Vision Podcast #23– Machine Vision Renaissance w/ Cognex

MVPro Media – The Vision Podcast #23

Guest – Matt Moschner, President and CEO, Cognex Corporation
Guest – Brian Benoit, Director of Advanced Vision Products, Cognex Corporation

Matt Moschner, President and CEO of Cognex, and Brian Benoit, Director of Advanced Vision Products, join Josh Eastburn to discuss what Cognex describes as a new machine vision renaissance: one shaped not only by AI capability, but by usability, faster deployment, edge-based tools, better image formation and lower barriers to adoption on the factory floor.

---

On this page:

• Podcast player
• Guest information
• Useful links
• Episode chapters
• Episode transcript

---

Listen to the Episode:

---

About our Guests:

---

Useful Links:

• Matt Moschner LinkedIn profile
• Brian Benoit Linkedin profile
• Cognex Corporation website | LinkedIn page

Matt Moschner will be at the Automate 2026 Executive Roundtable: “State of the Industry” on June 22, 2026: Explore the agenda.

---

Episode Chapters

Click onto the chapters to access the relevant sections of the transcript below

1. From Engineering Roots to Vision Leadership
Matt explains how his background in product and engineering shapes his approach to leading Cognex, staying close to technology while focusing on delivering real customer value.

---

2. Why This AI Wave Is Different
After decades of technology cycles, Matt explains why AI stands apart, not just for new capabilities, but for making machine vision significantly more accessible and easier to deploy.

---

3. From Months to Minutes: The Shift in Deployment
The discussion moves into how AI is collapsing deployment timelines, reducing complexity, and allowing engineers to solve applications dramatically faster with far less experience.

---

4. Scaling Vision Across the Factory Floor
Brian shares how lower risk and easier maintenance are enabling manufacturers to expand machine vision beyond initial use cases, adding inspections where they were previously not viable.

---

5. OneVision, Edge AI and What Comes Next
The conversation closes with how Cognex is combining edge deployment with cloud-assisted development, improving collaboration, and looking ahead to no-learning AI, better workflows, and more automated vision systems.

---

Episode Transcript

Matt Moschner – Cognex

When we started in AI vision, weeks, months, thousands of images, our best people to solve a problem. We’re now putting a new engineer in an hour meeting to solve a problem, collecting a few samples of a few parts. Amazing. That is to me the shocking thing. And as much as we’re leaning into solving the problems we’ve always wanted to solve, we’re actually probably more focusing on the problems we can solve today, but solving them 95% faster with virtually no experience required.

Josh Eastburn

Welcome to the MV Pro Podcast. If you spent any time around machine vision, you know that the gap between what AI demos look like and what actually ships on a production line is real. My guests today have spent the last several years living inside that gap, and now one of them is running the company. Matt Moschner is the president and CEO of Cognex Corporation. He came up through product and engineering at Cognex. Electrical engineering background from Duke, MBA from Kellogg, time at Boston Consulting Group and Boeing before that, and was named CEO in June of 2025. Joining him is Brian Benoit, Director of Advanced Vision Products, who leads vision software OneVision and LOCA at Cognex. Brian’s background is worth noting. He started as a research scientist in biotech and biopharma, moving into product management through robotics, including time at Rethink Robotics, and has been at Cognex for 9 years. The path from lab bench to vision software product owner gives him an unusual perspective on what it actually takes to make complex technology usable in the field. Cognex is the world’s leading provider of machine vision for industrial automation with over $900 million revenue, 30,000+ customers worldwide, and more than 1,400 issued and pending patents.

Josh Eastburn

They’re also celebrating their 45th year in business, which means they’ve watched every technology wave come through this industry. Today we’re talking about the current big one, but I wanted to get specific. We’ve heard the story of why AI is great for vision. I wanted to know how Cognex was actually putting it to work and how those tools are changing what’s possible on a production line. Please enjoy my conversation with Matt Moschner and Brian Benoit.

Josh Eastburn

So Matt, you came up through product and engineering before moving into your CEO role, which was just last June. Is that right?

