Whether you’re programming robots, working in a clean room environment or a dirty steel mill, people working in manufacturing jobs can make a fantastic living and should be proud.

Industrial cables are the silent technology that keeps the PLC connected to the HMI and the robot connected to the controller. In some ways, the only time that people think of cables is when there is a problem with one, according to Jay Swank, Business Development Manager – North America, Lynxeo. In this episode of Manufacturing Matters, Swank joined TECH B2B Marketing’s Jimmy Carroll to talk about a range of different topics, including attracting more talent to manufacturing jobs today – despite the proliferation of automation technologies – as well as the latest trends in robots, AI, and humanoids.

Jimmy Carrol: [00:00:06] Hi everybody. My name is Jimmy Carroll. I’m the vice president of operations at Tech B2B Marketing, and I’m here on a special episode of the Manufacturing Matters podcast at the A3 Business Forum. I have the pleasure of being joined by Jay Swank from Lynxeo. Jay, thanks so much for taking the time. Really appreciate it.

 

Jay Swank: [00:00:21] Yeah, this is exciting. I’m happy to be here.

 

Jimmy Carrol: [00:00:23] Yeah. So I want to say that Jay is the first example of somebody that, sometimes at the A3 Business Forum you have to go to the back side of the bar to be able to get a drink. And it just so happened that we just kind of instantly started talking. We said, “Hey, this has been a really interesting conversation. Let’s bring it onto the podcast.” So, glad you’re here.

 

Jay Swank: [00:00:39] Yeah, that’s exactly how it happened. I still am often more an engineer than a sales-minded person, and there was just too many people on the front side of the bar. I needed to find a little reprieve.

 

Jimmy Carrol: [00:00:49] Absolutely. Yeah. All right, so Jay, for those who don’t know, tell us a little bit about Lynxeo and what you guys do there and what you’re excited about right now.

 

Jay Swank: [00:00:57] Sure. Lynxeo was very recently a company called Nexans, which is a large cable manufacturer. Lynxeo was a carve-out from that. And what really is exciting with that carve-out and now standalone business is that we are hyper-focused on an industry that we’ve been in a long time for industrial automation. Really the CliffsNotes version would be that if you need a cable that is going to bend, twist, anything that you wouldn’t think would be a good environment for a cable, well, we make the things that the cable product will then excel in that environment. So it’s not that fun really of an industry compared to many of our components. But we’re the silent little thing that keeps the PLCs to the HMIs, the robot to the controller. We keep that properly functioning. And really, you only ever think of us when there’s a problem or a fork truck has driven over a cable.

 

Jimmy Carrol: [00:01:50] Yeah. Fair enough. And again, we talked about that a little bit yesterday. And it’s true. Robots kind of steal the attention, even away from sometimes the AI or the machine vision or whatever. But there’s also cables, there’s optics, there’s lighting, lenses, all that. And it’s all equally important when you’re trying to accomplish certain goals. So, on to some of the fun things: What are some of the applications in robotics and industrial automation where evolving needs are pushing product development on your end? What are some of the things that have caused your end users to say, we need these cables to be able to do x, y, z?

 

Jay Swank: [00:02:24] Yeah, if I was to pick one single category, we see an increasing demand for hybrid types of technology. So as we have moved away from your traditional control signals and cables that would carry power in some level of current, we moved into many different types of field buses, some proprietary, some just your normal industrial Ethernet, but our devices still use quite a bit more current. So adding really high-frequency rates for a Cat 6–plus type of communication but then pairing that with what’s traditionally noisy power signals, square wave AC for a servo motor or for some other device, and all of the different variations of that. Whether someone has a five-axis arm, a three-axis arm, they also are adding a vision system. Really what we see is more and more customers that are looking for a very dynamic application still but for all sorts of different conductors into a cable.

 

Jimmy Carrol: [00:03:16] Now what are some examples of the industries where you’re doing that and maybe some application types. Not just limited to cobots, obviously, but I’m picturing a lot of cobot applications where the robot arms are making a lot of different types of movements, and like you’re saying, there’s twisting and just a lot of force being put on these cables. What are some primary or notable examples of those types?

 

Jay Swank: [00:03:37] Yeah. You know, anytime you see, whether it be in the distribution space or manufacturing, where you see a cat track, it’s usually our type of products that will be in those cables. Again, what they power: the sky really is the limit. I find the cable industry, I am somewhat new to it, to really still be very reactive to the overall market. The technology is limited, and whether it can be done, we come in there to say now we can produce that. So really the hybrid technology for servo motors, many manufacturers today no longer have a power and separate encoder cable. They’re putting those in one thing. That was a major development for us. But as far as where you would see it in the factory, if something’s moving, there’s really only three ways in the factory to make something move. Two of them are fluid power. We’ve got hydraulic; we’ve got pneumatic. Both of those are cheap ways of powering. Everything else is electric. And there’s a cable there.

