The Benefits of Asynchronous Design Review for Engineers & Manufacturers w/ AllSpice.io

This week we are joined by Valentina Ratner and Kyle Dumont, co-founders of AllSpice.io. Valentina and Kyle met while doing an MBA program at Harvard and realized they both shared a common frustration - the lack of asynchronous design review tools available for electrical engineers despite such tools being readily available for other engineering fields like mechanical and software.  Thus they joined forces to create AllSpice.

AllSpice is a git platform for hardware engineers that helps them to accelerate their time to market by enabling modern revision control and collaboration for native electronic designs.

We had a great time having them on the show to learn more about AllSpice as well as Valentina's hatred of sharp-edged rectangles

Show links:

• AllSpice website: allspice.io
• The Git for Hardware Guide: https://content.allspice.io/en-us/git-for-hardware-guide
• Questions / Demo: info@allspice.io

pickplacepodcast.com

Transcript

0:08

​Chris: welcome to the Pick we talk about electronics, manufacturing and everything related to getting a circuit board into the world. This is Chris Denney with Worthington.

Melissa: 0:16

And this is Melissa Hough with CircuitHub.

Chris: 0:19

Welcome back, Melissa.

Melissa: 0:20

Welcome back. Chris,

Chris: 0:22

Hey, I'm, I'm in a much better mood this week. We're

Melissa: 0:25

you were telling me Yeah. You, things have been a little, a little bit less hectic, right?

Chris: 0:30

Less well. Less stressful, just as hectic, less stressful because yeah, things are just popping off and we're just firing on all cylinders and, the learning curve is less steep. I remember, I remember hearing about this woman who moved to the United States from the Middle East, and she really struggled with pronouncing words properly and being understood. And she had such anxiety just going to like Starbucks to order a cup of coffee. And it made her so nervous the first few months that she was here. But then after she lived here for a couple years, she got the accent better. People couldn't understand her better, and she was able to navigate the world so much better. That's sort of how I'm feeling at the moment about introducing this new equipment. It's like, okay, I finally know how to speak the language a little bit. It's like, I'm, I'm not so nervous that I'm not gonna be able to do this. I, I think we'll

Melissa: 1:17

Yeah. You understand each other.

Chris: 1:19

Yeah, exactly. So it's it's going very well and those guys have been great supporting us really well, and, and our teams just, they're just killing it. So it's good. It's all, it's all moving in the right direction.

Melissa: 1:30

Love to hear it.

Chris: 1:32

Yeah. How about yourself? Everything. Everything going good on the circuit? Upside?

Melissa: 1:36

Yeah. Everything going good on the CircuitHub side? Very busy. Especially before the holidays. Yeah. Oh gosh. And before we know it, Chinese New Year is gonna be here too, so Yeah. It's gonna be a crazy season very

Chris: 1:48

Yeah. Well, yeah. We should have a little refresher on that too., speaking of things being good, we, we actually I think we can talk about it. Why, why shouldn't we be able to talk about it? We can talk about it. We had to shut off all orders for, for December. Basically. It was like, it got to the point where we just, we have so much to do. We can't keep up and just shut off all our orders for the month of December, which is great, but awful at the same time. Cuz that means you.

Melissa: 2:11

some. Yeah.

Chris: 2:12

in December, they gotta go somewhere else. Yeah. Which kind of sucks, but we're working our butts off to increase capacity and we're hiring, we're building a second shift and, and that's going really well. We got some star players on second shift, which is really exciting. So, all good things. Life is good and I don't know what this recession thing is that people are talking about, but I haven't seen it yet, so let's hope it stays that way.

Melissa: 2:34

Yeah, Sure.

Chris: 2:38

we're recession proof, right, Melissa? Of course we are. Everybody's recession proof. Yeah, well, I'll tell you what two people who certainly are recession proof are our guest today. We have the co-founders of All spice.io Valentina Rattner, and Kyle Dumont. Valentina is an engineer and former Amazon pm I assume that stands for product manager cuz I'm reading this off of the marketing details about her She's managed infrastructure projects and internal productivity software. She holds a BS in mechanical engineering from Boston university and an MS in engineering and an MBA from Harvard. Passionate about hardware, space and cross team collaboration. She co-founded All Spice in 2019 alongside Kyle Dumont, who is also a very skilled engineer. Has 10 years of experience launching mass production Consumer products. Has a BS in electrical engineering from Northeastern and an MS in engineering and an MBA from Harvard. I feel like I'm, I'm like. In an echo chamber That sounds very familiar. like I've said that before. And recently Kyle spoke at Git merge here in 2022 on the title How Git is transforming the Future of Hardware Development. So, we'll be sure to include a link in our show notes to that video. It was a, it was a nice video. I didn't get through the whole thing, but it was very nice. So welcome to the show, Kyle and Valentina.

Melissa: 3:52

Welcome guys.

Valentina: 3:54

Thanks. Thanks for having.

Kyle: 3:55

Glad to be here.

Chris: 3:56

Yeah, very, very excited to have you. I was excited you guys reached out to us and we, we actually get a fair bit of people saying, Hey, you know, I have somebody interesting you should talk to. And we look at it and we're like,

Melissa: 4:08

Hmm.

Chris: 4:08

totally unrelated to what we do. And then, and then I saw it from you guys and I was like, oh yeah, no, no, for sure. We gotta get these guys on the show . So, super happy you guys reached out and I think it's gonna be great. I'm looking forward to chatting about all things Git, which I am. I'm a total complete novice on, on, on this whole subject, . So I'm excited to chat about it. Before we get into the product itself and what you guys are working on and how you got to that point. Maybe Valentina, if you don't mind sharing a little bit of your background and, and kind of like, where you're from and how you got started and, and all that kind of good stuff, and how you ended up working with Kyle on Allspice.

Valentina: 4:48

Yeah, absolutely. Super excited to be here. A little bit on my background. So, I'm originally from Argentina. Grew up there in a small town, northwest part of the country. Very warm year round by the mountains. And I moved to the

Chris: 5:03

So just like Boston

Valentina: 5:05

just like Boston. Exactly. I had never experienced snow before. If you compare it to the US it will be from, it will be a similar maybe landscape and weather somewhere between West Texas and Arizona.

Chris: 5:19

Oh wow.

Valentina: 5:19

that was kind of like my, my upbringing and I moved to Boston for college. Learned about snow winter, winter proof gear, good boots and good, good jackets. And I studied mechanical engineering, actually concentrated in manufacturing. Worked at Automation lab in school. I was one of those people that everyone had their, their hobbies and passions. I was really into conveyor belt lines. That was like my thing. Yeah. Loved it. We had like a set up in school and like a pick and place and in LA then a cnc and I, I just like, loved that lab. So, graduated college,

Chris: 5:53

which school? Which school had had that lab?

Valentina: 5:55

Boston University.

Chris: 5:56

Okay. Yeah.

Valentina: 5:57

The, the lab was called Epic, very appropriately engineering product innovation center.

Kyle: 6:01

I was gonna say that has to be an acronym.

Valentina: 6:03

It has to be, , it was perfectly placed for the lab. So it was brand new. So we, we got to play with a lot of cool new equipment and kinda an, a miniaturized version of what, like, state of the art manufacturing technology looks like. So really, really enjoy that. Those last few years of school and I worked at Amazon after graduation and what I usually explain as the backend physical infrastructure that supports the online shopping experience. So if you think of all of the physical parts and convey rail lines and machinery and trucks and logistics are kind of like the magic that happens between you click purchase and you get your package at home.

Chris: 6:44

so you're, say you're talking about the physical logistics, not just, cuz when I think of the ordering process, I think of the, the you know, the servers and everything that run it, but you're talking about what happens after I click order.

Valentina: 6:54

right, right, right. Like the actual physical moving of goods and kinda getting, getting packages from, from the original warehouse to, to customer homes. So,

Chris: 7:03

Which is really, in my opinion, that is the miracle of Amazon. Like that really is what makes Amazon, Amazon, you know, anybody can throw up a website and accept orders. It's how do you, how do you deliver

Valentina: 7:14

Yeah, and it's just, it was a, a very exciting time because well Amazon had been growing a lot and we were opening capacity in all of the states I got to basically open a two day shipping delivery in a lot of new cities and towns. And and one of my useless talents is being able to read Amazon labels and tell you which states and, and buildings your package went through before I got to your home.

