On-Demand Webinar: Why Robotics Hardware Can’t Keep Up with Robotics Innovation

Transcript - On-Demand Webinar: Why Robotics Hardware Can’t Keep Up with Robotics Innovation

Jim Mayer
Hello, everybody, and welcome. My name is Jim Mayer. Innovation in robotics is moving super fast right now, like speed of light type fast. We’ve got humanoid systems. We’ve got autonomous platforms. We’ve got AI enabled machines.

Things on the software side are accelerating at a very fast pace. The hardware side, sourcing it, manufacturing, scaling, that’s where we see a lot of pressure building, and that’s what we’re gonna talk to talk with you about today. I have three people who are gonna give different perspectives from different parts of this problem. First up, I have Ted Rozier from Festo Didactic.

Ted, welcome. I’ll let you introduce yourself here in a moment, but welcome to the panel. I’ve got Courtney Fernandez with Relativity Space. Courtney, thank you very much for being here today.

And last but certainly certainly not the least, I’ve got Fathom’s very own Bobby Youssefi. Bobby, thank you for being here. Great to see your face.

So like I said, I’m gonna give you a chance to introduce yourselves. Ted, you can do your intro way better than I can. So go ahead and introduce yourself. Tell tell the people who you are and and just real briefly what you do.

 

Ted Rozier
Yes. First of all, thanks a lot for, the the the opportunity to to to be a part of this. It’s exciting topic like you mentioned, Jim. But, yes, Ted Rosier. I’m a director of digital advanced technology and robotics at Festo Didactic.

Leading up to my twelve years at Festo, I spent a little over eighteen years designing, developing robotic solutions for automotive, pharmaceutical.

We were a system integrator all my life. So at Festo now, I drive the engineering team as well as business development when it comes to custom solutions. So anything having to do with emerging technology comes across my desk, and we build that out for education, always following the principles of industry.

 

Jim Mayer
Awesome. Awesome. Thank you. Courtney, you’re up. Let’s hear about you. A little brief intro.

 

Courtney Fernandez
Oh, so I’m Courtney Fernandez, and I cohost a podcast called Automation Ladies with two of my BFFs.

And my day job has me playing with robots for manufacturing purposes.

Most of my career has actually been like systems integration of some form. I have several years of machine vision, but I mostly am interested in like six axis robot arms because I think they’re the most fun.

And I have even kind of further in my past some embedded systems experience but I got out of that pretty fast because I was mediocre at it and not super enthralled by writing software and then wondering why it didn’t do the thing. So when the robot doesn’t move, it’s a little bit easier for me to understand what’s happening. So I like machinery and manufacturing and that type of thing so much more than I ever did, you know, designing circuit boards.

 

Jim Mayer
Awesome. Awesome. Bobby, again, last but not least, give us a little intro on on you.

 

Bobby Yousesefi
Hi, everyone. I’m Bobby Experiencing some technical difficulties right now. Apologies, guys. Let’s take that from the top.

I’m Bobby Youssefi here. I am with Fathom’s robotics and autonomy vertical. I have been in manufacturing for about a decade now. And similar to Courtney, I got my start in the software space, had the opportunity to jump into the hardware world again about a decade ago and haven’t looked back since.

I was listening to Tim’s, excuse me, Ted’s intro, and it was interesting. He mentioned that he’s touched multiple industries. And I feel like in manufacturing, are blessed because we have the opportunity to work with some of the coolest companies and the coolest customers in the world, whether it’s an aerospace, automation, automotive, biopharma, you name it. And so very excited to be here today with you guys as well. Before Fathom, I have kind of jumped around the manufacturing space. I was an early sales hire at Xometry. And really my biggest exposure to robotics came through my time in Makena Labs, I was there for about almost two years.

 

Jim Mayer
Awesome. Awesome. Thank you very much.

We’ve got a quick poll for the audience. You now know the panelists.

Before I get into Courtney’s first question, I do wanna just take sixty and ninety seconds, and that might be too long, and really understand what your role, our dear audience member, what your role is, in robotics or advanced manufacturing as a whole.

And, panelists, you can see the answers. Right?

 

Ted Rozier
Yes. We can.

 

Jim Mayer
Alright. So fifty per forty percent are adjacent or curious about the space. And I kinda like that for, an audience when we’re going to be diving deep into a pretty interesting topic.

Operations.

Okay. So we’ve got engineering. We’ve got executives. We’ve got operations.

Nobody from supply chain, and they might be the ones. So all of you executives who are on here, make sure to send the recording of this to your supply chain people.

So let’s get into the questions. My first question is, of course, to the entire panel.

But as I was researching for this this discussion, one number kept jumping out at me, and and that’s the fact that lead times for robotics grade components right now are running about twenty six to fifty two weeks. Like, that’s a long time for components for robotics. And it’s what I have seen, it’s that’s, like, routine. It’s it’s not an exception. So, Courtney, for you, for someone working in hardware development, what does that mean for your program?

 

Courtney Fernandez
A lot of times, it means that you’ve you’ve gotta draw up your best guess at what stuff is gonna be when you get the hardware and hope that that lead time is still the same lead time later on in the program. So you have to kind of be flexible in changing the plan multiple times throughout the program, first of all. Or to be able to have divergent paths like in your program. So you can develop towards one solution, but be ready to kind of pivot.

I don’t wanna say throw it all away, but take the other fork that you hopefully didn’t fully abandon.

