originally designed as an instrument. Sounds beautiful, though. Fine. It is beautiful. I'm not disputing it. It is very beautiful. I'm headed somewhere with this.

No, sorry. The particles coming in. There's a sense of wonder there. There's, you know, a sense of being connected to things in the universe. And so that sterile artifact produces that response. What is your goal with putting intentional art on top of that kind of detector? What do you want to mediate in the experience beyond just the, I'm looking at a science thing and having a sense of wonder? I'm asking people who experience this installation.

I want them to imagine future experiences and to imagine connection and communication and also optimism and hope because that's the futures that I'm presenting in this art piece. Okay. How does the optimism and hope come into it? well we could definitely read a gazillion science fiction books and listen to a lot of dystopia that's all around us i i tend to ignore all of that stuff so all of the fiction that i'm writing is about is is optimistic and also just realistic like it's not so hard to imagine a future let's say 10 years from now. And then if you went on top of that and said 20 years, 30 years, or even 100 years,

which is a pretty big jump because 100 years ago, what we weren't here, it was wildly different and we would not have been able to imagine what was going to come. But I think there are some things that we can imagine. Like some of the stories that I've written are about getting a heart transplant. How does that work? How does a hip transplant work? how do our knees work what does it feel like when you have like not just a eye surgery i've forgotten what that's called cataract not cataract the other lasic what happens when lasic actually lasts forever

what happens when it's genetic modification and you can just be born with 2020 vision or a third eye like some of these things i think are they're optimistic and they're not that far out of what could be possibility. So these stories are not science fiction in the sense of like a totally different world. It is the same world, but what I imagine things are going to be like in 100 years, if we are still here. Where do the stories come in? We've talked about sounds and lights, but they are, so if there are, well, there will be 37 of the sculptures in your walking down a path, and let's say every fifth or sixth one, or each cluster will have one that

tells stories. Okay. Yeah. So, oh, and then a question I get asked a lot about the sounds and the stories, and I think eventually you'll land on this, so I'll just answer it. Why is it not cacophony since I did say that they go off all the time? Oh, yeah. And the answer to that is I've tuned them so that they go off every tenth.

time. And so, oh, we get some meows and some purrs and then a cat saying? It's not a cat. The way they're written right now, there's only about six or seven stories fully written and edited. And actually, I'll put a link to that. I'll give you a link to that for the show notes so people can listen to the stories. They're dialogues that, yeah, they're between 60 seconds and two minutes.

And do they need to go in order? No. Okay. But if you walk into the biological cluster, you're only going to hear stories about how we deal with things like our food, how do we eat, how do we see in the future. If you walk into the mechanical cluster, we might find conversations about transportation, which I picture, which I picture to be. like that podcast, Car Talk. Okay. Okay, so as a Cosmic Ray comes and hits the designated storyteller in each group,

it talks, it plays one of the audio snippets that is a dialogue being held, and it doesn't have to be in order. Is that right? That's right. Okay. I'm getting it. I'm getting it. Be easier if I came to see it. When can I come and see it? Sometime in 20, 26. You said that. That's like 30 years from now. I know. I'm hoping to be able to present this project in its full form sometime about halfway through the year.

Do you have a site? Do you have a place that's going to go? I don't have anything confirmed. Cool. But if you are a listener who would like to host this piece, please be in touch. Yes, you can contact the show and we will forward your information to Sophie, or you can probably reach Sophie from her website, which will be in the show notes. Okay, so that's art. And clearly there's electrical and software in there. But you've focused on that for the last year, and now you just need to, you know, manufacture a couple dozen more. Trivial problem, not a trivial problem. I mean, it's kind of trivial at this point to manufacture boards once you've done the first one.

Well, you also have to do the sculpture and make, it sounds like you have the story, so you've made a lot of the decisions. Yes. Yes, it probably isn't as hard as conceptualizing at all, but it's also not going to be. I'm just going to crank out 30 more systems. Oh, that's true. Yeah, yeah, that's true, yes. And there is a production aspect to it.