Chris: 7:35

that, that is a cool, like you're going over, like your friend is like, Hey, come over to my other friend's house and you see an Amazon package sitting on the table. That's like a party trick you got right there,

Valentina: 7:45

Yeah. And, and depending on where they're at, if to say that I worked at, I'm like, this went through my building. You're welcome. Now two day shipping. Really exciting time. Lots of lots of work. But I also worked in I was in specifically in like the last mile part of the business, which was relatively new. So a lot of things kind of building from scratch. Kinda developing this infrastructure. And I was working on multiple projects in multiple states and needed to be in Alabama at once, which was basically impossible. And that's when I got kind into more of like the internal software productivity. Like how do we scale this? How do we manage these projects in several places with engineers? And through that, I, there was kinda a two part process that happened. The first one was seeing Man, there is, there's not a lot for us, for hardware engineers and everything was posting notes and a lot of email threads and pictures of whiteboards and things like that. Yeah, it was pretty pretty stunning that that was kind of the, the, the status quo. So first one was like, not, not allowed to, to start with. The second one was we started when working with a team for our own software tools. And once we rolled this out in the different projects, it was just incredible, like the impact that it had both on the quality of the project, the speed, the number of defects was going down, the number of problems was going down all the way to the employee satisfaction. People were happier. It was easier to talk to each other. They collaborated more so it just had such a widespread. Impact that I got very, very passionate about kind of working at this intersection of software tools for hardware engineers. And watching the software team collaborating, you can see all of the infrastructure that has developed and it's a typical match of, as a software engineer, you have the skills to be build your own software tools. So the ecosystem has moved a ton as a hardware engineer. Not everyone has the software skills to go build their own software tools. So it's a, it's a pretty, pretty big gap. And so I, I wanted to work there and I went back to school, to grad school and that's where I met Kyle. And the, what you said, look, the eco chamber at Harvard? We did, we were actually the first class of this dual deer program which is a combined MS and mba, so an MS in engineering and an mba. And guests call it, call it Destiny, call it fate. We actually were randomly assigned to be partners on our first project together in our first class, and that's how we met.

Chris: 10:11

No kidding. That's too funny.

Kyle: 10:12

It was perfect cuz we, we definitely you know, commiserated over some of those issues around hardware engineering and in collaborating on hardware designs. My background's in electrical engineering product development. I did EE development as, as you said, Chris, for, for about 10 years before going back to grad school. I spent roughly like the first half of my career working for iRobot and Consumer Electronics. Primarily, although I actually started in, in industrial robotics at, at that point they had some e od systems, explicit ordinance or robotics that they were designing for government applications. So I got the really cool opportunity to work on some of those before switching gears into the, the Roomba which, which folks are probably a lot more, more familiar with very different world of course.

Chris: 10:53

There's a very, there's a very funny inside joke around here about uh, about our Roomba. We, we , we have, because it's an e s D environment, we have like these chains that hang off of all of our, our carts and everything.

Kyle: 11:07

So have a little Van Degraf generator on wheels that are

Chris: 11:10

always gets hung up on all the chains and we find her the next morning just upside down, flipped herself over She's like, ah, I don't Roomba gonna on the factory floor

Kyle: 11:20

I thought this was gonna be an e s d horror story.

Chris: 11:23

No, no, no, no, no,

Kyle: 11:25

that was, that was certainly a bit of my

Chris: 11:26

had one yet. Yeah.

Kyle: 11:28

Is, is one might imagine having a little brush with electronics on it that just goes around a carpet sweeping things up. Especially in the, the dry winter in Boston. They, they pick up some, some some serious electrons and and under

Chris: 11:44

gotta pick it up and you'd get quite a shock.

Kyle: 11:47

absolutely. You know? Yeah. So we, we always had to test against that. That was that was always a, a big journey in the development cycle was, was ESD testing. And we had tested at 20 kv, 30 kv. We, we actually, I, I wasn't super involved in this, but there were folks at the company that were developing I don't know if you're familiar with like human body models for like E S D E s d specifications, but we were actually developing robot body models. So like what does the what does the, the E s D profile look like for a robot as it picks up a bunch of electrons and then, and then lets them go

Chris: 12:22

The human body model. Familiar with it as far as I know that my suppliers are familiar with it. You know what I mean? Like they have to worry about that, and I just have to buy it knowing that they went through all that. Yeah.

Kyle: 12:34

Yep.

Chris: 12:35

At iRobot? We have a friend of the show. I've been, I've been meaning to reach out to him to get him to record with us. Last name? Miller.

Kyle: 12:43

Scott.

Chris: 12:44

Scott Miller. Thank you. Wow. Did you, did you work with Scott?

Kyle: 12:48

I, I overlap with Scott A. Little bit, although I will say at the time I started, I think Scott was something about VP of hardware. I was engineering intern. So suffice to say we didn't have a lot of a lot of overlap, but since since leaving iRobot, Scott's actually was one of our first advisors of

Chris: 13:06

Yeah.

Kyle: 13:07

he's very familiar with what we're doing. We

Chris: 13:09

Basically, if you're a startup in Boston related to hardware, you know Scott Miller at this point.

Kyle: 13:13

Exactly, exactly. And he's just been a great person. I, if you've ever had him on the show, that's, that's just prime listening.

Chris: 13:21

We haven't, but we want to. I mean, we, he's been to our factory a bunch of times and we've worked together on a lot of things, but I, I wanna get him on the show. I think he'd be a fun interview. Yeah.

Kyle: 13:30

Oh, great. Yeah, absolutely. So yeah, that was, that was a really good learning opportunity for me early in my career. And then I went and joined a 3D printing startup company. Much different experience because this was really, you know, green field opportunity at least as I saw it coming in to, to grow an EE design process and a design team from, from the ground up. So really, really neat process there. Of course, then ultimately decided to go back to join the MS N B A program. Valentina was

Chris: 13:57

I'm, I'm gonna, I'm gonna ask you about your experience with that 3D printing company, cuz I think there's a very good chance the two of us met ages ago cuz you worked at Voxel eight, is that

Kyle: 14:05

I did. Yeah.

Chris: 14:06

And we, we spent a ton of time at Greentown Labs and I know that Voxel eight has had spent some time at Greentown Labs there. And I guarantee at some point we cross paths there.

Kyle: 14:18

Oh, sure. Yeah, actually,

Chris: 14:20

Yeah.

Kyle: 14:21

yeah, it was with Voxel Eight for a while. You know, both when we were in, in the, the labs of Greentown Labs, we didn't even have like proper desk space or the software team did the e team. We just had like lab seats. And

Chris: 14:32

what's with the EEs? Always getting the worst. You know, the software engineers, they, they get all the fancy jobs at Google and Facebook and all the EEs. You know, we're always like tucked away in a basement somewhere.

Kyle: 14:42

yeah, we, I, I think they just give us a, a fancy new oscilloscope and we're happy. We're like, okay, I'm good.

Chris: 14:48

We're very content people. That's all it is. It doesn't take much to make us happy. Yeah.

Kyle: 14:52

Exactly. So it's really cool and actually all spice so are, you know, all spice. We have a desk space in Greentown Labs now, so, yeah. Quick, quick plug for Greentown. I suppose because yeah, we've had a great experience there in, starting this company. It. Well, a, there, it was just a default to kind of go back to what, you know, we, we knew Greentown, we knew we had a great experience there with Ate. And for this company, for us to surround ourselves with hardware engineers that are building really cool things you know, especially with the, with the Green Tech Initiative was, you know, a no-brainer for us.

Chris: 15:24

Here, try this dog food, see how it works,

Kyle: 15:27

Yeah, exactly.

Chris: 15:28

from it and keep spinning it. That's super cool. So then you were, you were saying earlier before I cut you off about Greentown and Vox eight getting into school, and I assume we're gonna hear an echo chamber again. You, you had a project with Valentina. I take it

Kyle: 15:42

Yes. Yes. Although I think our, our initial project is, is probably less interesting for this show is blue Apron. We were like doing a you know, we, we did get pretty immersed in the, the proper MBA a case studies. So I think that one was not, not super. Super hardware focused, but later on there's a strong emphasis to basically, you know, in the program to, to start, start companies, start businesses clearly at that intersection between like, you know, the engineering degree and the business degree. And so that's exactly what we did. I mean, Valentine and I, of course, commiserated over this issues. I talked a lot about the, the electrical engineering design release process and what that looked like. Especially at the, the startup company when we were trying to build this thing from the ground up and I thought, , here's the opportunity to go and build the system I always wanted and dreamed of, and then just ended up really being so hamstrung by. Available tools and opportunities. We, like a lot of companies, we use GitHub to manage our electrical engineering designs. And that served the very, very base need of having a centralized, even though Git is decentralized, but having a, a centralized environment for our team to work out of, to always be able to know that they're on the same same version of, of a design. But beyond that, there's almost nothing, you know, you couldn't get any of the benefits that software did in terms of you know, diffing designs certainly no, you know, CI or, or analytics or any kind of advanced capabilities.