 

Jim Mayer
Yeah. Yeah. Ted, what about you? Is there something you would like to add to that answer?

 

Ted Rozier
Yeah. I think that, like Courtney mentioned, you can go back to previous projects that you’ve worked on and you try and hold on to a library of suppliers and put yourself in a position where you can pivot.

That’s the main thing.

It’s really important that we like standardization.

That’s something that everyone knows that when it comes to standardization, that’s how you start to make your money.

So being able to forecast, kinda push fast forward on your process and then rewind.

That kind of brainstorming and that type of culture, if you build that up, you do put yourself in a position. I like to relate it really quickly to basketball.

And there’s a position we call triple threat. You can shoot, pass, or drive. Okay? As long as you can take that mentality into your engineering process, got a fighting chance.

 

Jim Mayer
Absolutely. Triple threat. I like that.

Bobby, Ted, we’ll talk basketball after. Bobby, there seems to be this very real mismatch between the iteration cycles of software and hardware. You and I spoke at length about this.

Software team can push an update continuously or do push out updates continuously, but the hardware team has that twenty six to fifty two week lead time. How does that play out in the world of robotics?

 

Bobby Yousesefi
I think it’s actually a really interesting point that Courtney made earlier and Ted kind of added to it, which was this flexible state versus a stable state. And I think our supply chain today is really more set up for the latter, which is a stable state production. So how do we support innovation, which requires flexibility and agility, right? It’s a pretty big challenge here.

I think finding the way to kind of thread the needle there, where you’re allowing your teams to work within that sixty two, excuse me, twenty six to fifty two week timeline, those are for probably your longest lead items, right? But within that, there’s a lot of hardware development that happens here. And we alluded to this earlier where you need a lot of flexibility, right? So finding the right partners who can enable you to continue your builds to work through, let’s say these parallel or divergent paths that you’re exploring simultaneously is gonna be a big enabling factor for these companies who continue to prototype and advance and develop.

Look, training models is happening all the time. Getting hardware in hand isn’t right. You have a set window to get your build so you can actually deploy these things so you can start collecting data and training the models. So I think it’s finding the right approach that kind of threads that needle that takes those long lead times and allows your teams to be really effective in the interim throughout that.

 

Jim Mayer
Awesome. And so, Ted, you specifically, you you have two unique perspectives in my opinion, and maybe I’m wrong.

But because you’re inside the industry and you’re on the education side over at Festo, you have two different perspectives.

What do you think the the bottleneck is that’s showing up most clearly right now?

 

Ted Rozier
Well, yeah. So when it comes to the education side of the business, And you mentioned in your leading comments about how fast technology is moving.

The one thing we see in education is, so as we’re a system integrator for education, and then we have an industrial side of the business, automation side of the business as well. And so the perspective on the education side, content is king. Okay? So you can imagine we can develop really cool solutions.

Put them in a school and they hit the start button. Well, they’re gonna those solutions will fail. And documentation means everything. Curriculum content means everything.

So not only do you have to deploy and demystify emerging technology. Okay? But you also have to make sure that there is a paper trail and a teacher lesson plan and curriculum to align to it or else it just sits and no one’s learning. So that has to keep up.

Right? On automation industrial side of the business, to Again, documentation suffers.

That’s the one thing that allows someone else to pick up what you’re doing and for them to be great. And no one wants to really do that. Everyone like loves the Okay. Everyone loves the moment. The clean documentation in order to make others great. That that’s that’s that’s that is a bottleneck.

 

Jim Mayer
Got it. Got it. Good information. And, yes, everybody loves the the flash. Right?

Courtney, to you.

Pardon me.

Design decision that works in the prototype prototype stage of product of manufacturing sometimes can create very real problems at the production stage.

What is the kind of choice that seems fine at five at making five units but becomes a massive problem as somebody scales and gets to five hundred?

 

Courtney Fernandez
So it’s fun when we have started with, you know, I don’t wanna say inexperienced or younger engineers are the only ones that do this because I still do this all the time. But you jump onto Digi Key and hunt for specs and buy stuff, that you know is gonna work but you don’t really check, the manufacturer lead time or what’s actually in stock. So it’s really easy to build something that you got in one order, and verify that it works on the bench and then find later that everybody else also bought that thing, because it was actually the only thing. And that’s kind of happening with a lot of supply chain fun things right now is if you’re on DigiKey or Mouser or McMaster, so is everybody else. And they’re not necessarily replenishing as fast as we think they are. And if you’re not checking that manufacturer lead time, they’re like, Oh crap, this has sixteen weeks on it.

And thought it was fine because they had a thousand in stock and suddenly they do not. Like very abruptly they do not. And that bites me still to this day. And even like for people that don’t have, you know, the years of getting bit like this, it’s really easy to look at something and see like a thousand or two thousand in stock. Like, oh, I need to make eight hundred of it, we’ll be fine. But everybody else, if that’s thing, if scarcity is a thing, then you’re gonna run out.

 

Jim Mayer
Yeah. Yeah. Bobby, anything to add to that?