Matt Moschner – Cognex

Yeah, that’s exactly right. Yeah. And I’ve been with Cognex exactly 9 years. Funny story, Brian and I started on the same day at Cognex. We were in the same orientation. Oh, wow. My background is electrical computer engineering. I was a double major in economics. And that’s, that’s been a through line in my career where I’ve always tried to find roles where you combine cutting-edge technology with practical applications that deliver value. And that’s very much what we try to do at Cognex, and that’s what I’ve enjoyed most. But yeah, as you said, rose up through product, uh, engineering, and trying to bring, you know, that perspective to this role, which I stepped into the role of CEO in effectively July 1st of last year.

Josh Eastburn

Yeah, that’s kind of where I was going with that too, is like, how has that background do you feel shaped the way that you’re looking at your role now?

Matt Moschner – Cognex

Yeah, definitely. I mean, I’m, I’m still very involved, I would say, in our product roadmap. Uh, as Brian knows, we get together almost daily where I have a new wacky idea for him, but I chair our product approval committee. You know, the three engineering leaders we have report through me. And so I try to stay very close to the technology trends that are happening around the world, but specific to our industry. And gosh, is it changing quickly. And so I think that’s very much an area for me. And, you know, I still try to use our products on a regular basis and make sure that I can try to put myself in the shoes of our customers to see how are we delivering value and how do we need to change and evolve to do that even more easily in the future.

Josh Eastburn

Yeah, awesome. I think that’s important for a tech company, right? You’ve got technical people in those roles. Brian, your path into machine vision has not been the typical one. Is that fair to say?

Brian Benoit – Cognex

That is very fair to say. Yeah, I started my career as a bench scientist in the thriving biotech industry out of Boston. But interestingly, pretty quickly in my career, I gravitated away from the lab and more toward kind of developing kind of benchtop machines and that sort of like interplay between software and hardware was some, uh, an area pretty quickly that I focused a lot of time on, you know, kind of developing UIs, taking really complex, you know, bioassays and trying to figure out how to present those to users in a very kind of clean and elegant way. Right. And so naturally from there, you know, I went to business school and then when I got out of there, I didn’t want to get in the lab anymore. And my first job after business school, I was immediately pulled back in, said, I got to get out of biotech or else I’m going to continue to be using pipettes every day. So got a job at Rethink Robotics, my first experience within the FA space, and just fascinating, like, just incredible amount of innovation going on in that area, people trying to automate as much as they could.

Brian Benoit – Cognex

And lucky for me, there was a Cognex camera embedded inside the arm, and I got to know a lot of Cognex folks. Got kind of obsessed with machine vision. And here I am after a few years at Rethink. As Matt said, him and I started 9 years ago, and it’s been a pretty incredible journey to kind of learn more and more about vision, how our customers are deploying, but still that interface between hardware and software, that’s where I live. And all of our products, we have to have a very— it’s incredibly important for us to think through both of those sides, right? You got to really think about how users are using it day to day, but also your image formation. If you can’t get that right, you don’t have a product. That’s where I spent a lot of my time.

Josh Eastburn

So I’m curious, how was Cognex showing up in the work that you were doing in the robotics space?

Brian Benoit – Cognex

So it was more pick and place. You know, you have a part coming down a conveyor, identifying a pattern, X, Y, theta, and then in grasping it, right? And so what was really cool about it, one of the, most tricky parts of, you know, robotics is, you know, the hand-eye calibration. Once you have the camera mounted inside the arm, you don’t have to worry about that anymore, right? And so we were at Rethink, we were developing pick and place applications where the, you know, the image was front and center within your GUI. But of course that was, we were just using pattern tools and that’s not enough for most robotic applications. You need much more access and that’s, you know, I think, as we’ve gotten more sophisticated, we’ve understood to expose those tools is one of the critical pieces when it comes to going after robotics applications.

Josh Eastburn

Continuing with that theme a little bit, actually, Matt, Cognex just celebrated 45 years as a company. So that’s a company that’s seen a lot of technology cycles, right? Introductions of lots of different application areas becoming sort of falling under the umbrella of machine vision. And right now we’re in this wave of AI, right, or deep learning applications within machine vision. Is there anything that you think is different about this wave compared to some of the big trends that have come before?