Chris: 17:07

Okay. Well, I'm gonna, I'm gonna slow you down for a second because as, as I like to say, I'm, I'm the dummy in the room that has to ask these kinds of questions. So, as a lifelong manufacturer, one who has not had to work on product development Git is the kind of thing that I hear about constant. but really don't understand. Now, I understand that Git is not, so, GitHub is a business, right? Git itself is a technology, though. It's not owned by anybody. Is that right?

Kyle: 17:38

it. Couldn't have said it better myself. Yeah, it's a, it is a free open source technology is basically a, a protocol. So, you know, if you're using Git as a software, as a protocol it, it essentially details how your data is gonna be structured and sent over, sent over network, and, and there's a whole lot of functionality built on top of it. But that's exactly what GitHub will employ. To build GitHub as the name kind of implies. You know, they're hosting your your design files and then you can use Git to basically like push and pull from there.

Chris: 18:11

Mm-hmm.. Okay. All right. Now, the one thing that's always sort of confused me, I guess is like, so at a very, very basic level, you can, do your engineering, you're at your keyboard and you're mouse in a way, and you're working on something. You're like, okay, I think I'm ready to present this to somebody else. And that is when you commit something using Git. Right? And that allows other people to see it. And then they can then take whatever it is that, that you have done and pull that then into whatever engineering product it is that they're working on. Is that at a very low level, a very basic understanding? Did I get that right so far?

Kyle: 18:52

Yeah, that's, that's pretty close. I, I will say the difference between git and some other revision control systems is get to, to commit. and push it is actually those two steps. So it's actually two steps. Commit basically says, you know, for your copy,

Chris: 19:08

Mm-hmm.

Kyle: 19:09

in your computer, add a new revision with these designs.

Chris: 19:13

Okay, so we're going from Rev one to Rev two now? Yeah.

Kyle: 19:16

yeah, push basically says like, okay, now take that and then sync it with the server. So like, now I've done it on my computer, you can kind of pause there, check things out, make sure you know it did what you intended to do. And then that second process is, is push, and that's when it syncs it. Similarly on the, the pull side of things you can optionally have two steps. One where you just pull it down and one where you actually say like, okay, now sync it over to that branch that I'm currently on. But oftentimes people will just shorten that to, to a single step called well pull is actually that. It's a, it's combines those two steps, which is fetch and then merge into one steps where it just kind of squashes those together and then syncs it.

Chris: 19:55

Okay. Okay, so. that gives me anxiety, and I'll tell you why. That gives me anxiety. So I'm thinking to myself, okay, I'm in Altium, or I'm in in Eagle or something and I'm drawing stuff and there's a, there's my, there's another engineer I'm working with who's also drawn stuff, right? But like, how do I know that we're not scribbling over each other? You know, so I've got U1 and I've gotta, I've gotta fan it out and I've gotta put my resistors where they need to be and everything. How do I know that he didn't go do something on this board? And then I, if I try to pull that in, it's just totally obliterated all this other work that I've done. I guess I've never quite understood how you do work collaboratively in that way where you don't step on each other a little bit.

Kyle: 20:39

Yeah. Yeah. No, that's a great question. And, and honestly that's something that the software, the software discipline has had to figure out as well and that's really what we're, we're pulling from and learning from with this experience. The reality is if you do have two people changing the same thing , you're gonna have to resolve that. You're gonna have what's called conflicts. So you know, one way or the other, you're gonna have to pull up both of 'em side by side and figure out how as a human you're gonna consolidate those cuz

Chris: 21:01

So does giit have that technology where it goes, whoa, whoa, whoa, whoa, whoa. You can't merge this. You've

Kyle: 21:05

Yeah,

Chris: 21:06

something going on here.

Kyle: 21:07

And that's where you get the big red flags that are conflicts.

Chris: 21:10

Okay. All right. See, this is the beautiful thing about your, your Google Docs of the world, right? I can see everything you're changing as you're changing it. And that's, that's like the, the ultimate in my opinion, you know, is, is being able to do that. But I understand that that's not totally feasible when you need, you know, a core i seven and 32 gigs of ram to be working on something, you know?

Kyle: 21:30

Yeah. And, and there is that, but in, in that actually analogy also exists in software where, where you have sim coding, like simul SIM development. But what we find very commonly in engineering teams is that the last thing folks generally want when they're working on something is somebody else changing it. That's,

Chris: 21:50

Yeah. Yeah,

Kyle: 21:51

benefits for in, certainly in some industries that has really worked really well. And there are tools that, that do make that more manageable in hardware. But what we find is generally somebody wants to say, you know, I'll work on the power supply today. You work on the, you know, somebody else can work on the layout or someone else can work on, ingress controllers or something like that. You kind of have this, the specialist and you're working on areas and then using git you, you get revision control. You know, kind of immutable revision control. So instead of having this like very free form, you wanna build a system that is, flexible and customizable and lets you do what you want to do, but in a controlled way. Because it is important to know if you change something, what did you change? Who changed it at what date, at what revision? So that's kind of, you know, we're trying to combine those, those two sides of the world. And, and that's what I think software has done a really good job at with, you know, with the tools I get

Chris: 22:45

I want that for manufacturing. who changed the vision data on this connector? I had this thing working perfectly last week and now I now, now my machine won't place it. That's we need after all Spice io. We need all place io. That's what we need next. Actually, I am curious about how you came up with the name allspice io. That that's a, it's a very you know, aside from a very delicious seasoning what's

Kyle: 23:08

It, it yeah, very, very timely. It goes great in mulled wine. One of my, one of my favorites.

Chris: 23:13

does it really

Kyle: 23:14

It does, yeah. Yeah.

Chris: 23:15

No. But how did you come up with a name? Do you have a, do you have a story there, how you came up with it?

Kyle: 23:20

so, so Spice of course you know, has been around since the, the seventies when you know, spice simulation software was, was first developed in, in release and then just gone through multiple iterations. So from the beginning we saw, you know, there's this whole world of possibility of, of, of simulation in analytics and design data that we could start to kind of automate and tie together. Like, that was, that was the early on vision, and we ended up actually building the product that we have now. To build a platform to start to build some of that analytics and simulation on top of that's a huge direction for us going forward that you know, we can definitely go into in, in more detail. So spice for, for all spice for us was kind of the, the unifying unifying spice, right? You have your, your LT spice, your NG spice. And for us all spice was like, all right, you know, we're, we're gonna have a bit of a fun play on words as kind of the, the unifier of, of electrical design data.

Chris: 24:20

it's a good name. It's a

Valentina: 24:22

it's also a bit of an inside joke. It's always a,

Chris: 24:24

so. I was hoping there was an inside joke,

Valentina: 24:27

it's a, it's, it's a great, it's great to find your audience, you know, when people ask like, how is this related to simulation tools? It's like, great, lemme tell you all about it. And if people think we sell food, it's like, you're not a good fit for what we're doing. It's like, it's a, it's a great conversation starter and of a audience segmentation.

Kyle: 24:47

Which was kind of disappointing to some people at times when we were doing pitch competitions and we didn't bring in food demonstration like food food samples.

Valentina: 24:57

Especially at business school where we were competing with a lot of frozen meal plans and we will show up to, to our, our pitch with this full of judges audience and

Kyle: 25:07

Hungry judges.

Valentina: 25:08

Yeah. And, and the two of us were like, let, let us tell you about hardware design

Chris: 25:14

Yeah. Data simulation,

Valentina: 25:16

Yeah. And all these very technical words that we had to work really hard to kinda generalize and, and draw analogies that people would understand. But I think half of the time probably there was always like someone in the audience premier space that will totally get it. And, but I will say the majority people like just, it was that not for them. Similar with the name, it's like probably a lot of people think we sell some kind of food and not, not for you, but if you, if you are in the space, it's, you kind of get it.

Chris: 25:45

Oh yeah. Well you throw the IO in there and then it's like, oh, okay. I know what these guys are doing then. Yeah, yeah, yeah. I get it. I get it. Now they're in, they're in engineering for sure. The IO is like the universal sort of extension for We do, we do engineering things. Alright, so let's, let's talk about now, now that you've, you've gotten this tiny brain up to speed here on, on what exactly get is used for. Can you, can you bring me along a little bit? Kyle, you mentioned you work for 10 years in electrical engineering. I think the best way that I can understand, and I'm hoping our listeners can understand, , all spice had to had to exist. And why you felt you needed to build it. A great place to start would be what was your version of the status quo, right? Like what, what, what was your development process like, what was your team like? How did you guys work together? How did you make sure that you were staying in sync before you built this, this thing? You're both welcome to speak about your previous experiences.