 

Bobby Yousesefi
I think you guys kind of covered it pretty well here. In terms of don’t have any further comments on kind of that particular topic right there. I think as far as bottlenecking is concerned, you know, you do have the challenge posed by COTS components. I think when it comes to custom parts, sometimes what we see on the bottleneck on our side is raw materials, right? So kind of our upstream isn’t necessarily going to the McMasters of the world and buying the hardware that we need to add to these parts. It’s kind of first principles, it’s raw materials, right? So a lot of times what we’re seeing right now is what is our material cost or what is our material lead time as being the constraining factor that’s kind of bottlenecking manufacturing from a first principles basis.

 

Jim Mayer
Yeah. How about you, Ted? Anything to add to either of these points?

 

Ted Rozier
Yeah. Something that jumps out at me.

We have to always look at purchasing components direct or through a distributor. And sometimes you can cut costs going direct, but a distributor sometimes has those components on the shelf too.

 

Bobby Yousesefi
Yeah.

 

Ted Rozier
Okay. So trying to take a shortcut and go direct, sometimes you get caught up. So so so, again, you you you you have to it’s not a one one and done decision. You you have to stay adaptable and constantly watch, you know, your your partnerships.

Yeah. So it’s it’s always evolving. It’s up and down. You know?

 

Jim Mayer
Yeah. And it is wild when you go to, like, the Adidas website just to use something that I just happened to me. Went to the Adidas website, was looking for a pair of shoes. They were stocked out, put it into a Google search.

Amazon had them on the shelf. Right? So same same similar principle. Right? That OEM just doesn’t have it.

Courtney, what is the cost of bringing manufacturability into engineering conversations too late when it comes to robotics and the build out of the hardware? Sorry. I halfway asked that question.

 

Courtney Fernandez
I I think I I understood it. But a lot of times we bring in, like, the actual, manufacturing team that’s gonna use equipment too late in the conversation. And I’ve been a part of that process a few times and seen us bring them in way too late or bring them in early. And if you know what your process looks like, you can repeat it with a robot really well.

If you don’t know what it looks like and you’re just guessing and then you bring in manufacturing later, what you’ve designed could have incorporated their expertise from day one. And there’s little silly things that I’ve seen on the floor that we didn’t know until we actually walked the floor with somebody that worked there every day. And there was a problem with orienting a bottle that they solved by cutting a zip tie and tying it to a pole and a bottle would hit the zip tie and turn. And they had already solved their problem.

And there was a huge design taking place for orienting these bottles. It was gonna cost a lot of money. And if they had talked to the people that use the machine every day, they could have seen like little tiny things like, Oh, there’s a bottleneck down there at two o’clock every day that you just, maybe you don’t know, but they do because they use it every day.

 

Jim Mayer
Yeah. Yeah. Ted, anything to add to that?

 

Ted Rozier
Yeah.

When it comes to overall system integration, there’s so many pieces.

Right now, everything requires visualization.

Everyone wants to collect data.

Okay. And I’ll just speak to one. You know, when it just comes to visualization, visualization is to it’s it’s supposed to make you, faster. Okay.

It’s supposed to be powerful for the operator, the technicians, and leadership. Everyone should be able to use visualization and make great decisions. Well, in order to build out visualization so that it’s actually touching off three of operator technician leadership. You gotta bring them all together.

And and that’s really difficult to do. Okay? So what happens is trying to build things out at the end, you you just don’t get a satisfying solution and cost has just gone up on everything.

 

Bobby Yousesefi
So

 

Ted Rozier
it’s extremely important and vital piece in getting everyone to the party earlier in the process.

 

Jim Mayer
Yeah. Bobby, from your side on on the manufacturing side or your side now on the manufacturing side, what is the most common thing that you see that an engineering or MPI team did not see coming when they hand off from prototype to production?

 

Bobby Yousesefi
You know, I I’ve been thinking about this since, Courtney and Ted started offering their their feedback. And one of the things I’ve seen that happens when you involve a manufacturing team too late is that you’ve moved in from preliminary design review to critical design review. You’ve gotten this part approved. It’s this beautiful five axis part that when you needed quantity five or ten of was no worries.

And then suddenly you’re scaling up to quantity a thousand and you’re looking at that five axis part sideways because that cost is starting to give your supply chain team some heartburn, realizing that this could have been designed at some point during this iteration as a three axis component. And suddenly time goes down, cost goes down, savings increase, deliverables improve. And it’s little adjustments like this, better yet, was thinking of another example. There’s a team I was working with that was using one particular photopolymer during their prototyping process that was working great.

Here’s the thing with some photopolymers, they continue to cure, you know, as they’re exposed to UV rays or environmentally they might get, you know, a little moist or they might get dry and brittle and crack. And what they were noticing was these components were starting to crack once they actually deployed them on their systems and started to, you know, try to get data out of them. So engaging manufacturers earlier on and because, you know, a lot of times no one knows the form fit or function of a part better than the designer. But as Courtney mentioned, sometimes you have younger engineers who are new to this who are just trying to figure this out, and they’re throwing whatever material they know works on their printer at home, and they’re being like, okay.

Let’s scale up with this. Doesn’t necessarily work that way. Bringing in a partner who can help kind of show you that light at the end of the tunnel saying, okay, well, you might want to start with three d printing, but then we have all of these other other processes we could offer, whether it’s urethane casting, whether it’s injection molding. And so having that trusted voice to help guide you on that development path can save a lot of pain downstream.

 

Jim Mayer
Yeah. Makes total sense. Courtney, Ted, what do you think about Bobby’s, answer there? Anything you’d like to add?