It's never easy to make 30 of anything, except maybe cookies. Yes. I'm also a great baker, if this falls through. I've heard your birthday cakes are really good. But you've spent the last year developing a lot of this concept and other artworks. But you also said that next year you're going to finish this and be working as an electrical engineering consultant. Yeah, it was pretty magical to take the year to focus on this project. And I worked on some other projects too.

I also got very, I got pretty fluent in 3D modeling. Which I had long thought with for many years. How did you get better at it? I start and then I just, I get so frustrated. Which software? I mean, I have started. Autocat is the one I always go back to, even though I kind of hate it. But there's a free online version.

Yeah. But then I like make a box and then I'm like, okay, I want the box to do this. and I'm just look at it and think, okay, it would be nice if it did that. And the truth is, what I need to do is sketch it out on paper first because I just don't have a good mental model, let alone what the actual tools are. But if I could get past the, I don't know mechanical engineering hardly at all, what would be the tips and tools to get me further with modeling things? I'm coming back to your idea of writing, drawing it on paper.

My first job in engineering was with civil engineers. And my job as a drafter was to view different things from photos in different views of the same thing. And I just could not figure out how to get a 3D thing onto a flat piece of paper. It was terrible. So I cut pieces of paper up and folded them to make these 3D models to make that connection in my brain. And it took me a long time, but I got there. And so coming back to the first question you asked me was about a tip. And I said having a chat GPT or clawed window up while you have a software is a superpower.

And I had fought with Blender for years. Like, I think we're up to Blender 4.5 now. And I started trying to learn it at maybe, I don't know, 2.5 and gave up every time. It was just, I feel your frustration, Alicia. You are seen. But this year, just having the AI window up and fighting it. And I really, I worked in Blender almost, I would say almost every day. this year and became proficient with the help of a chatbot.

You're telling me I have to put a lot of time into it because I wasn't signing up for that. The thing that that helps me get better at Kat, and I'm not good at it, but I can make simple boxes and mechanisms and things. The thing that helped me was that I had to do it for a client and I had to make this work. And so therefore I sat down and I didn't use a chatbot, but I had a lot of YouTube videos and stuff open and it was very specific okay i've gotten this far and now i want to do this how do i do this and then i would learn to do that little thing um and then but there was there was i think in blender too there's there's some fundamental principles like okay you start with 2d shapes and you extrude them and then you subtract things from them by intersecting with other things and once you

kind of get that idea once i got over that it i felt like all the cad stuff was like okay these are all variations on the same theme. You're extruding things, you're taking cuts out of them, and that kind of thing. It gets much more complicated, of course, when you're making something real. But the basics I didn't understand. The first few times I tried, it was like, what is all of this? I don't understand. And there's a new basic. Oh, like you, Chris, I had a client a few years ago who requested that I use Inventor, which is an engineering CAD software for mechanical engineering. I think what happened was their mechanical engineering intern left, and I was there. So suddenly, I mean, the intern can do it.

Right. They were like, oh, we have a warm body here who can also do this. There was no AI back then. But I also, like you, Chris, I had to fight through it just to finish the project. But to add to the extrusion and 2D stuff, there's another thing that I do. on the art side, which is to sculpt the thing by hand in clay and scan it. Oh, cool. And get in a, what's like a mesh model.

So it's a surface model. And then do you manipulate that further digitally? Yeah. As I get better at digital sculpting, I'm feeling my analog sculpting mind, I don't know, surrender or something. I mean, as you were talking, Chris, I was thinking, well, I mean, that'll make sense because that's what I would do in clay. So Sophie saying do it in the clay, I'm there for that. And there are applications. I think there are pure sculpting applications, too, where you start out with a sphere or a cube and you go in and you rotate the thing around. And I found those difficult because it's one of those 2D to 3D mapping things in your head as you're spinning it around and trying to do stuff. But yeah, you can do that. Yeah, you can do that with Blender. You can pull in, like, spheres or cones.