Kyle: 26:45

Yeah. For, for us you know, I had the experience at, at, at certainly a larger company in, in seeing what they did. Of course, we, we used PLM systems pretty. pretty heavily for releases. But one thing I found was that these were, these were built for Waterfall very long long, and I would say long thought out product design cycles where we would specify in advance what are the release dates gonna be. And what I saw during, even at that company was that there's this huge push to basically be more agile was the word that got thrown around a lot. But the reality is we wanted to be more flexible. We wanted to be able to release designs more quickly more frequently. We wanted to be able to make design decisions and qualify them in a more flexible manner. In that really, that really created friction. You know, we saw for instance, you know, mechanical engineering teams, software engineering teams were really evolving their. Mechanical engineering team, for instance, we got every me at and when I was at iRobot, had the option to get a 3D printer on their desk. And that just made it so easy to like rapid prototype something, qualify it, and then they could use their, EPDM tools to like, you know, to sync. And it was much, much more streamlined process. And then of course, when I joined the, the startup company that was, it was that on steroids. You know, we really needed to move quickly. We really needed to be able to change to customer demands or, changing market conditions cuz we couldn't, we couldn't throw around our weight as a big company to say like, we need to get a thousand of these chips and the company's gonna, you know, this seems gonna make it happen. Something goes outta stock. We had to design around it. And so that, that meant a lot of flexibility that we needed. As I might have mentioned, we, we used GitHub for our electrical designs. That's what we saw pretty much every hardware startup was doing.

Chris: 28:24

And when you say your electrical designs, you're talking about your like legit literal Altium files.

Kyle: 28:30

Yeah. Yeah. Schematic files. PCB files. And then there's all the questions of like, oh, what about all the other information, the

Chris: 28:36

Sorry. I'm assuming Altium, by the way. I, I don't know what tool you were using, but

Kyle: 28:40

actually were good guess. Yeah, good guess. Yeah. Very common for small and startup companies. Even now

Valentina: 28:46

yeah. When we, when we were talking to, we talked to hundreds of companies and what we saw, it's basically will fall in one of three buckets you had on kind of the upper end your custom build system. So large company hires a firm to build them, a personalized, very bespoke software system to do exactly what they want. You saw on the other end, The general purpose tools string together with duct tape. So some, some version of Google drives meets paint, meets email chain meets OneNote, meets something else. So that was kinda like the two extremes. And, and that's what we saw. And there was very little in between. And so as a company, you were either doing things on Excel, in PDF, or you needed 5 million for a two year contract to, to be able to do something. And so that's where Kyle was talking about all his teams kind of ended up going to other disciplines, which is in this case software and being like, can we, and that's kind of the third market was let us as electrical engineers tag along to someone else's system. So very common adding electrical files to mechanical software or adding electrical files to someone else, some other team software. And what we saw over time is this trend of actually shifting to putting it on software tools. So GitHub, GitLab, Bitbucket, and more and more of this team, special electrical engineers that worked at this intersection where they're spending 5%, 20%, 50% of their time doing firmware design. So kind of working in this GitHub get world bringing their electrical files as well. So their PCBs schematics their bombs and using those tools and actually really liking it and what we saw, it's also, but you're getting a fraction of what, as a software engineer, you get out of GitHub because this is built for text-based code rather than for your drawings. So that's where kinda the allspicee idea kind of focus came about. It's like, well, we see these teams already using this protocol and this kind of workflow of. Pushing, pulling committing, but they're getting a very small fraction of what, what you actually can do if tool is built like natively for your like underlying design work.

Chris: 30:59

Can you gimme an example of that? Like what, where is it breaking, using those general purpose tools? Like what are they, what are they missing out on there that they have to find, you know? Okay. So they're trying to use GitHub,

Valentina: 31:12

Mm-hmm.

Chris: 31:12

what's broken there?

Valentina: 31:14

Yeah. One of the first things we built is actually a diff tool. Something that's ubiquitous in software

Chris: 31:19

Hold on a second. That , that doesn't exist., oh my gosh.

Valentina: 31:26

It's it's pretty, pretty mind blowing, but as you were working on GitHub or GitLab, you, you can kind of use it as a placeholder for your files, but a lot of the teams kind the what has changed, not only that my file has changed, but what in, it's different from the last version and that's kind of like the kind of going from the file level to the object level. That's where we wanted to kind of add value ourselves. So the first thing we built is actually a dif tool, which we release on our website, free for the community to try kind give us feedback and that basically does that highlights what has changed between the last version and the news version and color codes it and, and tells you like this is what's different in your file.

Chris: 32:03

and that diff tool, you're, you're, you're using it on your schematic files and your PCB files. That's what you're using it on. No kidding.

Valentina: 32:11

Yeah.

Chris: 32:11

guess.

Kyle: 32:12

also bill of materials.

Chris: 32:13

I guess I'm kind of shocked that that didn't really exist. Like , I'm sort of like, what, because we use diff tools all the time here. Like we live and die by csv files com separated value files. Like it's, everything we do is in a copy com separated value file. So we're constantly diffing things, you know, so, wow. The assumptions we make, huh? Holy moly. Okay,

Kyle: 32:35

I mean we, and we saw a lot of hardware engineers would like scrap together, and we actually did this a little bit for, for Gerber files. You could kind of scrape together like this, like at least swipe diff or something like that to but really this would be pretty hacked together. And, and we saw like, well, why don't we just codify this and, and make this part of the workflow tool.

Chris: 32:53

Yeah, cuz I guess if I was to, cuz a Gerber file is nothing but a text file at the end of the day. It's, it just follows the Gerber format, which has a fascinated backstory by the way., absolutely fascinating backstory about how Gerber was developed. But anyway, I digress. The if I was to diff that I would get, in my world, if I just use a, a, a text-based diff tool, I would get, here's all the text that's different, but that doesn't really mean anything to me. Like, I don't, okay, well, all right. What, what is this text? What is it really on the board? Where do I find this on the board to know what is different?

Kyle: 33:28

Mm-hmm.

Chris: 33:29

I'm, I'm really hoping you're gonna tell me that that tool will help me connect those dots.

Kyle: 33:34

And, and, and it will there's really two steps to the process. Like if it, you know, we convert it a file if it's not already in an ask format, like you said, to like a text-based format. And then we run the diff and then we convert that diff back usually to a, a visual format that you're familiar with. So with the schematic with the pcb, and highlight the components on there. And then if it is like metadata or something like that, like say you change a resistor value, you can click on the resistor. It'll highlight that. You can click on the resistor and then see exactly what's changed. And that's when it'll show you it, you know, we generally call metadata, but like the, the attribute value.

Chris: 34:10

That's really cool. And and you're talking about by using the PCB file the Altiium PCB file, diffing those, and yeah, sorry, I, as you were speaking, I was thinking about Gerber data, and then you said, you said, click on a

Kyle: 34:20

Yeah, I, I switched gears.

Chris: 34:22

I can't click on resistors. Okay. Wow. That's, that's fascinating. That's cool. Sorry, I totally derailed it. This, this will happen. This is, you know, I get, I get hooked onto one particular thing I find fascinating and then, and then we, we go off the chain there. But, so you b that was kind of the first step. You built the diff tool.

Valentina: 34:39

Yeah. And then from there, kinda the next is we, we talked to a lot of our, our users and asked. are the problems, right? Like what is the last reason your prototype failed? Like what went wrong the last time your board didn't work? And what we found out is the majority of the times it was a what I will call a, the process or communication gap. So it was very rare that someone, you know what? I used the wrong power or I used the wrong component. It was most of the time I ordered the wrong version or I updated this and my mechanical engineer counterpart didn't know and it didn't fit in the closure or we added this component to this version and then this other component to this other version. And one of the changes made it through, but not both. So it was a lot of those type of things. The kind of the the place where a lot of the things get caught. It's design reviews. And the other thing we found is design reviews are a very powerful kind of moment in time to kind of take stock, look at your design, see what's going, what's working, what's not working, and kind of give you the, the official stamp. And they were happening too infrequently and too late. So a lot of the changes were happening like sometimes months after you place the component and the chances that you're gonna go back and change a decision you made like three months ago, it's pretty low, right? set at this point. So, there were these very heavy, bulky meetings where you have to get everyone in the room, schedule three weeks in advance, line up everyone calendars, go through like 327 items. There's 557 post action items, maybe half of those make it through, the other half are gonna fall through the cracks. And there was just these monumental, kind, very Very heavy points in time. So we we kind of wanted to introduce the concept of kind of by running this more often and more frequently in the smaller chunks. Like you catch these things earlier, you catch more of them and you catch them like, well, you can actually make a difference. And this is what in software happened through pull requests or merge requests, which is, it's not like a three week in advanced meeting, it's more of like, I've made three changes, one change, whatever might be, I'm gonna request some feedback and I'm gonna merge it in into our brand. So kind introduce more of the kind of process management tools and introduce this r reviews to kind of wrap around what you do have the diff functionality, which is telling you what are the changes, but also kind of helps you in a quick way to get feedback, validate your design. Now that you're on track, kinda introduce your changes and move on. So by doing that in smaller chunks and more often we see teams kind moving their overall development timeline and getting their products to market faster..