 

Courtney Fernandez
I think design for manufacture is a thing that even I kind of learned a little bit late in the game. And I think making design decisions that make the part easier to make for everybody are skipped very frequently. And it’s like even putting a hole for a tool, to go in and work on stuff, saves everybody time and energy. And you you gotta have those, you know. I’m new to the preliminary and critical design review stage. Some companies go really fast and design the machine after they’ve already built it because that’s the documentation order.

But part of the life cycle is super important is redesign this so it’s easier to make a lot of it.

 

Ted Rozier
Yeah.

And I would just add to that by stating that, specifically speaking to robotics, end of arm tooling, it’s so critical.

It’s so critical.

It’s one piece that I think we’ve put a lot of time into the software side. We put a lot of time into how the solution will communicate to things. But the interarm tooling, that is what touches the work piece every single time. And the material, what you use so vital. So, yeah, trying to make sure that you buy out the time to develop that piece and give it its true maturation, you might say. It’s got to mature.

The smartest guy in a process, we build a lot of solutions where pick and place with machine tools. And the smartest guy at the party is the operator. He can hear that machine. He knows what that machine should sound like.

That’s the guy that if he learned to program, he would do best. Okay? And he’d tell you right away where the birds in that star and so much, so much more. So no.

No. It’s very important in looking at the gripper, clearly understanding where it plays its role in the life cycle of the solution.

 

Jim Mayer
Got it.

 

Bobby Yousesefi
Makes total sense.

 

Ted Rozier
Hopefully, didn’t derail the subject there.

That kind of jumped out, but

 

Courtney Fernandez
Hey, Ted.

End of arm tool is the whole reason you have a robot. It’s just there to move that around. Yeah.

 

Jim Mayer
Well and, Ted, what I noticed is when as soon as you mentioned end of arm tooling, both of the other heads on this panel started nodding along. So I don’t think you derailed. I think you added something very valuable to this part of the conversation.

I’ve got another poll for the audience.

Poll number two is where does your team feel the most pressure moving from prototype to production?

We’ll take, you know, another thirty, sixty, ninety seconds to to let answers come in on that. While those answers are coming in, I I wanna talk to all three of you about what engineering and MPI team should be evaluating before they move beyond prototype that most are probably not evaluating today?

Bobby, I’m gonna pick on you. You go first.

 

Bobby Yousesefi
I would say evaluating alternate vendors and parallel pathing or multipathing or dual pathing as early as you can only lends itself to your benefit as a product designer. Right? The more options you have to work with qualified partners who take the time to invest and learn in your process and your product and your part, think it just shores up any potential supply chain gaps that might develop down the line. I mean, look at the end of the day, manufacturing is hard, right?

And I think what we’re trying to do here stateside is a really noteworthy endeavor. So I think reaching out to as many qualified suppliers as you can. Now finding that list of qualified suppliers, finding the right people becomes a challenge in and of itself, right? But I think early on developing these relationships and moving away from a transactional model and truly bringing in qualified partners just helps shore up your supply line for those future sprints, whether you’re just rushing to go from prototyping to production or you are changing a revision mid run and you need to work with someone who really understands what you’re working on your tolerance stack.

So I think taking the time to do that, so many people don’t wanna rock the ship because while the system’s working fine as is, right? I say, don’t wait for it to break down, be proactive, and that’ll set the stage for success in the future.

 

Jim Mayer
Yeah.

Completely agree. And and so does the audience. Forty percent said finding the right manufacturing partner is their biggest pressure.

Also tied with that design decisions that don’t scale and bringing up third place component lead times in sourcing.

But going back to the question, Courtney, Ted, is there anything that either of you would like to add to to Bobby’s answer there?

 

Ted Rozier
I would say, you know, for a while there was a buzzword out there in digital twin simulation. Okay. And we’re all still trying to get there.

I won’t get on my soapbox on that, but what I what I what I will say is, you know, the more you document. Okay? And there are there’s a lot of tools that we can utilize now.

So to simulate things, there are a lot of errors we can catch on the simulation side. So trying to standardize on simulation between the engineering team.

And then you have different sectors And your partners.

Because if you can standardize on all of your simulation, once you push a change, they’re at least starting with the core foundation that measurements are have been qualified already. And now here is where the change is. You know there’s gonna be some errors, but if there’s synergy and everything is you you know, you’re not dealing with paper, but everything is to scale. You’ve got a fighting chance. You know? So I I think that really trying to work with simulation and making sure that you’re you’re dealing with a a true model to scale, this is gonna go a long way.

 

Jim Mayer
How about you, Courtney? Anything to add?

 

Courtney Fernandez
Yeah. I think, simulating and knowing that the the basic hardware is going to work, just proof of concept wise is huge. Cause a lot of times you’re even selecting say a robot arm to run this path through some machine that you have CAD models for. You can see what it’s gonna look like before you ever even commit to purchasing that particular form factor of a robot.

And to Abi’s point, multiple vendors is a huge decision to make early on if you work for a company where you have like an approved vendor list and you’ve gotta go through the paperwork of adding a vendor.

I think I didn’t really understand when I was a little bit younger trying to work for myself why these guys would tell me, I know I wanna go with you but I have to get three bids. I’m like, Why do you have to get three bids? That’s the craziest thing ever. But now I understand like why companies make you get three bids.

Because you’ve got three people in the ABL and you’ve got three bids already done. So if something crashes and burns, the other two are already good to go. It’s a really smart decision to do early on. And young me would hate hearing today me saying that, but it’s a fact of life.