 

Jimmy Carrol: [00:04:31] Okay. Yeah. So I want to ask about, again, a lot of times I might keep it a little bit focused towards cables and the different applications that we’ve already touched on. But given the conversations we’ve already had, I just want to talk about some overall trends and different things in the industry. So one is the labor shortage, and this issue persists. And you heard it this morning in the 2025 Global Economic Update from Alan Beaulieu, which is tremendous as always. The next five years will be good. There are tons of open jobs still, and there aren’t enough millennials to fill the gap for boomers and all that stuff. So how can companies today attract young people into manufacturing careers?

 

Jay Swank: [00:05:11] Yeah, I think you said a key word: “attract.” I have two sons that are older. And as they were going through school, I had a fantastic career. I’m very proud of the fact that I went from a maintenance technician into then being an engineer to clean sheet design before I moved into business development and sales. I’ve worn a lot of hats, and I always found it to be a very fun industry to work in, but I feel like most of our youth sees the factory, certainly in that millennial range, they see the factory as a remedial place to go. No matter how neat the job is. Whether they’re programming robots, whether it’s a clean room environment, or whether it’s a dirty steel mill, they feel like they didn’t succeed when they do it. And I’m a huge fan of Mike Rowe and the “Dirty Jobs” and some of the philosophies that he puts into it: “You can make a fantastic living and you should be proud.” And I see this today. Lynxeo is a global company. So I work with folks that are in Germany and Italy, and the folks that are in the factory, they carry such pride for what they do. They’re proud to tell you what their job is. They’re proud to talk about how they work in a factory and what that looks like.

 

Jay Swank: [00:06:19] And I don’t see even in the highly technical jobs, folks sort of are under their breath, and they very quickly get to sort of glamorize the neat aspects without being very real about it. And I think this is the first step. It starts back at education. It starts back with our parents. We really need to educate. And I think it does come down to the companies themselves. They have the responsibility to get active in the community and really showcase their factory. Show what it’s all about, show the technology. There are companies that are doing a very good job of that. You know, we have a partner, if you visit their factory, their large facility in Providence, Rhode Island, they’re all the time in the local newspaper. They’re established in their community. They invite the community in. So as you start to see the local students say, I would like to work for that company. I like the ethics, I like the culture, and it’s a place that I want to go. So I think first we need to make it attractive, to come full circle back to that and not make people feel like they haven’t succeeded when they move into the space.

 

Jimmy Carrol: [00:07:18] It’s jobs like doctor, lawyer, astronaut, and things like that are romanticized and duly so. And by the way, it helps that they exist in a town of their own, given how large they are. But anyway, when it comes to jobs, and you mentioned it, you touched on it a little bit, but do you think that the ability, and maybe it’s the younger generations below millennial, but do you think that the ability to work alongside the robots will do any benefit?

 

Jay Swank: [00:07:55] I do. I think the more that happens, although I think that there could be some pitfalls to overall acceptance in some of our robot technology. So I’ll circle back to that. But I do think the more the robotics are present in the factory and the more that becomes the job rather than putting lug nuts on a wheel (this really doesn’t happen now. This has been automated for some time, but it’s an analogy that many folks use), I think it does make it much more of an interesting job for folks to say, “I work with robots every day.” Now we have some risks with technology now as well. You know, I think technology leaps forward when there’s really good social acceptance of technology. And today, and topics on the show today have been have been around humanoid robotics and whether that’s good or whether that’s bad. I personally feel like the more we are trying to turn our robotic methods into a human, the more we will pull people back from that social acceptance and make them then really want to adopt that technology. It’s threatening to the existing factory worker that is there. So as a programmer, we would love to give them personalities and names. You know, that’s Thelma and Louise. Obviously, these are common pairings for names, right? But I think that comes with some caution because that suddenly really does become Thelma and Louise. And then we see vandalism and other things as folks try to protect their jobs. So yes, but also with caution.

 

Jimmy Carrol: [00:09:18] Yeah, yeah. Fair enough. And I’m definitely going to circle back to that humanoid question because that was very interesting. And we did see a discussion earlier today, this morning at the forum, that was interesting. I will give a quick shout-out to organizations like A3 and the ARM Institute and some other folks that I’ve talked to this week that are also trying to help bring awareness, make young people aware of super-rewarding careers that you can be proud of. So anyway, buzzwords are everywhere: “humanoid,” “AI,” “deep learning,” “3D” — it goes on and on. One of them that gets put out there a lot is “ease of use,” “usability.” And obviously a lot of that’s happening on the software side. But you guys are a big company and you’re global. So what are some ways that you are seeing companies today lowering the barrier of adoption to robots and automation?