Chris: 37:20

You know, this sort of reminds me of you know, you're describing tightening the timeline, and shrinking it up and doing it in smaller chunks. I remember hearing this story about when the computer spreadsheets were, you know, kind of. The Kill the Killer app at the time were, you know, the ability to do spreadsheets. Basically, the Excel of the day, I forget what it was called. And somebody, you know, somebody in the accounting department was like, oh my God, this is, you know, this takes me a, a, a month to do and I'm doing it in 30 seconds now. I, I can't believe this. And the crazy thing was, it was so difficult for them to rethink their process, that they still only did it once a month for a long time, And then somebody came along and said, well, if it only takes you 30 seconds, just do it every day. If it only takes you 30 seconds, just do it every hour. If it only, you know, and all of a sudden they're like, oh, we can just do this real time. You know, we can just constantly keep up on this. Because now they had a tool that allowed them to do that. Right? It never existed before. It was impossible. So what, you had to do it once a month. Now you can do it constantly., now you'll do it. It's no longer like, oh man, we gotta get a design review. Oh, you know, oh man, Kyle's on vacation next week. Okay, we'll push it out. And, you know, all of a sudden you can do this like quicker and, and more rapidly. It almost, it almost doesn't become a design review at that point, right? It almost just becomes workflow.

Valentina: 38:49

Exactly, and the kind of the exciting part is that the other players in the stack, which is, well, maybe before you could only do it once a month. because the manufacturing will take a month. So you

Chris: 39:02

Hey, hey, hey, take it easy. Take it

Valentina: 39:04

But that's the exciting part. It's like now manufacturing is, it doesn't take what it used to take, so now you can move a lot faster. Right. So the other players are, and we have like really quick turnaround prototypes. So you, you can run design reviews every day and you can get

Chris: 39:20

where you could get.

Valentina: 39:21

more often. Exactly.

Chris: 39:23

get those

Valentina: 39:24

So that's the other thing we've seen is like kinda by introducing like 3D printing to mechanical and kind of all these technologies that help you prototype and move faster is like, well you need tools in your systems where before again, you get me to run reviews more often. Often cause you are kind of bottleneck by something else, but now you're not. So you can move your team at the speed that you want.

Chris: 39:46

So that gets me sort of into my next question. I'm always very curious because, so, I've mentioned this on the show before, and listeners, I apologize for repeating myself, but you know, a lot of times manufacturing is sort of the black box that everything goes into, and then poof, the product comes out, right? And we're not really sure what happens inside the black box. And the whole point of this show is to, you know, peel back the cur, I'm mixing my metaphors here, but it's to peel back the curtain and, and show people what is happening in manufacturing. And the more collaboration that we can have with, with all disciplines, the better all disciplines get. So, for example, we, you know, we've talked to people on the show before from from e d a tools, right? People designing the software, you know, your ltms, your Autodesks of the world. You know how rarely they actually talk to manufacturing. They're talking to, to engineers. They're talking to people who are using their software. They're not talking to manufacturing, not, I mean, they are, but not very often. And yet we're having these conversations and all of a sudden it's like, Hey, did you know, are you aware of X? No, I didn't know about X. Well, man, if you could do Y for me, it'd be, of course we could do y That's the easiest thing in the world. And all of a sudden you have these things that take place. I'm curious about those kinds of conversations being you know, between, I would say between product designers and engineers and manufacturers and using software like this to, to make those conversations even easier. Is, is there, is there a hook? I guess it's all about me at the end of the day, Valentina and Kyle, it's all about me. I wanna know what I can get out of all spice as a manufacturer. where, where, where's it gonna make my life easier?

Kyle: 41:22

Yeah, absolutely. I mean, and as you point out manufacturing is, is a huge part of the process and in dfm, I I is a lot of how we hear, how we see it. Like just design for manufacturing step. You know, here's yeah, we got your design file. So here the issues, maybe it's some valor runs or some other tools or, or, you know, pick and place design scrub or something like that, or components that we can't get and we're seeing more and more. You know, because we are a cloud-based tool. We initially made it intentionally made it very easy to org owners to be able to create teams for external collaborators. And so that means that that designer review process can happen for your internal teams. And you can also create different designer view processes with potentially different files. Maybe it's gerber and export files and you can start to pull in in contract manufacturers such as yourself. And so that all that information kind of live in the same place and the, the, the pieces kind of flow through a little bit better.

Chris: 42:20

So I will have visibility into what this P C B looks like, what the geometries are. I can see if through hole pin is too close to a surface mount device. I can see potentially if you have a footprint issue. My dreaded 0 4 0 2 footprint listeners of the show are probably laughing their heads off right now because they know how much I hate bad oh 4, 0 2 footprints, So I, I'll have visibility into that. I can, so I can get a login from my customer and they, they can say, Hey, come, come check this out. What do you think?

Kyle: 42:49

Yeah, that's definitely a world where we're getting into this especially recently and, and going into going into 2023. This is a big, big part of our, our expansion.

Chris: 42:58

Cool. forward to that. You gotta let me know when it comes out. I wanna check it out.

Kyle: 43:03

Nice.

Chris: 43:03

So it's because like, the truth is and there's this toss it over the wall thing that happens a lot of time in manufacturing. Right. And , I was actually just on a, on a another podcast recently about engineering and you know, it was, it was a mechanical engineer who hosts the show. And here I am coming in an electrical manufacturing engineer. And you know, it's, it's, both engineers, but we speak two different languages. You know what I mean? So he goes, he goes okay, well if I'm if I'm designing this hardware product and, and there's a PCB on it, can I just give you the Gerber files and you guys can build it? I'm like, oh, gosh, no, no. Like , you know, there's this, there's this disconnect and, and I, I have to imagine when you have different disciplines, you have your mechanical engineers of the world, your electrical engineers, your software engineers of the world, I'm hoping that software like this, like All Spice can help these teams work a little bit more closely together and they can see, oh, you've got an eight millimeter radius on this corner. I, I, you know, I gotta make sure to square off my, or put a rounded rectangle on my PCB so that it can fit this eight millimeter radius on the, on the corner of your enclosure here. And, and, and are you, are you able to, to bring those things together?

Valentina: 44:16

One of the things we very intentionally did about our product and sometimes electrical, it's the same way, right, that you're talking about. It's a bit of a black box and it's that toss it over the wall and toss it over The wall normally comes down to pdf or an email, right? And it's a you, you lose a lot of information. So one of the decisions we made for Allspice was to actually allow collaborators to have free access to the platform, right? So for teams, they can add as many users as they want and kind of bring them into their own space and kind of give the full context rather than sum summarizing it to like the, the lowest common denominator and tossing it over the walls. So we see a lot of teams actually bringing people even earlier in the process. So rather than waiting till things are ready, improve, bring your mechanical engineers to your design review and get those questions answers like before it becomes an issue. And, and we see the same for, we have these features called releases, which it, I guess is surprise for us was how many of our teams use it to talk directly to their manufacturers and kind of make it more of a, of a two-way street collaborative process rather than wait until everything has been decided and kind of bring people earlier on to, to be part of those decisions.

Kyle: 45:28

And the collaborators can, even, as you say, if they're mechanical engineers, they can come in and they can comment directly on the PCB and say, Hey, this mounting point is, you know, too close or off size, something like that. And that's all still included in that, that collaborators piece. They can attach files and documents and stuff like that. We only charge, basically for for users that can, can push, you know, that get, commit,

Chris: 45:48

Mm-hmm.

Kyle: 45:48

access,

Chris: 45:49

But you can comment.

Kyle: 45:51

comment, touch files, create issues.

Chris: 45:54

All right, move this oh 4 0 2 resistor away from my through hole pins. I need to solder of these things, and I don't wanna wipe out your oh 4 0 2 resistor

Kyle: 46:03

Or for the, for the operations folks, find a second source for this component in this one, in this one.