Matt Moschner – Cognex

Yeah, you know, you said it, we’re celebrating our 45th year in business, which is pretty amazing to think about for a high-tech company to survive that long and thrive, frankly, that long, particularly in an industry like ours that changes so quickly. Yeah, we’ve seen all of the technology trends over the years. We, you know, we like to think we invented in large respect the category, Dr. Bob Shillman, our founder. And so, yeah, we’ve been along for the ride. You know, whether it be the emergence of 3D, the emergence of color imaging, high-speed line scans, et cetera, et cetera. And what’s interesting now with AI, which, you know, we started our AI journey about a decade ago, right? The world is kind of having its AI moment now. We had ours, you know, about 10 years ago. What’s different is all those other technologies about capability, right? Gathering new types of data in different ways, higher fidelity data, different spectral data. This is much more about not doing more necessarily, but making vision technology holistically more accessible to more customers. And so we, we have this slogan, Advanced Machine Vision Made Easy, which is more than just a slogan.

Matt Moschner – Cognex

It’s kind of like our belief system now that, you know, for forever Cognex has been known for pushing the boundaries of technology and we were one of the first to kind of embrace AI. But we’re seeing that not really be enough at this point. And we’re seeing AI actually as a key enabler to not just do the thing you could, you’ve never done, but do it 90% cheaper, faster, simpler, uh, with someone that doesn’t require, you know, 10 years of experience in, in our technology area. You know, we launched some new tools last year in the consumer electronics space that we basically allowed someone with an hour’s worth of training to complete a sophisticated vision configuration that would’ve required someone with 10 years experience. That to me is exciting, right? Where of course we’re able to solve new problems, but Now we’re actually, I think, turbocharging the market itself in terms of who can use vision, who can deploy vision much, much more accessibly maybe than even 5 years ago.

Josh Eastburn

What is it about how that technology is changing that’s making it more accessible?

Matt Moschner – Cognex

Yeah, well, think about it. I mean, so, you know, you’re moving from traditional tools, we call them rules-based tools, which I’m sure the audience is familiar with, where they’re really programmatic in their nature, where you write lines of code or you’re configuring, you know, parameters to configure a tool to do a thing, and that requires a level of domain expertise, right, that you build up over months, years in many cases, you know, configuring a lighting setup to accommodate for the specifics of that station. And AI abstracts that complexity, right, where you’re learning or you’re training in this case, much more like a human would learn something by example. And then the power of these tools, right, they’re so robust that a lot of the precision required, particularly in the image formation systems, which, you know, really means the lights, the lenses, the sensing, doesn’t have to be as specific, perhaps even as dialed in because it can accommodate sort of the natural variation of the scene. And so you put those two things together where you’re training a system much more intuitively in the way your mind thinks versus how a computer thinks. And then you have a level of play in the system, or what we would call robustness, that you don’t have to be as precise and as accurate.

Matt Moschner – Cognex

Which is allowing us to take, you know, in some cases when we started in AI vision in 2017, weeks, months, thousands of images, our best people to solve a problem. We’re now putting, you know, a new engineer in an hour meeting to solve a problem, collecting a few samples of a few parts. Amazing, right? And so I, I think that is to me the shocking thing and as much as we’re leaning into solving the problems we’ve always wanted to solve, I’d say we’re, we’re actually probably more focusing on the problems we can solve today, but solving them, you know, 95% faster with virtually no experience required. I don’t know, Brian, would you agree?

Brian Benoit – Cognex

Yeah, no doubt about it. And it’s always shocking to me as I go out, visit customers, and they’ll show me some program that they had to solve some problem, and it’s multiple pages and multiple lines, and you’re trying to figure out connection A to connection B and how this chain works. And then they show me the output from their AI tool and it’s basically just one cell with some, you know, some data downstream and that’s it. You know, once that train by example kind of paradigm has just shifted everything. Like normally you’d have to, the engineer needs to think, you know, exactly what the output of A into the output of B and then all the different kind of measurement parameters to get there. And then they have to tweak the settings of each to get the output that they’re what you’re looking for, that’s gone. Like you just do not need to do that anymore when it comes to the way that we’ve set up our AI training tools. And so that’s when I was talking about the user experience, that’s one of the big pieces to this thing. It’s not just about the training in the, you know, classify A, B, and C, but it’s also the validation.