With a brush, right? So you're actually kind of using a brush. Mm-hmm. And clay tools, which are surprisingly, they surprisingly map to the analog sculpting tools that I learned with. Hmm. What software is that?

Blender. Blender. So I use... Blender's big and scary. Blender is much better. than it used to be from the 1.0 days or so, but it's, it's, it does everything now. So it is a little big and scary. It'll do video editing too, so. That's right. The thing you might like, Alicia, is that it's all Python based under the hood. So you actually can write snippets of code to do

what you want and chat GPT taught me how to do that, which was pretty amazing. I could take my origami, which is Python generated. Put it into blender, fold it in clay, and then 3D print it. That's right. And actually, that sounds like an amazing project. There's some weird new origami. I don't know how you get the folding step. That's kind of step two of the question mark.

Well, there's the origami simulator, so I'm just assuming that I can mash those together. I find the, as someone who really is not that familiar with coding or with Python, I found getting Blender to do stuff in Python with the help of AI was pretty straightforward. One of my problems with using AI for tools is that they, the AI always seems to be using a random collection of versions. And so it says, go here and do this. and I go there and it's not there. That's right.

Yeah, that is super frustrating. I often very sternly tell my AI to, you know, I'm using this version. When I say my AI, I primarily use Claude now, but I started out using chat GPT the most. I don't use Gemini just because, like, why need, who needs three AI's telling you what to do? There's no reason to have really, I mean, once you can. get really comfortable with one. You're right. I tend, and I, you know, I tend to reach for Gemini because it's the one that first worked

for me. And I don't, yeah, I should try the others, but yes. Okay, but so learning mechanical, but you have been an electrical engineer for a long time. Yeah. And are you worried about having taken a year to focus? See, but your art is still very engineering. have this question about you taking a break and going back to tech, but your break was full of

tech. Yes, it was. This isn't the first time you've had a break from doing deep technical work. No, it's not. I worked at Hackaday on the business side for almost seven years, and I didn't work in a technical role there. But prior to going to Hackaday, I had a lot of very technical roles. primarily worked as an engineer with scientists. So the expectations were pretty high that both I would be technical and also be willing to learn all kinds of minutia about the specific science. I was worried. I mean, and I think, you know, I think a lot of people worry. You get out of your technical role, you go into management, you go into business. These are kind of like normal pathways for people who have been in engineering for a long time.

and you worry about both getting back into tech and then also, what does it mean when I become an individual contributor or I am a one-person company or even a three-person company? What does it mean to not be in charge of a big department or a manager or in business? But I just, I really wanted to be back in engineering. For me, it's very creative. I love it. I never stopped loving it. I honestly just missed it a lot.

And it's been very fun to, I feel very grateful that I was able to get back into it. And I still love it as much as I ever did. Were you worried? I mean, I guess, I don't know why I have this fear that if I put down technical work and do something creative, I will never be able to go back to technical work. But I also know I'm not alone with that fear.

Yeah, I was very worried, and I wasn't sure that I wasn't sure that I would be able to get a technical role because I hadn't done anything technical in a pretty long time. However, if you join a startup or a small company, they don't care. You just, you know, it's like if you're not, if you can't do the work, you're not going to be staying. And if you are doing the work, they're not necessarily noticing. if you're, like, I was spending weekends studying, and I still spent a lot of extra time studying or learning what the actual task is. And maybe that's just my nature and because I'm into it, but part of it is also, or at least in 2021 when I came back to engineering, part of that was just a lack of experience. But like anything, if you want it, you're going to fight for it.