 

Jim Mayer
We get wiser, Courtney, we get wiser. Doesn’t mean we get better, but we get wiser. So with that lead time and and having to to kind of diversify your supplier portfolio, but with the lead time, does that make companies have to make decisions earlier than they may feel ready to make them?

Bobby, you’re nodding, so you go first.

 

Bobby Yousesefi
I’ve actually been thinking about this a lot, and I’ve been thinking about it in the context of the greater American manufacturing context, but also specifically as it relates to robotics and adjacent industry defense. What we’re seeing right now is a ramp up in defense spending. And so you’re going to have a lot of DPAS rated orders flown down to a lot of the existing American manufacturers out there today. So now there’s this complicating factor with, you know, the there’s a lot of impetus being put behind rebuilding America’s arsenal today.

And look, some of the robotics companies out there that we’re kind of talking about are ITAR companies are working in this space, but many of them aren’t. So what does that mean for them practically speaking? Well, it means that look, there’s going to be a strain on capacity that’s kind of being flown down by the federal government, by the defense sector. And so I think robotics companies would be smart to get ahead of this and really start to, again, invest in these relationships, try to pull in as many partners as they can, because I don’t wanna say you’re gonna be competing for capacity, but it’s a reality of where we’re at today in manufacturing.

 

Jim Mayer
Courtney, Ted, anything to add to that?

 

Ted Rozier
Think that something that we’re seeing right now on the educational side of the business, going back to what Bobby mentioned, funding comes in education by way of a grant process.

And the understanding where that funding is going to come from allows you to forecast.

But the thing about grant funding is you can lose it. Okay? There is a stop date on it. Okay?

So that does put the pressure on you. And and that’s the good thing about education. The stress level is, can I deliver by this date? Because that funding, it there there’s no discussion.

It goes away. So you know when to invest ahead of time. Okay?

On the other side of the business, other side of the house, everything doesn’t typically work that way. It just goes up. Okay. The funding might not go away.

It’s just that the cost will start to drive up higher and higher. So back to your to your questions specifically, do they have to make a decision earlier? The the the pressure is there to to to to come up with who you’re gonna work with. And, yes, by all means, you you you have to to to to definitely make a decision a little earlier in the game just just due to the jam and and and suppliers.

But it depends on, you know, what sector of business that that we’re looking at.

 

Jim Mayer
Got it. Courtney, you look like you were about to say something.

 

Courtney Fernandez
Well, because I’m trying to think of how much I really wanna say about money, on that one. But there is this need right now. Like, there’s a speed, to things. And if you’re like if you’re on the smaller business side of and you need to like pounce on stuff right now, you know, to Bobby’s point, there’s a lot of contracts coming up that are gonna put demand on all of these shops doing various pieces of manufacturing. And if you wanna get in there and do stuff, they’re bidding and closing bids like by the time you even entertain the thought. So I think there’s like a lot of pressure to make decisions and buy stuff and get going on projects before you’ve really even finished evaluating. Like I have a long, for myself, a long process for evaluating what I’m gonna do on a project and the bid closed.

You gotta like make a decision, make a bomb and get it out fast.

 

Jim Mayer
Yeah.

 

Courtney Fernandez
I think that’s happening to smaller people a lot right now.

 

Jim Mayer
Yeah, absolutely. I’m just gonna ask this question.

It’s kind of the whole thesis of this conversation, but I haven’t been very direct yet.

Are supplier limitations slowing down commercialization of robotics?

 

Courtney Fernandez
I mean, Yeah. If, just thinking about, being a hobbyist, have you tried yourself, to buy like a Jetson anything?

 

Jim Mayer
Yeah.

 

Courtney Fernandez
It’s hilarious. Yeah. So I don’t wanna say like I’m an innovator, but like smaller companies, you can almost see like just individuals and they’re trying to compete with companies that have huge contracts. Right now, spend lots of money and they’re the ones buying up the thousand of something that you wanna get your hands on. So I think that is limiting what the, not necessarily smaller groups can do, but the more funding limited groups can do with stuff that are actually probably really smart and able to do cool stuff. They just don’t have access to the equipment right now.

 

Ted Rozier
Yeah.

Yeah. You know, the the one thing that I’ll say, you know, we’re used to getting things on consignment.

Okay? And if you can get something on consignment, you know, that allows you to do a lot of your r and d. Okay?

And, you know, you you returned it after six months or so.

Or hopefully, when you get something on consignment, you’ve sold it within that six months. So it allows you to budget. But right now, what we’re seeing is when suppliers don’t have that luxury, you can’t get anything on consignment. So now you’re spending money, which does it it changes the the the whole process.

 

Bobby Yousesefi
Yeah.

I think a lot of this has to do with the state of the supply chain as it is today.

And if you look at it, you know what we have today is a supply chain that’s really catered towards stability. And what we’re seeing now is this surge in new companies trying to innovate, trying to get new technologies, whether it’s humanoid robotics, whether it’s other robotic systems, drone systems out to the market faster than ever before, right? So we’re being asked to do more with less.

I think there’s a couple positive things. There’s this like resurgence in investment in American manufacturing that’s going to open up more capacity, more capabilities. And I don’t even know if it’s necessarily, there have been some capabilities that just like we don’t really have in the States critically currently that we need to reinvest in. But I think there’s something like thirty thousand to fifty thousand American manufacturers like machine shops in the States today, right?