Brian Benoit – Cognex

Am I getting the outputs that I was expecting when I change or I add a new image into my dataset? That’s a key part of this, right? And our customers require that, like they’re, they’re used to being able to see, right, within their toolchains when they make a change to something, how that’s affecting their output. They have— it’s a hard requirement to do the same sort of thing with the AI workflows that we’re designing. And I think that’s something that we, by knowing our customers really well and adding these new AI tools with how they use it day to day in mind, I think we’ve It’s just amazing the level of complexity that we can solve with someone with a couple of hours of, you know, of stick time using the software.

Josh Eastburn

Yeah. I’m wondering if you feel like, has this been a, let’s say, traditionally a hurdle for companies in adopting these deep learning-based tools, just that there is this learning curve or this accessibility curve?

Matt Moschner – Cognex

Absolutely. Yeah. And I’d say, you know, again, we’ve been at this for about a decade where I, Brian doesn’t like this comment, but I think it is true. We’re on our fourth generation of tools, our fourth kind of generation of deploying AI through Cognex products.

Josh Eastburn

Oh, sounds like a good thing.

Matt Moschner – Cognex

Good marketing, right?

Josh Eastburn

Yeah.

Matt Moschner – Cognex

But it is true. I mean, each iteration we’re trying to solve for a few things. We’re trying to do more on device, right? Because the user experience from our perspective at least is best when you can start and complete your work on the embedded system with fewer images to collect and to train on. With less exotic and costly hardware and with less overall experience required in the general kind of application design. So each of— so, so for sure, when we started, it was quite complex, took a lot of time, and every generation we’ve tried to get closer to those goals. And to your point, those early days, far more of the project cost was in the engineering of the solution than it was in the cost of the solution. I wouldn’t wager to say where we are today, but that has definitely shifted or at least reduced that upfront engineering as well as the post-deployment support, right? Those systems tended to be a little more fragile, right? Requiring more retraining, Brian, right? Where we’d have to send someone to site, collect another 1,000 images, validate and deploy. You know, not only are the tools more robust, meaning they don’t need retraining or babysitting, but even when they do, you know, you just have someone go to a line, collect another couple images, train and deploy.

Matt Moschner – Cognex

Much simpler. No PC required, no GPUs, no cloud connections, nothing. It’s right there on the line. So as that has come down, the total cost has come down, the ROIs start presenting themselves again. And so a lot of the conversations I have with our customers is like, go back to that project you thought wasn’t economically viable 6 months ago, right? Because 6 months later, We might actually be able to solve it. It’s not a 3-month science project. It’s actually, let’s bring the part to the meeting right now. We’ll solve it right now kind of thing. That’s really cool. That’s really, ’cause that grows market, that engages customers, that opens their minds to where else they can use this technology. I talk to investors a lot and they think, oh gosh, isn’t the industry fully penetrated with machine vision? And the truth is it’s just not, right? Like, ’cause we’re coming from a place where, you know, the technology was quite complex. The total cost of those projects was quite high. Both of those things are just in freefall right now. The cost is coming down, the complexity is coming down. Our customers are like, oh my gosh, now it’s like a renaissance. Like, I can now take these systems to so many more places and know that I can deploy them, know that my team can maintain them, and on top of it, we can solve problems that we couldn’t solve before. You put those three things together, it feels like day 1, not year 45.

Josh Eastburn

Yeah, that’s pretty exciting. What it makes me think is a pattern that I’ve seen in other industries too, especially at enterprise scales where probably 10 years ago when people were looking for these types of technologies, it was to solve a particular problem that they couldn’t solve using the traditional approaches, right? So I’m wondering what you’re seeing now. Do you feel like there’s an inflection point here that you’re seeing in terms of adoption and like enterprise scale or just using AI more broadly across organizations?