Chris: 46:10

Yeah. That is, let's, let's be honest, that is the only thing anybody cares about in the past two years is trying to get components. We've, we've seen some customers get really creative, like totally like they're designing their products now with the idea in mind. It's like, okay, I can only get this, this device in an s o t 23, but I have another device that'll work, but it only comes in an s o eight and I have another device that'll work that only comes in s o t 25, and they'll lay it out with all three, you know, and then they'll just tell us, only populate U one here on these boards. But then when you run out of U one start populating U2 and , it's, it's man that has been, that has. That has been my past two years of, of trying to figure out like how to help these customers with sourcing parts. Believe me, I get it. But, so I can, I can type super passive aggressive comments on there and be real snarky and everything inside jokes, all that kind of good stuff.

Valentina: 47:09

Yeah, you can tag people. Send them notifications

Chris: 47:11

Oh, you can tag 'em too. Oh

Kyle: 47:13

Yeah. And we've, we've introduced emojis to the world of electronics, so I'm not sure. Yeah. It might be a dangerous, dangerous tool at our, our disposal now

Chris: 47:22

that's a dangerous tool. But, but if you really want a dangerous tool for Chris Denney to gets hands on animated gifs if you allow me to put an animated GIF on it, there will be an animated GIF on it. I promise you, , there is an, there is an appropriate animated GIF for every conversation. I'm just saying they

Kyle: 47:39

It is. It is. It is amazing.

Chris: 47:41

It's an art, it's an art . I, this is, this is great.. I have other questions. Is there other things you, you guys wanted to hit on? Did we not get a chance to talk about a particular subject you're, you're particularly interested in? Especially as it concerns the product you guys are building?

Kyle: 47:55

Yeah, I mean, you hit on a really good one there at the end too, which is chip and supply chain. We're, we're really focused on in the coming months. Not only that, that dfm in, in kind of CM integration, but also in, of course as you mentioned, supply chain and library management. So we're really taking a closer look at that, working on a lot of, of new, new features. We are., especially as we start to look at, again, bringing the All Spice into it. Continuous integration and continuous deployment, ci, cd and what that means for hardware and things like supply chain analytics, we feel falls squarely into this category of, you know, now that we have this information, we have these design reviews that are running in the cloud. We're bringing everybody to the table early and often. Lets start to now, you know, not only run diffs and not only should let you, you know, comment on files and stuff like that, but start to automate all of the, a lot more of these steps that you're, that you're already doing. And you're doing manual, which is like, you know, download a bom and upload it. Or start to do some of these validation of, of, it could be something as simple as, you know, do all these components have you know, multiple manufacturers or something like that. Certainly a big big part of the world we're in.

Chris: 49:00

Paul, we call those cross references.

Kyle: 49:02

Yeah. Cross refs.

Chris: 49:03

yes. Super cool. That's, that's, that sounds very exciting. I, you know, a lot of times we are given, you know, I guess it depends on the engineer it depends on the customer. We're usually given one part number, right? It's like, use this Yago, RCO 6 0 3, blah, blah, blah, blah. But honestly, they'd probably accept., any Panasonic, any Samsung, anything with equivalent values on it, you know, and any tools to help help our customers validate those cross references and then supply us those cross references will make everybody's lives easier. Totally.

Kyle: 49:38

Yeah, it ranges so much as, as I'm sure you've probably seen, I mean some companies are probably like find me and 0 4, 0 2, that's a hundred k, and like that's it. And some, I'm sure, like this is the manufacturer, this is the part number,

Chris: 49:52

Oh yeah, yeah, absolutely. Well, you know what gets even worse, what gets even worse is you have I don't know if you're aware of this, but like, when it comes to resistors, it's pretty easy. It's like, okay, it's 10 K resistor, 5%, whatever. But in the world of diodes, you'll have a diode with, you know, A one N 4 1 4 8 W for example, and there'll be 10 different manufacturers of that exact same part. but they each have different value. Like they're, they're not all the same. That's the, that's the maddening thing. So we do get that, we do get customers like, no, you must use on semis one N 4 1 48. You cannot use Toshiba's. You know, and because it has a different value, it's not gonna work for our product. Right. It's, it's gonna affect the form, fit and function of our product. You cannot

Kyle: 50:42

feel like capacitors kind of, kind of gets into that realm too sometimes.

Chris: 50:46

They can't, well, the, the ceramic capacitors, you know, the resistors are the easiest, the ceramic capacitors are the next easiest, and then it sort of falls off a cliff after that. But yeah, it, it, it definitely can for sure. So anything to help, like what are the key values for this location? So, for example we had a customer recently who came to us and they, you know, obviously it's fresh on my mind. That's why I'm talking about the 4 48. They had like 13 of them on their board. 12 of them really didn't matter., they're like, yeah, it's fine, but this one location, we really need just this one manufacturer. And I think what would be key, you know what, what would be key for, for your manufacturer to know is why, right? What is it about this location? What are the key parameters about this location that I need? You know, you can't, you can't just use any temperature coefficient, blah, blah, blah. It needs to be this temperature coefficient. And, and, and like, can you, can you highlight these are the key parameters and those sorts of things? Because we can take that and we can go, Hey, look, we're experts at sourcing. We we know way more suppliers than you guys do, trust me. We can find these parts for you if we just know what to look for, you know, and, and we can help you find those cross references. So,

Kyle: 52:01

you mean that that customization's really important And, and that's exactly it, is that different companies have different requirements, have different regulations in, in letting something be flexible enough that you can make those specifications, you can do checks for, you know, that one, that one diode. If, if you need to make sure this one remains this. But, you know, some, some companies and products might not have those, those requirements.

Chris: 52:23

yeah, yeah, yeah. But let us know that it's not required. You know what I mean? Let us know, like Yeah. Any, any 10 K will work, you know? 5%, 1%. It's all good. Alright, so this sort of I, I always love to ask people we have on the show about, you know, sort of, obviously you're working on something, you're trying to build something, you're trying, you're trying to create something to make everybody's lives easier. And, we've spoken, this whole conversation has been at a very high level. It's, it's teams working together, it's development. Right? Let's get into the nitty gritty. Please tell me you've got some kind of a horror story or some specific example that you've experienced where you thought to yourself, man, if only I had all spiced five years ago when we totally screwed this up and lost thousands of dollars.

Kyle: 53:07

You think So early on we talked with, I know, I'm like, I'm like kind of passing, but I remember one of our conversations very early on in our kind of like discovery phase was a, a engineer at a, a large like Fortune 10 hardware company lost an entire you know, pre-production run. So not full production, but this is like, we're, we're gearing up, we're getting close because of a, a connector that was rotated in swap and they saw, you know, you see a hundred sub assemblies built. I know when you're at the production line and you're going, these things don't plug in together. You know, what are we, what are we looking at here? And it was purely because of a communication meltdown. I saw the same thing at a past company where, where we had, yeah, the diff would've just helped out so much. Even that alone would've helped because we renamed. Net renamed to the same net that art existed on my board. Whoops. Pushed that to pcb and when we passed it over to the layout person, they happily connected, you know, to I, I think it was, yeah, two very very you know, discontinuous, you know, segments of the board together

Chris: 54:16

Ground in 3.3 volts,

Kyle: 54:18

yes. And bad things happen. And again, of course, this was like late in the design process. This is not like an early prototype. It was, you know, a quick, easy change that unfortunately spiraled.

Chris: 54:29

you, and if all Spice existed, that, that would've highlighted itself as like, Hey, hey, hey, hey, this thing changed here. Is everybody okay with this change? And somebody could have raised the red flag and said, no, no, no, no. Do not do that.

Kyle: 54:40

Yeah. We're just seeing the, the board light up probably on that diff and going, mm, this seems like more things have changed than we expected.

Chris: 54:48

So

Valentina: 54:48

Yeah, it happened even in our demo. We were putting together a demo for we have an open source demo that people kind of go check out, play around with it. And we, we were doing a component swap validation for something that was out of stock and I guess like real life, we didn't know. The casual, I don't know what happened. There's this test point that was removed in a totally different part of the board and we kinda left it in there cuz this is like the exact example of like how actual engineering works and it's like this is something would've order, would've gotten in house and it would've spent three weeks trying to debug this and figured out what was missing. And it was just so beautiful to see the, like the bright red do being like, did you really intend to leave this? Is this how it's to go? And we kinda left it in there. You know what it's like, this was supposed to be a demo, but we, this is how it actually goes. So we're.