Somewhere around that ballpark. So I think that the state of the industry is slowly shifting where before you’d have a company who’d have this relationship with a client for twenty, thirty years, right? The shop owner was actually the salesperson and had a relationship and has grown this. Now you have a lot of shops who are developing this agility.

They need to keep pace with, you know, these leading robotics companies that are on the bleeding edge or the cutting edge of innovation, right? So they’re getting more responsive, their systems are updating, they’re investing in different technologies, whether it’s automation of their own, right? Using palletizers, you know, doing things within the shop, you know, moving away, let’s say from paper travelers, things like that digitizing to be more efficient and effective. So I do think we’re also seeing a change in the state of the supply chain that ultimately is going lead to really good things.

But, you know, it’s a work in progress.

 

Jim Mayer
Bobby, what what happens when a robotics company engages with a manufacturing partner early in the process, like before a prototype is locked in? What what happens? What changes in that process?

 

Bobby Yousesefi
Yeah. I mean, a a a lot of things can change. I think that there’s an added confidence on their part that when we do push a design forward and we do, you know, because there’s internal stakeholders, right? Everyone is coming at it from a different angle.

They’re all trying to bring these unique perspectives and combine them together to create a new product and bring it to the market, right? I think that, you know, having a manufacturing partner that allows you with a high degree of confidence to say, hey, we think that not only do these parts pass simulation, like all the analysis we’ve run is great, we get the parts in hand, they’re within tolerance, and it actually matches our tolerance stack up. That means that we can deploy that much quicker. That means that we can gather data that much quicker.

Again, lot of these new humanoid systems, for example, know, there isn’t enough data out there yet to actually deploy these in the world and commercialize them per se. Now everyone’s really moving toward that at a rapid clip. How do you keep pace? Getting your parts in quickly, making sure they’re good, not losing time on the QC side, like getting parts back that are bunk because you realize, okay, my build isn’t just like this arm is not gonna be able to extend because this part is not in spec.

These things all compound and eat into your lead time and eat into your competitive market advantage. So the people who have the strongest relationships on the front end, getting stuff made are gonna be able to get stuff to market that much quicker.

 

Jim Mayer
Yeah, absolutely. Courtney, do you think that there’s a difference between having a manufacturing partner and a manufacturing vendor or is that just semantics?

 

Courtney Fernandez
No. I think having a partner that really actually takes part in the ownership of what you’re working on is very different than somebody who just delivers your bomb. So I think there’s a lot of terminology I’ve heard over the years of value added resellers, the distributors, the integrators and then people that like to blur the lines and we give them nicknames like distributors.

And then I had one who referred to his as a disintegrator, which I always found very entertaining.

But having somebody that actually takes part in servicing the machine after you deliver it. Say, you have partners in the areas where you’re actually servicing, maybe they’re not where you’re headquartered. That’s an important relationship to have. You want people in the same time zone when a machine breaks. So that’s a big one, I think when I talk to companies.

And there’s a lot of ways that companies can partner with you versus again, just delivering you your part list.

 

Jim Mayer
Yeah. Absolutely. Ted, anything to add to either of those two points?

 

Ted Rozier
Yeah. Courtney did hit on a value added reseller.

I really feel like having a partner allows an extension to your internal team. Okay?

Having the communication that way and an understanding, a relationship is always going to beat out.

Not having that partner. There’s just the definition of partner.

When you look at, they’re going to have your best interest at hand. You’re going to get the truth out of each other, deal with tough situation. That’s where it matters.

It’s not the easy. It’s not when it’s easy. It’s when things get tough. That’s when you’re gonna see the biggest difference.

 

Jim Mayer
Yeah. Absolutely.

So when you three look at the companies that are successfully making that leap from prototype to production, what are they doing on the manufacturing, the supply chain side that the ones who aren’t able to make that aren’t able to do?

Did that question make any sense?

Did I walk myself into a circle?

 

Courtney Fernandez
No. I think, I’ll take a stab at it, but I think, companies that are doing well are tapping into that, know, we’re geared towards stability right now. So they’re tapping into like existing manufacturing processes to make theirs work. And I can’t remember the name of the company right now but there’s one in China that’s gonna produce a pretty cheap humanoid because they have tapped into their electric vehicle manufacturing network. So they’re buying electric vehicle motors and everything in the supply chain that already exists for their EV market. Plus there’s already facilities manufacturing the EVs that can turn around and also manufacture these robots because they’re using the same parts. So they’re kind of utilizing an existing manufacturing network to make a new part, but not really with new components.

 

Jim Mayer
Is that what we’re seeing with, Elon and and him making the switch of some of his Tesla plants to make humanoids?

 

Courtney Fernandez
I can only theorize, but it would make a lot of sense, to, you know, tap into something you already have, a whole manufacturing system for.

 

Jim Mayer
Yeah.

 

Ted Rozier
One thing I would say to add to that is, can’t do everything.

So as much as, when you think about system integration, there’s a lot going on out there, but it’s important to pick partners that know their wheelhouse.

It’s easy to say yes to things, but you learn real fast that there’s a robotic company that started out and all they did, for example, is three kilogram, five kilogram, ten kilogram. And if you said anything else, they said, absolutely not. It’s not me.