Brian Benoit – Cognex

I could give you a quick anecdote on that one. From an on-the-line perspective, right? So in many cases where we’ve initially penetrated, there was some inspection already happening and we’ve used AI to do all the things Matt was talking about, you know, lower the barrier and make it easier for folks to use and, you know, get your ROI that much quicker. But ultimately kind of replacing what was already there for an inspection. But now that folks are getting more and more comfortable with this, they’ve done a few deployments, they understand it’s easy to maintain once it’s done. You’re finding inspections added to the lines that you wouldn’t even thought about before, right? Why not? You know, like I’m doing this subassembly. If I don’t do an inspection here, then if something makes it through, it’s going to cost me X amount of dollars to scrap it or to replace it. There was no inspection before, but now they feel way more comfortable adding a new one in because they know there’s just barely any risk there. And so as you, as we expand and as we add more cost-effective solutions, depending on your resolution and your field of view and what’s required to, from a lighting perspective to solve it, people get a lot more comfortable kind of deploying more and more because again, the risk profile, it’s completely different than it was for that very first one.

Brian Benoit – Cognex

And so we see that quite a bit. Like there’s nothing better for me than going to a customer and walking the floor and start identifying, you know, oh, have you thought about putting one there? And I know you can do a part presence in absence there because we can handle all the variation that you’re seeing. You couldn’t dream of that 5 years ago. And so that’s an area that, you know, it’s super exciting to see these vision systems popping up in areas that they just didn’t have a chance to 5, 6 years ago.

Josh Eastburn

Fantastic. I wanted to ask Brian, OneVision is your product. It just launched in June of last year, and I feel like there’s probably a scale story in there as well, but, maybe tell us, do you feel like that fits into this conversation as well?

Brian Benoit – Cognex

It absolutely does, right? And I think, you know, when we’re kind of specing out an application, determining if vision is the right solution, we always want to start on the edge, right? With our edge devices. And from, you know, acquisition to output, we want to solve it as much of that application as we possibly can with our edge tools. And our edge, what we call our edge AI tools, are the best in the industry. From classifications to defect detections to OCR, very easy to use and ultimately can solve 80-90% of the applications that are targeted toward it. But what OneVision gives you is if you need a little bit more, you have a little more variation than you’re expecting, or you need to train, instead of 10 images, you need to train 100 images, or you need to develop a model from scratch, OneVision gives you the capability to do that, within your exact same UI that you would normally be using. We’re using obviously cloud resources with GPUs and validation tools that allow you to get ultimately the, the most sophisticated tools that our engineers are developing, but ultimately deploying them on the same edge device that you started with.

Brian Benoit – Cognex

And that’s really what OneVision gives you. It’s designed very specifically to work in parallel or in conjunction with our kind of edge workflows. But gives you that little bit of extra that you might need to solve some applications without having to move to a PC or a programmatic environment, which is kind of what you have to do today. So OneVision just opens up. We want to solve more and more sophisticated and complex applications with our edge devices, and OneVision gives us a very easy way kind of to lower that barrier for anyone who’s using this kind of standard device to do that.

Matt Moschner – Cognex

I think, Brian, it’s also fair to say, you know, as much as OneVision was about creating better vision, I think as we’ve drawn more users of vision into the market, we also found that they were wanting better ways to collaborate on model creation and vision tool development. So when I talk to a lot of our customers about OneVision, I say, you know, how great are the vision tools? They’re like, yeah, yeah, those are nice. But what’s really great is the collaboration, the revision control, the collaborative annotation. And labeling where it, you know, as a multi-tenant cloud, like you can have multiple people collaborating on a project potentially at the same site across sites around the world, whereas that just wasn’t really possible. And so one of the things that’s emerging about OneVision that’s really interesting is the collaboration aspect in addition to, obviously it’s a very powerful way to train new vision tools.

Josh Eastburn

Do you find, uh, you encounter any resistance from people who obviously, I mean, people understand cloud now, maybe 10, 15 years ago that wasn’t the case, but is there a case to be made for people who are used managing everything on-prem? Is there a learning curve with like, yeah, this idea of, oh, I can share these vision models, you know, with other sites?

Matt Moschner – Cognex

For sure there is. And I think Brian said it, OneVision’s not for everyone. Our perspective is start at the edge, start your project at the line, at the station. And our job is to try to help as many customers solve it there and then, right? And not require the cloud. I think we’d very much agree with that. But the unfortunate truth was up until when we launched OneVision, if you couldn’t solve it, you had to completely change the environment and almost start over. And that bothered us, that discontinuity in the, in the user experience, like, really was a problem. And, and that, that’s where OneVision was born of. So think of it as a service that’s kind of got your back so that if you start on something, you know you’re going to finish it, even if it’s a little more complex, has a little more variation than you expected. And I think when we position it with customers that way, they get much more comfortable, like It’s not gating anything. It’s not where you start, but it’s a utility, you know, you have if you need it.