I guess this is one of the burnout things is that at some point I feel tired of technical work. And then I think I have to continue because there's no way I would ever pick this up again. And the truth is if I would actually give myself some space to heal a little from the burnout, picking it up again would be fun. I think so. And I don't think that you ever lose your ability to reason and problem solve. And even things like your specific technical skills, whatever those might be for anybody, mechanical engineering or coding in Python or electrical engineering, the laws of physics don't change. The tools change. But as a technical person or as an engineer, you're the, the expectation and the necessity is that you're going to keep up with whatever the tool of the day is, especially if you're working on a team and everyone is using that tool. And I went back to college for engineering at age 36. And I think if I had not been in school with people who

were 19 at that time, I would have missed the whole idea that your value as an engineer is being able to hop on and off different tools. Yeah, I think that's true. It's a paradox, right? Because stuff does change, but the stuff that changes tends to actually be, I don't know, minutia. It's like there's not a lot of,

oh, we're fundamentally changing how computers work this month. Right. And you missed it. Sorry. A lot of people will say that the AI is that, but I don't think it is. But, but, and that's a, same true of electrical engineering, probably more so for electrical engineering. It's not, you know,

physics breakthrough, turns out wires don't exist anymore. You have to learn to use, you know, pretzel straws or something. That doesn't happen a lot. And what we learned as engineers and computer programmers and whatever is kind of how to learn those things. That's right. Yeah. And I lose sight of that because I'm currently, like I said, coming back to doing more technical work. I'd done some client work this year, but it was mostly Python, and it was mostly kind of glue code, you know, piecing things together. It wasn't a lot of code doing logic and things like that. But now I'm moving back to embedded stuff and back with C and C++ and microcontrollers. And yeah, I was like, I don't know.

I don't remember how to do any of this stuff. But your biggest problem so far have been tools, which... The biggest problem has been tools or... Would have been a problem a year ago. Learning a giant library that I don't know and what's wrong with it and why doesn't it work on this microcontroller, which, No, I didn't. Maybe I lost a step here or there briefly like, oh, I can't remember, you know, my debugging workflow or something or what is a common problem here? But that's stuff you pick up and I just, but like the fear of it is real because when you jump back in, it is a little slower, right? You are a little both really because you've maybe forgotten a few things that you'll pick up quickly, but also because you're, you know, hesitant and and trepidacious, at least I am. There's nothing like anxiety to make everything go a little slower, much slower. Right.

That's definitely true. But I think also the people that you're working with, my experience mostly in technical roles has been the people that I'm working with are also perhaps they're struggling with the same thing or they very much understand that it takes a certain amount of time to get familiar with something. Yeah. And so there is no shame in being like, I'm working hard and I'm going to understand this in X amount of time, if anyone's even asking. I find that usually everyone's too busy to even ask. Yeah, everybody thinks that everyone's looking at them. When in truth, everyone else just wants you to finish what you're doing so that they don't have to ask about it. I think that's true.

I think it's an additional quirk of contract work sometimes where, you're not, at least I feel guilty about, oh, man, I thought this would take five hours and it's going to take me 30, which is, you know, a lot more money for the client or whatever. And so there's that aspect of it too, which I didn't feel that so much doing full-time work because I'm salaried and everybody's working together and, you know, whatever. But when it's a small client and it's like, I'm billing you hourly, I want to be efficient. Well, you can also remember that starting in 2026, that's a small client and it's like, I'm billing you hourly. small client is going to be paying their full-time people a lot more for health insurance.

But I'm paying me a lot more for health insurance. That's true. That's why you need to work 30 hours instead of five. And that's why our rates are going up. Don't bug us about the higher rates. Please see all of the health insurance and just how much it costs. But you are going back to work in an engineering capacity. Do you have clients lined up or are you looking for some? I'm looking for something. What kind of projects do you want to do?

So what I've done in the past is work for early stage startups, often wearing a lot of hats, design engineering or testing or marketing even or writing content. So in the 2026, I'm adding a new service, which is helping hardware companies validate their products are ready to launch. So not just testing the hardware itself, but also the complete ecosystem, like, does your documentation work, does your app work, do the accessories connect? Can, if you have an embedded platform that you're putting out into the world and you have, you spent a lot of money on marketing and your big launch, can the people who have it? Can they connect successfully? That kind of thing. So like an external reality check before something gets missed at launch.