 

Jay Swank: [00:10:16] I think most companies don’t necessarily have the barrier. It just comes down to cash flow, time-to-market with products, and things like that become other factors for whether you have a human operator or whether you have a robot or in most cases that I see, some combination thereof. We now I think have a primary focus where we put robots and we put automation in a space that is highly repetitive and not a position that an operator would want to be in, meanwhile allowing those more human skills to be focused in different areas of the factory. For me personally, I don’t believe it has taken jobs. We hear this topic a lot, and it’s obviously one when a factory is talking about adding automation. It’s something I’ve had to deal with a lot is, “Well, you’re stealing jobs. You’re taking jobs.” I don’t believe that to be true. However, it does change the dynamic of the jobs in the facility. We are replacing general laborer type of positions for higher-skilled positions, and not every factory worker in the United States is interested in a high-skilled job. Some people do just want to come in, punch a clock, do a good day’s work, punch the clock again, and leave.

 

Jay Swank: [00:11:25] So I think it comes down to each individual situation. But, again, I think it comes down to more of budgeting on acceptance. For us, global, we obviously look at new machines. We always want to invest in technology that’s going to help us with our products. I find for me personally, some of the AI tools that we have today with the language models, they benefit me. My team members are spread out across the globe. I interface with folks in Italy. I interface with folks in our headquarters in France and another factory in Germany on a regular basis. You know, I’m from Ohio. I sometimes get a Kentucky drawl. I sometimes get a little New England speed about my delivery, but that also comes along with jargons and different deliveries. So honestly, most of my emails I pass through ChatGPT or some language model, where part of my instruction is to use vocabulary that then is accepted. I haven’t gone so far yet to say, “Translate it for me,” because then I can’t proof it. And I don’t think we’re there yet with technology.

 

Jimmy Carrol: [00:12:25] Yeah. Well, it’s almost inevitable that I was going to ask you about AI, but since you brought it up, I will ask. And again, we saw a presentation earlier that we spoke about briefly and it talked about AI. AI has been around for a long time. But in the last seven or eight-ish years, it’s become a huge talking point. And a while back it was maybe being overhyped a bit. I think it was maybe in 2016, 2017, maybe around Automate or the Vision Show in Stuttgart, and a number of companies come out with deep learning tools, and “These deep learning tools are going to change everything,” and they haven’t. But AI is adding real value in different ways. What are some of your thoughts on the way that AI is adding value today?

 

Jay Swank: [00:13:11] Yeah, I’ll build a bit of background here. A good portion of the presentation that this conversation came up about really revolved about the realism and some of the gaps and the hype that comes around, where we’ve been at these phases before. In fact, there was some timing shared that some of the things we talk about that are new today were from the ’20s and the ’50s and the ’70s. But there are differences in some of these tools as well. I am on the hype train to the large language models. I think it’s going to be really good for programming of PLCs, even robotics, I think where we are in the short term is for someone to not necessarily, I may speak my age here, but back in the CNC days, where you needed to know G-code to move everything around, and now it’s a much more user-friendly language. I think the same thing is upon us for robotics, where users can manipulate the arm where they want it to go and speak, “Close gripper. Open gripper,” and program in a new way and really bridge the gap between the languages that take so long for someone to work and really allow them to focus on what they want that robot to do. The other key, the pivot point here, and why I’m on the train of I think it’s here for us, particularly with AI, is social acceptance.

 

Jay Swank: [00:14:28] We now have a large body of people that are interested in using language tools, and that’s different from what AI has been in the past. It was just two years ago there was a presentation that showed us how AI learns, but that presentation was largely based on how it learns for itself, and in this case it looked like Pong, the old Atari game, but basically it was teaching itself how to play soccer. And then it was rewarded. I don’t know what a reward looks like to AI, but it was rewarded for when it got a goal. And then they added team members, and it learned all on its own. The difference today, there are so many people using these large language models, we are teaching, we have added another layer of human approach to it. We correct it, and it’s learning not only on its own but then how we want it to respond. So part of the conversation was that this exponential growth isn’t real. I have a hard time accepting that because humans are actually what causes it to taper off, and I would like to think that we are not the limiting factor to our technology. I would like to think that we will continue to bring ingenuity and new ideas and really bring it to that next level.