Brian Benoit – Cognex

Yeah, great point, Matt. And one thing we always have to make crystal clear to our customers when we talk about OneVision, it’s a design-time-only environment. When you’re running on the line and you need maximum speed and you’re trying to tweak that last couple of milliseconds, there is no cloud connection required. This is just once you have your model, you can deploy that on your edge device for months and years in some cases. And so that’s a kind of a key part to this thing. When we talk about the cloud, people go, oh boy, I need to be connected to the internet on my line. Of course not. Right. That’s not what this is.

Josh Eastburn

Yeah, that’s a good clarification. I also read that this year for 2026, you have three new AI vision products launching maybe in the next month or two. What are you able to say about those at this point? Either of you?

Matt Moschner – Cognex

I mean, it’s, it’s, it’s a little premature, right? They’re not launched yet. But yeah, Cognex, we invest over $100 million a year in R&D. And so every year we try to get game-changing products in the market. This year I think is, to your point, especially exciting. We have some bigger leaps, bigger swings this year that are coming, you know, very shortly. And yeah, I think they’re very much of the vein of the conversation we’ve had, which is how do we drive our most advanced technology? Much of it is AI-based in a great package, which is embedded systems with OneVision supported. To deliver both performance and simplicity. And I think, yeah, there’s a couple new projects that are completed that we’re very excited to bring to market shortly.

Josh Eastburn

Yeah, I’m thinking about maybe themes, right? Every, every company is looking at these technologies and thinking about how to apply them for their customer base or for the problem space that they’re in. As you think about that, where this technology is going within Cognex over the next, you know, 12 months, what are some of the things that are top of mind, either because you’re excited about that technology or because you’re getting feedback from customers that’s making a direction?

Matt Moschner – Cognex

I mean, so much of our work has been in vision tools development, right? And making a better tool, a more performant tool, more flexible tool. Where we’re doing, I think, some really exciting work is on the workflow of our products and the user experience and how can you automate more, have fewer button clicks, have less complexity overall, have the system adapt, I would say almost on the fly. So I think that’s pretty cool. We launched at the end of last year our first no-learning AI product. No learning meaning no training required. It just out of the box, you enable a tool and it does the thing you want it to do to, to a very high degree. I think we’re gonna be extending that, that concept of no learning. You can imagine what comes after that, which is real-time learning, right? That sort of more dynamic learning where the system is running, collecting its data and actually making adjustments. That’s maybe a little farther away, but yeah, I’d say our focus is actually less so on the tools themselves. And now we’re taking a step back to say, How do you kind of rethink the entire product experience with some of these tools and doing that in a responsible way, in a way that customers can actually accept?

Brian Benoit – Cognex

Yeah, I think those are two excellent ones. And, you know, I think we’re spending quite a bit of time making sure we get both of those right. In addition to that, we’re going to be introducing some new ways to do really good image formation, right? Ultimately, Every vision application starts with a good image, and that could be— you can have the best tools in the world, but if you don’t have the right lensing and lighting, or if you make that piece really complicated and leave it 100% to the customer to figure out on their own, then you haven’t really solved their problem. And so we treat image formation kind of as a first-class citizen in our software, plug and play. Couple of buttons and you’re getting a good image. And we’re gonna use some AI tools as well as some, some new hardware platforms to bring that barrier down even lower. And so it’s a critical piece for us. If you think about the entire work chain from trigger to output, you know, a lot of the really interesting things are happening on the tool chain side, but ultimately if you don’t help the user and help the customers get a solid image 100 times out of 100, than even the best vision tools in the world are.

Matt Moschner- Cognex

The corny analogy, Josh, is like if we’re making a fine bottle of wine, a $1,000 bottle of wine, it really does start with the grapes. The grapes being high-quality pixel data.

Brian Benoit – Cognex

Exactly.