And I thought that they would not be around long, and they are they they they they they knew what they were doing in starting off three, five, ten. Okay.

Yep.

 

Bobby Yousesefi
I think we’re seeing a wave of investment in the robotics space, but I was trying to just take a look, any publicly facing numbers, I was trying to take a look at where this is going. And I don’t think the bulk of it is going towards manufacturing. And in my mind, the companies that are doing the best are actually making that investment now. So again, it’s speed, it’s all about speed, right?

Speed wins. So when you have all of these companies vying for market share in a previously kind of uncreated brand new category, this humanoid space that we’re about to be venturing into, it’s the companies that already have these partnerships in place are working to put them in place that have the leg up, right? I keep thinking back to this is like, how do you unconstrained innovation? And it’s getting people what they need in a quick timeline, right?

They have to be able to try things. Try it, break it, fix it, try again, right? And that cycle can be unforgiving when you have really long lead time items like gears, like specialty actuator housings, like all the electrical components or sensors that go into a build. So the companies that are doing best are taking advantage of that time and they’re doing everything else, right?

It’s the boring stuff, the fasteners, the connectors, the arms, the chassis, so on and so forth. They’re really dialing that down and they’re collecting data faster than anyone else. So they’re training their models. They’re gonna be able to come to market faster.

And I think speed is the name of the game here.

 

Jim Mayer
Yeah. I get that.

Before we get into some audience questions, because I think I’ve seen a couple come in.

One last question for for each of you.

If you could sit down with the director of engineering or NPI program manager at any of these robotics companies that are out there and give them one piece of advice, what would

 

Ted Rozier
it

 

Jim Mayer
be?

Ted, we’ll start with you.

 

Ted Rozier
I would say as an engineering manager, okay, we can hold on to new things way too long and not get the opinion of others.

Being open, being a little more transparent on emerging technology really helps your partners to figure out if they want it.

And you figure out more of their needs because engineers geek out.

They think use a different side of their brain. And sometimes what’ll happen is you’ll think that what you’re working on is the greatest thing since slice of bread, but there’s no need for it. Yeah. Okay. Yeah.

 

Jim Mayer
Absolutely. Courtney, how about you?

 

Courtney Fernandez
Would say

 

Jim Mayer
piece of advice.

 

Courtney Fernandez
Let the let the engineers teach this technology to the people that don’t usually touch it and watch.

A, watch how the engineers think their stuff works.

But also watch how people that aren’t usual users of the technology use it and give you feedback because just unexpected feedback comes from all kinds of places. And really like even having the janitor come and use technology and give you feedback, I think is valuable. Not just the I mean the operator is obviously most important, but bring in everybody to use the technology and give feedback on it.

 

Jim Mayer
Yeah. Bobby.

 

Bobby Yousesefi
I would build off that. I’d literally literally say, like, bring in everybody. And what I mean by that is visit your manufacturing partners. Right? You know, you as the NPI director might be constrained for time, but you have someone on your team.

And not just the usual suspects, right? I’m not talking just pulling someone from supply chain or supplier quality engineer. Send someone from your team who’s responsible engineer, forward deployed engineer, whatever the new nomenclature is for that term, right? But whoever the stakeholder is involved with these parts or subset of parts that you’re getting made, go out and visit the manufacturer, check out the production room floor, see how your parts are being made.

And to your point, Courtney, about these unexpected, you know, conversations or points of feedback, sometimes it’s on the shop floor where you have the master machinist or the GM looking over a project or a part with the customer saying, Hey, by the way, if we adjust this feature right here, we can, you know, decrease our run time or decrease our material costs. I think taking the time to go out and actually have these face to face meetings, meet people where they’re at, but also see how they’re making your stuff. I think that those are huge.

 

Jim Mayer
I lied. I have one more question because it’s my last poll question here.

Great point, Bobby, but I don’t wanna forget this poll question.

What will limit robotic scaling the most over the next five years? And we’ll just do this as a rapid fire answer from the panelists.

We’ve got supply chain and component availability, manufacturing production capacity, AI software and capability, talent and workforce, or cost and capital. Bobby, out of those those answers, what do you think is the biggest limiting factor?

 

Bobby Yousesefi
You know, I’ve been thinking about this a lot. And while I’m inclined to agree that capacity is an issue, I also think the challenge frankly is a workforce issue. Right? A lot of the folks we have, the master machinists, the master welders, any of the skilled trades who’ve been really leading this industry for the last however many years, a lot of them are getting a retirement age, right? And so I think capacity and workforce are actually inextricably linked.

And I also think the companies that will scale the best will form relationships that are strategic and not transactional.

So the companies that are thinking long ball that treat this bottleneck as a vendor problem, I think they’re gonna fall, sorry, the companies that treat this like a vendor problem, they’re not gonna keep up. But the companies that look at this as strategy issue, I think they’re gonna, you know, keep pace with everyone else. So I think it’s tied between production capacity and talent in the workforce, personally speaking.

 

Jim Mayer
Got it. We’ve got a lot of mixed sports metaphors. We’ve got triple threat, and we’ve got long ball from Bobby.

Ted, let’s go to you.

What what do you think the biggest limiting factor is?

 

Ted Rozier
Yeah.

You know, Bob Bobby Bobby hit it. Okay?

And and and because I I play on, within the education space a lot, workforce development is key.

We all understand that the pesky skills gap has been around forever. It continues to get wider. And then what you tie software capability.