Chris: 55:38

So cool. That's so cool. That's so cool. Well, I then, I just learned how, how to make my life easier thanks to all.. I can't tell you how many times we get blamed for something not working, and I usually, I usually don't respond. Like I'll usually respond quickly. I'm like, oh, I'm sorry to hear that. Yeah, we'll help you out anyway. You know, we'll, we'll try to figure out what it is, but I don't, like, I don't, I'm not like, here's my u p s number. Send it over right away. Because nine times outta 10, they come back and they're like, oh, we figured it out. We accidentally did blah, blah, blah, and we did a jumper wire and had an match cut, and we're all good. Now it's like, yep, , it's, I, maybe not nine times outta 10, but it's, it's pretty common. Don't, don't get me wrong, we definitely make mistakes, listeners who have experienced our mistakes. I get it. Don't, don't go shouting at me saying that I'm blaming the, my customers for their problems, but no,

Kyle: 56:28

ha. Having been on that side, I, I can attest. Nine times outta 10. It was me.

Chris: 56:34

Yeah. Yeah, exactly. I'm actually dealing with something right now. I designed this little fixture for, for a product that, that we build here. Just trying to. Make life a little bit easier for our assembly team and just got the mechanical dimensions wrong and I'm having broken pieces. I'm like, ah, you know, it must be the manufacturing. The manufacturing is broken. It's not my design. That's force giving it too much force. It's gotta be something that they did in manufacturing, right?. Yeah, for sure. Alright, so I will say, you know, another thing we like to do, you guys are obviously very experienced, qualified expert engineers top of your field you know, poet laureates of mechanical and electrical engineering, right? Any tips or tricks or advice you, you wanna offer young people who are, who are just kind of getting started as mechanical engineers, electrical engineers, things that they can focus on mistakes you've made that you wish you had avoided? You know, here's, here's the opportunity for sage advice.

Kyle: 57:28

Some of this mistakes, especially when I was starting out in the industry, were some of the best, I don't know, maybe decisions, and I say mistake in the way of that I didn't know things wouldn't work. At one point I developed a really cool capacitive sensor for one of our robotic applications because I didn't, I just thought this was something we could do. So yeah, throw a capacitive sensor on the top of, you know, front of our robot and, and see what happened. And every senior engineer I talked to later on said like, that's, that should never work. And I'm, oh, look at those work. And knowing what, you know, I, I'd learned over the years, it was like, yeah, that, that should never have worked. But it, it did because I didn't know enough to. Yeah. I don't know at one point why I had a capacitor sensor in the front of a robot. It could detect everything for via proximity. Even down to like wood and plastic was the most difficult. But basically just because of the, the changing material properties, it could detect it like a fluctuation in I think it was, oh God, going back per permittivity, like the, the air gap permittivity and sense of change and then basically respond to that. And it was way outside the realm of what I had experienced and at the time, but just kind of threw it together. So I guess my, my advice is just like, go for it. Try it out.

Chris: 58:41

To the old people. Just go ahead and do it.

Kyle: 58:44

go, go for it. I mean, you know, on, on a desktop or, or benchtop task, it's like, yeah, just try it. Just do it. You know, once you get into manufacturing, it's a whole different story. I, I relied a lot on simulators, hence maybe part of the inception of, of our company name. But I try to simulate everything before I send it out the door cuz I, you know, I didn't trust myself enough to, to do that. And that also worked out really well. So, I guess two flip size, like one, try everything, but like, before you send it to production, maybe, maybe test it a

Chris: 59:13

Before you spend thousands of dollars on it, let's just make sure. Yeah. I, it's funny, your, your, your point resonates pretty strongly cuz I was, I, I literally on my drive in this morning, I was thinking about this exact thing. It's, I don't know if you, if you recall or have seen this i b m commercial where they talk about ideas and how fragile ideas are and, and it's this it's this, this like ugly looking monster that's progressively getting more beautiful as it as, as the commercial progresses. And then it's this like almost looks like a peacock at the end. It's very colorful and bright and it's like, hey, this is the ugly idea that nobody thought would work, but oh my goodness, we've brought it into the world and then it's amazing now. And . It's so true. And, and I had an experience recently where I realized I had shot somebody down, you know, and I said, no, no, no. Let's, let's don't do that. It's, it's, it's not gonna work. And I thought to myself, how stupid, because it, you know, if, if I let this person do it, it's not gonna cost hardly anything to let them do it, first of all. Second of all, if it works, freaking awesome, right? That's great. It worked. And if it doesn't work, they're gonna learn on their own. They don't need me telling'em it's not gonna work. They'll figure it out. You know, they're smart people, you know, so it, it resonates and, and I owe this person this conversation, which I will have with them later today. So rather than shoot them down. Yeah.. Yeah. How about, how about yourself, Valenti, do you, got you got any horror stories or real, real world examples of where you really wish you had had something like All space in the past? Or perhaps, cuz you have, I, I see that you have quite a bit of mechanical engineering background. where you did have these kinds of tools in your world and you're like, , Kyle, we had that 10 years ago.

Valentina: 1:00:54

I was trying to think of kinda what I wish I knew as a starting out and I was trying to think of where the majority of my problems happen, and I think it's some version of. Over communicate or, and I feel like a lot of my issues happen at the intersections. I was trying to think of like, what are some horror stories? And I remember walking into our warehouse once where the conveyor belt was 50 feet from the unloading dock. And I'm like, how on earth did that happen? Like, how did we not talk to each other? To there? There's some clear gap here, right? Of the, this is like both parts were done correctly. They just. Line up. They didn't work together

Chris: 1:01:41

See, that's where marketing

Valentina: 1:01:42

totally useless. Yeah. And it was totally useless. It's like my conveyor belt was like perfect and the construction was perfect. It just didn't work together and it didn't matter how great it was if it didn't play along with like the next step in the kind of the, the whole building. Right. So I feel like those type of things were some of my biggest lessons about kind of paying attention at, at the address, at the intersections. That's where a lot of my problems happen. Like kind of, it's not only about my, my part of the problem, but like who, who comes before and who comes after. And like those transitions and like making this move. Right. When you're talking about manufacturing, it's not about my design being great. It's also like, am I communicating everything that needs to happen for this getting manufacturer? It's like kinda like my job. When I started, I thought my job ended on like my product, but it doesn't, my kinda, my job ended like when the next stakeholder or the consumer or whoever came next, like. Fully understood and kind of that transition went smooth, then my job was done. And I feel like that was a mentality shift that I had to do after making a few mistakes then along the way.

Chris: 1:02:45

Mistakes. Mistakes are super valuable. You do learn a lot from 'em, don't you? You know, like they, they're super valuable,

Valentina: 1:02:51

That never happened again. I, I, I checked our, our docs every single time after that and I made sure that we overly communicated going forward. But that, that was great learning opportunity.

Chris: 1:03:04

See, I, I was gonna, I was gonna suggest you, you get marketing involved there and you, you call it, you call it, this is our rec center. The, the, the space between the dock and the conveyor. This is our recreation center where we, where people get a chance to work out and exercise and run 50 feet, 50 foot dashes all day.

Valentina: 1:03:20

Yeah. Yeah. In reality, we're not being more of a, of a, of a sprint of like, how do we fit this space with temporary equipment that we can then, and kind of how do we face plan, like actually getting this to the right place. But I think it worked out in the end, but it was just like a, I remember it was a shell shock, and I've been working on this for, for a long time, and it's like, how did I not check that before? Like that, that was like one step, like above my, my product, right? Like my scope. It's like if it's just one step and I just couldn't see, I was just so focused on what I needed to do

Chris: 1:03:52

Yeah.

Kyle: 1:03:53

was the end result to turn up the speed on the conveyor belt and just kinda launch 'em out to the the production gap?

Chris: 1:04:00

put a, put a little bit of an incline on it so it just launches it right into the truck. Yeah., that's brilliant. that's brilliant. Yeah, that's that's great. I appreciate you guys coming on the show. I love having these conversations. This is an area where I would never have even thought to have a conversation with folks like yourself, but it's super interesting and I really look forward to, I will say listeners, if any of you are already using Kyle and Valentina software please reach out, let us know. Like we'd love to hear what your experience has been and I'm sure they would love to hear it as well. And, and I'd be, you know, cuz again, once again, it's all about me. Have I built anything that has gone through all spice and design reviews? You know, that'd be really cool to find out. We'd love to know. As always, we love to we love to invite our guests to share my favorite segment of our show. Really, the only reason we do this show is to air our pet peeves. This is why the show exists, is to get our pet peeves out into the world. And I did not prep you guys for much, but I did prep you for this one little bit. we, we have, we have a plenty, so if you, if you're not prepared with one, that's totally fine. We could go off for days on our, on our own. But do you have any particular real world pet peeves? You know, something that just you wish, you know was, was better , please share

Valentina: 1:05:16

I have one, I cannot stand sharp edge buttons. It just

Chris: 1:05:24

edge button

Valentina: 1:05:25

like, if this, if it's a square with a sh it has to have the rounded edges. Like it's just gonna be sharp.