Matt Moschner- Cognex

Right? And even in the world of AI, these models are amazing. They still fall down if that image is poor or misrepresented. And so yeah, no, great, great point. I mean, the image quality still matters and that’s your lights, your lenses, your sensor selection. How do you make that easier to work out of the box, plug and play? It’s both. Yeah, I totally agree.

Josh Eastburn

Okay, great. Well, I feel like you’re painting a rosy picture for the future of machine vision here, just becoming more accessible. Yeah, more, uh, more predictable. I understand, Matt, that you’ll be speaking at a leadership roundtable on June 22nd at Automate this year.

Matt Moschner- Cognex

I will be. Yeah, I’m excited about it.

Josh Eastburn

Yeah. Okay. I’m wondering, what are you looking forward to about that conversation? Are we, are you going to be talking more about this kind of stuff? What do you think is going to be top of mind there?

Matt Moschner- Cognex

Yeah, absolutely. Very much a state of the industry, key technology trends. You know, the exciting piece for me is Automate is such a great show, right? It’s a great way to meet customers, suppliers, networking even with some of our, you know, peers, if you will, in the industry. We would have launched some of these newer technologies that we referenced by then. So really trying to tie our company strategy to those product releases, tie them to the trends we’re seeing in the industry. Overall. So yeah, yeah, it should be a good discussion. I’m looking forward to it.

Josh Eastburn

Right on. Brian, maybe a final word from you. How can people follow what Cognex is doing? Where should they go to dig into the technology or connect with your team?

Brian Benoit – Cognex

Oh, that’s a great question. Luckily, we’ve just revamped cognex.com.

Josh Eastburn

Oh, right on.

Brian Benoit – Cognex

There’s quite a bit you can find and do to learn about machine vision, learn about our product offerings. You know, we’re adding more and more kind of do-it-yourself capabilities for our customers as well. And so that would be a great place to start. And that’ll make it just dangerous enough for you to get started. And then once you’re in the Cognex family, the sky’s the limit for you.

Josh Eastburn

Fantastic. Well, thank you both so much. This has been a lively conversation. Parting words from either of you? Anything I left out that you really wanted to talk about?

Matt Moschner- Cognex

No, things are moving quickly. You know, at Cognex, we have a saying, move fast, and we are. And so like I said, we’re a 45-year-old company, but in many ways it feels like day one. And so that’s an exciting thing for our industry. That’s an exciting thing for Cognoids around the world. Always happy to talk about it. So thanks for the opportunity, Josh.

Josh Eastburn

Is Cognoids, is that the term? .

Matt Moschner- Cognex

It is, yeah. Cognoids.

Josh Eastburn

Is that for Cognex users or Cognex employees?

Matt Moschner- Cognex

Those are, those are our, yeah, those are our employees. Cognoids, sales noise. We have all sorts of noise. Uh, we have honorary Cognoids as well. So Josh, you know, that’s always an opportunity for you if we, if we continue to work together.

Brian Benoit – Cognex

Cognoids swag, Josh, so you can wear it around. Maybe we’ll see you at Automate.

Matt Moschner- Cognex

How do you feel about bright yellow? Yes.

Josh Eastburn

My wife loves it.

Matt Moschner- Cognex

Excellent.

Josh Eastburn

That was Matt Moschner and Brian Benoit from Cognex. The frame Matt laid out is worth keeping in mind. Accuracy gets manufacturers to invest, but usability is what makes AI vision stick. And scalability is what turns a working application into an enterprise deployment. Those are the three problems Cognex says they’re designing toward, and it is a useful way to evaluate any AI vision system you’re looking at. For more on the specific products covered in this episode, VisionPro, Deep Learning 4.0, One Vision, and the new AI vision tools launching this spring, head to cognex.com. And if you’re heading to Automate in June, Matt will be at the executive roundtable on the state of the automation industry scheduled for June 22nd. Worth getting in the room for that one.

Josh Eastburn

If you’re a vision professional or integrator with a field application story worth sharing on this show, please reach out. Josh.eastburn@mvpromedia.com.

Josh Eastburn

This episode has been produced by Big Robo and Flannery Creative Studio. For MVPro Media, I’m Josh Eastburn. Be well.