 

Jimmy Carrol: [00:15:39] Yeah, totally. I mean, my brain often goes to visual inspection with AI, but to your point, there’s just different types of AI on the factory floor and well beyond the factory floor. But no matter what form of AI you’re talking about, better data is better results. I know it’s cliche, but garbage in, garbage out. And so in visual inspection, for example — there’s that New England speed of mine. I apologize — you have to train your model based on lots of data, and you have to label defects and come to agreement on what’s a defect, what’s not a defect. And then the model can start to inspect things and make subjective decisions on whatever it may be — organic matter.

 

Jay Swank: [00:16:19] But even in that scenario, today is different again. For a factory to want to adopt this technology. We’ve had global events, whatever it is, there’s been enough global events where factories have learned that they need to diversify where their manufacturing is held throughout the globe. But if you have a common process, say in Germany and one in Asia and one in the United States, how awesome is it that your visual inspection system is not only learning at that one point, but the data collected there can be shared to also accelerate that learning on those other locations? That’s a different level, and that’s a different drive to use that technology than really what we’ve experienced. And then you come back into security. Well, I don’t know that I want those images. And how do you handle bandwidth? It’s all valid. And that’s where the comment today that technology is adopted with an S curve, not exponential — I agree, and there’s certainly going to be some holdups. But, boy, I think we’re far from seeing a slow in this.

 

Jimmy Carrol: [00:17:19] Yeah, and that kind of goes back to that point we were talking about earlier a little bit, right? If you take somebody who’s been at a job for a long time in a steel factory or something, and all of a sudden they bring in AI. Well, these guys, they’re calling them boomers, toward retirement age, from that generation, but they can now help program AI models without realizing they’re doing it by labeling what a defect is in these images on the steel, or whatever it may be, welding, or something. And being able to tell a young person, well, you’ll be able to help train these AI models, you’ll be hands-on with AI. That’s got to be at least somewhat attractive to them.

 

Jay Swank: [00:18:01] And there’s good examples of that. You know, largely the the automatic driving by Tesla cars has come from humans still teaching it that it made a mistake. For everyone that has a Tesla car that’s driving around, whenever the human jumps in and makes a correction, Tesla records that, and it evaluates: “Why did you decide to do something different than AI thought it needed to do?” Well, it’s really the same thing you’re talking about. You’re talking about having AI present while we still have that 50-year expertise in a factory to say, “No, no, no. That wasn’t right. You need to do this.” It’s really a very opportunistic time to capitalize on these really senior positions that have been there.

 

Jimmy Carrol: [00:18:41] Yeah, that’s a really good point. So I want to go back to that humanoid point. I’m cheating here. I will admit that Jay and I did talk about this a little bit off-line, but it was really interesting because I think humanoids are obviously very interesting. I guess I’ll start by asking you kind of a reset question: What are your thoughts on the short-term and long-term applicability and usefulness of humanoids and how do you see that playing out?

 

Jay Swank: [00:19:09] Yeah, I think today we’re still in a really highly romanticized approach, where when you look at most of the humanoid robots, we are trying to make them truly humanoid. And again, to circle back on earlier comments, I personally feel like this could be detrimental to the progress overall with adoption at the plant floor level. Again, I can see operators that would be intimidated by a robot to show up that looks like a human, especially when we go to name them. But I do think there’s tremendous opportunities to displace human-type processes. And if we were to focus not so much on the looks but just on dexterity of a hand and an operator position to enable them for when someone’s sick or to just be a third hand for some delicate processes, I think there’s a tremendous opportunity in that space. I think it could be one of the biggest growth areas. You know, I’m not so concerned if it looks like a concrete block, but if it’s able to manipulate the parts in a similar manner to an operator, well then we may just may look for an x, y, theta sort of positioning or something else to move that robot in and out. The other advantage here is that it does play into some safety aspects that are already proven with collaborative arms, whereas today and as spoken, I think accurately, robots that are standing and walking, we’re still a ways from that being what we would consider safe. These robots are heavy, and heaven forbid they do lose their balance. Now we’ve added a safety risk for that operator that could be present. So I think it’s a step back there. But again, if we pivot now, if we move away from the romanticized approach of making a robot human and just say, “Give me really good hands,” I think we would see applications be adopted very, very quickly. The factories today are still really designed for a human. So if we maybe just went waist-up with some regular-looking arms, I think we could see it go.

 

Jimmy Carrol: [00:21:03] So I guess one of the counterpoints often to the usefulness and practicality of humanoids is: Why not just use a cobot arm on an AMR? It’s kind of like what you’re saying but not quite. You’re saying that it could be two arms with fingers and a vision system in the middle. Is that what you’re picturing?