So, the software protocols, everybody’s open source.

Man, it’s a job in itself to commission and deploy.

You’ve got a workforce that’s very limited, and now you bring in someone new. They’ve gotta learn a lot. Yeah. Okay.

So a lot of things you can push off the table because of technology right now. There are things we can push off the table. And we can use tools like augmented reality to capture some knowledge. But then there’s a lot of things that using AI, there’s a gray area.

Again, things you can take off the table, but you still have to have clear competency. You have to understand how to do things and have courage.

And all of that falls back to the workforce. They’re just not there. The training is not in place. It’s extremely important right now.

 

Jim Mayer
Yeah. Courtney, what about you?

 

Courtney Fernandez
I mean, I don’t wanna be a third one to chime in with the same answer, but, yeah, I think the talent and workforce is gonna be the biggest bottleneck. And it’s something that we talk about like on our podcast a lot is how many people found manufacturing by accident. And like our nerds are all about how everybody wandered in here by accident. And it should be more like on purpose, where people should be in manufacturing on purpose. But all I saw when I was graduating in Silicon Valley was Google, Facebook, work for these companies and I think it’s only gonna get worse right now with everybody going, oh, I’m gonna work for Anthropic, I’m gonna work for OpenAI.

And just not knowing that this career path even exists in manufacturing because there’s really bright people that can, you know, do innovative, cool stuff in manufacturing if they knew they even could do it here.

So I

 

Bobby Yousesefi
don’t know

 

Courtney Fernandez
what the solution to that is, but we needed more people to know that it’s even a career path in the first place.

 

Jim Mayer
Yeah, absolutely. And this is why I love hosting these conversations because we started at one point. We are finishing on a completely other topic, and that’s fantastic.

So we are we’ve got two minutes left, which means we’re going to go and answer just one question.

My my friends from Fathom are gonna put up on screen. We got one in.

Five years from now, what does the robotics manufacturing landscape look like if the industry gets this right? So what’s that that forward looking view to to you all?

 

Courtney Fernandez
An actual digital twin, like like if we can actually simulate our our entire factories before we actually commit to making decisions.

So I think we talk about it a lot right now, but it’s really actually hard

 

Jim Mayer
Yeah.

 

Courtney Fernandez
To simulate all the factors that go into a plant.

 

Jim Mayer
Sure. Makes total sense. Ted, what about you?

 

Ted Rozier
Yeah.

I definitely think that we’re at an inflection point right now when it comes to robotics.

Humanoids right now, there’s a huge push.

And I think that there’s a lot of things that we can speak to, to say that humanoids might not, while they won’t work. Okay?

With safety protocols, I definitely feel like in five years, we’re going to see this space where you’re going to see more humanoids definitely.

There’s a whole section that’s going to be committed to just humanoids. That tells you a whole lot that you have that many suppliers right now, manufacturers that are trying to bring these to fruition. But then the one thing that drives it all is information data. And from the simulation side of things, to just speak to what Courtney mentioned.

We have been able to collect a lot of information and we’re learning what to do with it.

Data was hitting a black hole for many, many, many, many years in classification. It touches on AI prompt engineering. These things are starting to move forward and you’re gonna see a lot more on the simulation and digital twinning side of things.

 

Jim Mayer
Yeah, absolutely. Bobby?

 

Bobby Yousesefi
With respect to the software component, so I’m gonna focus more on the physical part of physical AI here. I think Courtney and Ted really covered the software implications pretty well here. I think that, look, we’re building a new industry, right? This humanoid space hasn’t existed before.

And similarly, the supply chain for it doesn’t exist. It’s currently being built now, right? And so I think in the next five years, as we see the advancement of humanoid robotics companies, we’re probably gonna see a lot of investments in rare earth metals, rare earth materials, the magnets, the motors, those kind of things. We’re probably going to see some more investment in some of these longer lead time manufacturing capabilities, casting, so on and so forth, things we don’t have.

And I think we’re going to see a burgeoning of American manufacturing in general. And certainly humanoid robotics is going to be part of this wave, right? There’s a lot of companies who are focused on building domestically. There’s a lot of applications that we can support here right now in the US.

And so look, I think again, American manufacturing is getting a lot of attention, a lot of investment right now. And I think we’ve already hit the ground running and that pace is going to continue in the next five years. So it’s going to be more and it’s going to be faster. And the companies that make these adjustments early on are, I think, gonna be the ones who will have the longest lead.

 

Jim Mayer
Yeah. Awesome.

Real quick, before I cut you three loose.

Thank you. Ted, Courtney, Bobby, this was a a really good conversation. I hope you liked it as much as I did, because I I really enjoyed being the guy who just asked you questions and learned a whole lot about this industry. So thank you three very much for being on the panel today.

 

Courtney Fernandez
Thanks for having us.

 

Bobby Yousesefi
You’re welcome. My pleasure.

 

Jim Mayer
And thank you to Fathom for hosting these.

Ted brought up Automate. Next week is Automate.

It’ll be a great show.

And Fathom is hosting an event out at their facility just outside of the city in the Chicagoland area. If you’re interested, please reach out to me or or somebody from Fathom for more information about that. It’s Wednesday night, the twenty fourth, I think that is. So go check it out if if you’re at Automate.

Again, thank you all very much for being here. Audience, I can’t see your reaction, but I hope this was a valuable use of your time. Thanks for being here.