Okay. Kind of a QA department. Or not that? Kind of. Kind of a QA. Adversarial QA department. Yeah. Black Hat QA. If we can't break it, no one can. Red team. Right. Yeah. And since security testing. Sorry. No, I realize, like, companies do have, they, I think companies, of course, do QA testing before they ever launch. But a lot of times companies are insular and they're testing it within themselves. And when it goes to launch, I'm thinking of several pieces of hardware that I have not, I'm just not going to name anything. But, you know, listeners, I'm sure, think of your favorite hardware that you were not able to get working and you abandoned on the,

first day. That kind of thing. And I think there's a psychological thing, right? Like, we as software developers shouldn't be testing our own code, not because we're consciously avoiding breaking it, but because we are unconsciously avoiding breaking it, I guess, for a lack of, but we don't know, we don't try the things that are necessarily a little bit off the wall or obvious to new users. Obviously to new users. That don't have our experience and know how that you're supposed to push this button before that one? There's a thing I learned in a mixing class recently where the instructor said,

once you feel like you're done with the song mix, invite somebody else into the room to listen, or just to be there while you listen to it again. But they're not allowed to say, they don't have to say anything, they're not supposed to comment on it. They're just supposed to be there and then leave. And he says that psychologically,

once that person is there, you notice a bunch of flaws that you wouldn't have noticed because somebody else is just there. And I think it's a similar thing. It's like this outside voice existing that doesn't have a connection to your company or an emotional connection really or not thinking the same way. I think that's a huge psychological thing for groups and people as well. So it's a great idea.

I mean, it's kind of rubber-ducky-ish. Yeah. Yeah. Yeah. And I think from the engineering perspective, I will be able to offer something probably a little bit deeper. Disco Duck QA Solutions. I'm looking for a company name as well.

Oh, I do like that, Disco Duck. Probably, I think it's probably a Disney trailer. Cosmic Duck. Yes. Sorry. It actually kind of, I'm going to pull it all the way back to the beginning where we tried to get onto Zoom. And you said, this happens every time.

Imagine if Zoom started out with a button that said, connect to this audio. Not made you click on some little icon that you don't necessarily. Mine was not on the screen, actually. I had to hover over it to see it. Well, actually. Well, actually. There's an option to do that. Is there?

Yeah, it comes up with a big thing. It says, connect to your audio and it shows you're, I don't know where it is, but on some of my computers it does that, so I must have it set in different places. When I connect to Zoom the first time, for a meeting, usually I get a pop-up that lets me choose all of that, but I don't know where the setting is. I don't either. And since I do shift my audio quite a lot, it's important for that pop-up.

And they could even just have Zoom podcast mode. Yeah, that would be a good idea. And then it would record to her without having trouble. Yeah, the recording stuff for them. Anyway. Anyway. Yes, making a common. Common thing's common.

Yeah. But, I mean, I'll also be offering hardware design and testing in the normal way as well. Cool. So if anybody's looking for hardware consultant, Sophie's website will be in the show notes. Okay. I want to go back to one more question about the art.

You did another installation recently called Transmissions from the Long. Yes. What was that? And who did you work with? That was a precursor to messages from space. And I worked with Justin Day to realize the software. He did the software for the LED patterns.

And with my partner and husband, Oliver Tanner, on some of the mechanicals and installations. That was a piece that was at two festivals over the summer. I usually make a piece that just goes to festivals locally. And this one was about getting the participants to feel an emotion for somebody that they felt like had slipped through their fingers or lost. And there was a story. I wrote a story that you walked through a path reading the story, not listening to the story. and I think we can link to it on the show notes if people want to read the story. And at the end of the story, you were asked to put your hand on a sensor,

which triggered a blinking pattern and asked the participant to think about somebody that they lost. And so this story was the more deeper thought behind it. It was about the deportations. So I had the project on my website in multiple languages, but in the insulation, just because the piece of paper only has two sides, it was Spanish on the back. My next question in my list here is, where do you get your ideas? But I feel like that one's an idea that really needed to happen.