 

Jay Swank: [00:21:24] Yes. Again, I think there’s a lot of factories for a lot of reasons that would prefer to still leave a human operator in place. But as we talked on earlier, there are some staffing shortages. There’s other dynamics to keeping the factory going. So having perhaps one robot — that is a less investment rather than replacing an entire line of maybe five robots — that just replaces the sixth person occasionally in multiple stations, is really more where my mind is on a slower adoption than saying, “We need to completely automate this line and eliminate those five operators altogether.” I don’t think that’s where anyone wants to be, including business owners. I think there’s many business owners that take pride in the staff and giving to community and the number of jobs that they create in a role. That being said, we also, especially through the supply chain shortage, saw factories with COVID that needed operators and then they weren’t there. So to me, that’s really more of what I speak about, having some device that could replace five operators not all at once but as needed. And then perform what would be otherwise highly, highly skilled smaller-type operations.

 

Jimmy Carrol: [00:22:31] Yeah. And it just kind of circles back to one of the things I was talking about earlier with buzzwords. Another buzzword that I hear and frankly talk about a lot is the idea of flexible automation. And that’s a great cutting-edge, innovative example of what a flexible automation system could be like, something you could repurpose, move as needed, handle different part types, handle different orientations. So that’s fascinating.

 

Jay Swank: [00:22:58] Yeah. The robot, once it’s trained to that position, it doesn’t have a limit. Its memory — this is the cheap part of our technology today. So you could over time teach that humanoid station to perhaps do 30 things. And once you’ve taught it once, it’s good. Where again, with the human, we do tend to get hyper-focused with skills in one area. And if they haven’t run Line 6 in 12 months, then they have to be retrained for that position, whereas the robot will be as good six weeks later, 12 weeks later, 12 years later as it ever was on that first day.

 

Jimmy Carrol: [00:23:33] Yeah. Or better probably, because if it’s using AI, it’s going to be learning over time.

 

Jay Swank: [00:23:38] I think fair point.

 

Jimmy Carrol: [00:23:38] Yeah. Jay, I’m going to venture a guess that — sometimes I ask this question and some people say, “Well I don’t have a crystal ball. I’m not sure,” but do you have any fun predictions or any general predictions for the next couple of years or even out further in terms of where you see automation evolving or growing into?

 

Jay Swank: [00:23:58] Yeah. I don’t know about a crystal ball. It’s clear to me on any economic forecast on anything, automation is here. Automation is being every day more accepted for a lot of different reasons. And I think we’re going to continue to see this and reshoring of manufacturing, particularly in the U.S. That’s where we are today. I think we’re going to see a lot of activity there. I think we’re in for a good couple of years. Where the technology goes? Honestly, again, I think we’re still just sort of figuring it out. I think many people — I forget the number of folks here, close to 2,000 folks here. And I would venture to say, as I’ve spoken with some, that about half really haven’t even still yet given ChatGPT a try.

 

Jimmy Carrol: [00:24:42] Yeah.

 

Jay Swank: [00:24:43] So I think as they understand more, ideas continue to grow. I think as the millennials and younger folks are coming into the factory, they have a good voice behind them. They have some creative thoughts. I think where we sit year by year will be drastically different.

 

Jimmy Carrol: [00:25:00] Yeah, and a couple of things earlier that kind of back up what you’re saying is, one is Milton Guerry, who was the president of SCHUNK, the outgoing chairman of the board for A3, said, “We’re only getting started. We’re only getting started here.” Not literally — automation’s been around. Then in another one, in a presentation later, the gentleman said that only 20% of warehouses today are being automated. And you think of warehouses today, and all I’m picturing is all different types of automation. But that’s just not the case. I mean, yes, the larger places are using it, but there’s a lot of room for growth. And it’s a very exciting time to be a part of all of it.

 

Jay Swank: [00:25:43] Agreed.

 

Jimmy Carrol: [00:25:44] Well, if folks want to learn more about the company, its lynxeogroup.com. Is that right?

 

Jay Swank: [00:25:49] That is correct. Imagine the cat, the lynx. You know, it’s dynamic, it’s nimble, and it’s sort of hidden. You never see it.

 

Jimmy Carrol: [00:26:01] Okay.

 

Jay Swank: [00:26:01] And then add an EO.

 

Jimmy Carrol: [00:26:02] All right. Perfect. Well, Jay, thanks so much for taking the time. I really appreciate it. If anybody has any questions or comments for us or for Jay, I’d be happy to pass those along or field them myself obviously. And you could reach out to us at manufacturing-matters.com. And thanks for tuning in.