Chris: 1:05:31

like on a website and

Valentina: 1:05:32

yeah, like a software in, like if you're filling out a form, if it has like sharp edge like corners

Chris: 1:05:39

you must really hate Windows eight then. Like Windows eight must just drive you crazy. Then

Valentina: 1:05:44

I think it just, it just instantly makes me feel like I'm going back in time, 20 years and it just canal, no, it just instantly makes me feel older and like no, I, he has to be rounded.

Chris: 1:05:55

I'm on all Spice dot aisle right now, and I see, I see the learn more button. That's a nice rounded rectangle. That's good, that's good. Schedule a demo. Nice rounded rectangle. All right, you've checked all your bo even your forms are rounded rectangles. Very good. All right, so you, you eat, you eat, sleep and breathe the rounded rectangles. That's good to see.

Valentina: 1:06:18

Like I can probably, I probably lost count of how many times I'm asked to redo a form that did not have rounded edges.

Chris: 1:06:27

we have on the new machines we bought, we have physical buttons, right? And they have , I'm not sure how to describe this, but they ha they have sort of a, a, a, not a cover. There's not a cover over the button, but there's sort of a oh, what's the word I'm looking for here? The buttons sit within this sort of a, a, a, a rectangle. So the button itself is a rectangle and the then the button sit inside these walls, these four walls surrounding the rectangle. So you can't accidentally push the button, you know, but it's all square and the, it's, you know, you, I've caught my finger on it a few times. I think I would prefer if it was a nice rounded rectangle rather than these square corners, you know, just saying it can work in the real world too. I'm all for it. Where I'm, see, cuz I know, like I'm looking, I'm looking Windows is I'm, I'm on a Windows PC at the moment. Full of square icons, square buttons, square icons as far as I can see everywhere.

Kyle: 1:07:26

I feel like design's an endless an endless circle. So at some point everything's gonna go back to, to sharp edges. So

Chris: 1:07:34

Ooh. I wonder if I wonder, but I wonder if Windows 11 has gone rounded rectangles. Cause I'm not running Windows 11 at the moment. Maybe. Maybe, maybe. The pendulum's still swinging that direction. I, I'm all for it. I love the rounded rectangles personally. You know.

Melissa: 1:07:47

I'll have to go check CircuitHub after this.

Chris: 1:07:49

oh, uhoh. Real time. Real time.

Melissa: 1:07:53

I, I, I think all the pages, all the marketing pages, I think we probably have run rounded rectangles. All the pages that I have control over I don't know about the.

Chris: 1:08:04

All right, we're gonna go to the secret internal page. Let's see, what do I have? Oh, no, I got nice rounded. Ooh. Even the tabs are rounded rectangles on our internal pages. Pretty nice. Very nice work. Very nice work. Valentina would be proud. How about you got anything? Kyle, you got one for us?

Kyle: 1:08:21

Yeah, I was gonna say my, my pet peeve even, and, and maybe this is. It's super close to home is proprietary final file formats.

Chris: 1:08:29

Mm.

Kyle: 1:08:30

what the heck? I'm sure you're the, you know, the software you deal with, Gerber Files obviously built on, on G Code, which I would love to hear that story from you sometime. So that's like, at least, at least it's text based.

Chris: 1:08:44

Yeah, Gerber is, Gerber is a proprietary format,

Kyle: 1:08:50

true. Yeah.

Chris: 1:08:50

but it is, it is stewarded by oh, I can't think of the name of the company right now. And they, they publish, it's a proprietary format that is, Open sourced ish, like it's fully published. Like there, you don't need, you don't, you don't need to sign an NDA to, to work with the format. I, I don't exactly know how it all works, but, you know, presumably the software that we pay for is paying this steward of the format for the rights to, to, to, to use. I have to imagine. But that does exist in our world. So we have we have Melissa, you probably remember the number better than I

Melissa: 1:09:29

IPC 2 5 81.

Chris: 1:09:31

There you go. 25 81. So IPC 25 81 is the open source version of ODB Plus, plus ODB Plus Plus I believe is owned by well, I'm not sure who owns it now. I think it's Mentor Graphics would own it now. Cause it was a valor format and I'm not sure who has it now, but ODB plus Plus. So, so to your point, see ODB plus Plus is sort of like the, the XLS of the world. Like it's the, it's the one that everybody effectively uses, even though the dot, oh, I can't think of what, I can't think of what the open source version of, well, let's say the dot doc of the world, and then even though Dot RTF exists, but everybody still uses Dot doc. You know what I mean? And so trying to convince people, they're like, no, no, no. Go to the ipc. It's just as good has been, has been a bit of a challenge. Yeah. I'm, I'm, I'm with you there.

Kyle: 1:10:19

That's great to hear. Great. That I, ipc 20 25 81 is, is maybe making some some gains

Chris: 1:10:26

It should, I mean, it's, it's, it's gaining wider support. CircuitHub supports it. Fun fact though, the, the, the e d A tools of the world, not everybody exports it

Melissa: 1:10:36

Yeah, they claim to support it, but they don't actually support it according to I P T

Chris: 1:10:42

Yeah. It's, or it's like five versions old or something like that, you know, it's like, come on guys. Because it's like, well, we have ODB plus plus nobody uses 25 81. It's like, no, what it, nobody uses 25 81 because you're not using it. Right. chicken and an egg problem

Kyle: 1:10:56

chicken in the egg. Yep.

Chris: 1:10:57

Yeah. Somehow it, it ends up, you know, I, I I, I, I tend to adopt people's pet peeves, so I'm, I'm fully on board with rounded recs. I'm fully on board with the open source file formats as well. Yeah. That must be a fun world you live in trying to support everybody's different formats and, and you know, I'm sure you have to sign all kinds of NDAs to get to the, the, the dot PCB files that everybody uses.

Kyle: 1:11:22

It, it, it keeps us, it keeps us busy. But that's kind of like our, our value prop. That's kind of why we're here to, to let us help with that because we see, you know, we've seen engineers start to try to take these things on. It's like, that's which, you know, when I was an engineer, I tried to take these things on rendering files and automating things and getting more data outta my designs that, doing things that I wanted it to do. And, and and yeah, that's kinda where we come in. It's like, you know, we'll, we'll work with you to, to make those things happen productize and just make it tenable for everyone.

Chris: 1:11:51

Beautiful. Beautiful. Well, I think that that is an excellent way to wrap things up. I, I've loved this conversation. I had a feeling this would be a fun conversation, just reading , reading your background and reading what you guys are working on. I'm like, oh yeah, I think we're gonna hit it off. I think this is gonna be good. And yeah. And I have lost Oh yeah, yeah. Too many tabs. Too many. There's a pet peeve for you. Too many tabs.

Kyle: 1:12:17

Yeah. In, in on our website we have a, a, a couple resources that can also really help. We just put together GI for hardware guide, which has been super helpful. If anybody wants to dig into all the, the, the nitty gritty and see how, how Git can work specifically for hardware that is on our website. That is, that is totally free as is the the diff tool up there. So

Chris: 1:12:35

very nice. Very cool.

Kyle: 1:12:37

my, my quick plug.

Chris: 1:12:40

Cool. Appreciate that. Yeah. by all means. So, if if you don't feel like looking at the show notes and you just wanna get in touch with these guys, whether you can email us at contact pick place podcast.com, we can put you in touch with 'em. As always, we are at CircuitHub and at w Assembly on Twitter. And we've been mentioning this at the end of each episode lately. We would love to hear from you guys, so please write in and let us know comments, what you think of the show, but suggestions, show suggestions because you know, we can talk about anything forever. But is there something we're missing? We would love to, we would love to know about it. And please tell a friend that's the best way really to spread the word about these kinds of podcasts. You know, ratings and reviews are great as always. We, we love those, but please tell a friend and somebody you think might be interested in it and we would love to have 'em as a listener.

Melissa: 1:13:28

Yeah, absolutely. Thanks for listening to the Pick Place podcast. If you like what you heard, consider following us in your favorite podcast app, and please leave us a review on Apple Podcast or wherever you get your podcast from. Thanks so much Valentina and Kyle.

Chris: 1:13:44

Yes. Thank you very much and thank you,

Valentina: 1:13:46

for having him.

Kyle: 1:13:47

Thanks. Our pleasure.