So let me actually ask the opposite question, which is I swim in an ocean of ideas. How do you decide which of your dozen projects is the one you're going to finish? I mean, you don't start a dozen, but how do you decide what's worth pursuing? And how do you turn away the ones that are good ideas, but you just can't do everything? I have a list like a mile long of project ideas. Yes. I mean, that's excellent to just have a Google Doc to put all the ideas into.

so that I feel like there's never a project that got away. But I'm pretty analytical about how I'll decide if something is going to be, if I'm going to actually pursue something. You know, first thing with an idea is, is it doable? Like, do I have the skills to do it? Do I know people who have the skills to do it to work with me? Who will want to work with me on this? So that's like a first gate.

Like, there's definitely projects, large-scale outdoor projects that I want to. to build that I don't have the resources for or the space to build. So those things kind of sit on the sidelines. Requires football field. Let's put this one to the side. Exactly. Yes. Or is like public art, I would love to pursue large-scale public art projects. And I'm sure I will at some point. But yeah, is the project doable? I think is a pretty good gate. And then how much funding do I need. I can do a pretty quick budget. Even if it's wrong, I can get within, you know, I can get within a window of money. And then from there, be like, our grants available is, am I going to get a rich collaborator, et cetera?

I see the grant sometimes. And this is all about me, not about you. I see the grants. And then I think that I should leave them for people. who can't do engineering work and make enough money to live on that and still do the art. I don't know. It's not really guilt. It's more, I don't know.

Where was it going with this? I think that that's an altruistic way to think about grants and you're right on one, or I agree with you in one sense. And in another sense, if you're pursuing a creative career as a visual artist, it's validation and it helps your career to get those kinds of, those kinds of opportunities. And if you, you, Alicia, were to receive a grant and felt like you didn't want to keep it for

yourself the money, you might then share it with, you know, collaborators or with other people. but I could still say grant winning. Yes. Which as we know from the rest of this podcast over the year, I need awards. That's right. Yes, I would also say that, I mean, just using the podcast as an example, whatever you've spent on that, be it time, equipment, all of that kind of stuff, the grants are often pretty small.

So you might not even break even. So I'm just saying like you might. This is so true. Right. Sponsoring is wonderful. And yet it still doesn't really pay us for our time. Exactly. It can't.

So even though you make a living, it's not wrong to want to get reimbursed for the time and supplies that you put out on a project. That makes sense. And I think I just need to hear it a few hundred more times before I managed to internalize it. But maybe Chris will just make a loop if you say in that. And every time I walk by somewhere in the house, it will, whenever there's a cosmic ray. Sophie, do you have any thoughts you'd like to leave us with? I think I talked a lot. I'm good.

Well, then I'm going to ask you one more question. When did you claim the identity of artist and how? Oh, I probably claimed that identity when I was a teenager. I mean, I first went to a couple of different art schools that I didn't graduate from. So, yeah, I think I was an artist from teenagerhood. All right. Our guest has been Sophie Kravitz, artist, electrical engineer, optimist. Thank you, Sophie. Thank you for having me. That was really fun.

Sophie, it's always wonderful to talk to you. Thank you to Christopher for producing and co-hosting. Thank you to our coffee and Patreon supporters. We really appreciate it. If you would like to get an ad-free stream or the occasional bonus episodes, join our coffee and Patreon group for $5 a month. But everyone who contributes gets access to the embedded listener slack. which is a pretty fun place to hang out. And now a quote to leave you with from Helen Keller. Optimism is the faith that leads to achievement.

Nothing can be done without hope and confidence.