Tomas Guillen on the Future of Automation, AI, and Virtual Commissioning
In this episode of Down with OEE, Patti Engineering CEO Sam Hoff talks with Gen Z controls engineer Tomas Guillen about the future of automation, AI’s role in programming, and the benefits of virtual commissioning. Tomas also talks about his early career and his experiences working at Patti Engineering.
Read more about Tomas in his employee spotlight.
The transcript below is available for those who prefer to read along. Please be aware that it may contain minor errors.
Sam Hoff:
Welcome to Patti Engineering’s Down with OEE podcast series. Tomas, thanks for joining our podcast. Down with OEE. I’m really looking forward to this podcast because everyone we’ve had up to this point has been kind of around my age. You’re Gen Z, I would assume, and really looking forward to getting a perspective on controls and the future of automation from somebody who’s fairly new to their career.
Tomas Guillen:
Yes, definitely new to my career. I guess in the grand scheme of things when it comes to controls careers because you’ll see and hear and know of people that have been doing it for longer than I’ve been alive, it seems. So it’s really cool to sink your teeth into it. I like it. It’s never ending learning, I guess.
Sam Hoff:
Yeah, I’ll tell you, as a 58-year-old guy, and I haven’t programmed anything in probably 25 years, I joke, so I haven’t touched code since age 33, but I can tell you it’s kind of an interesting balance because I think your brain is sharpest in your late twenties, early thirties, and I think after that, the brain’s a muscle and I just feel like I’m just not as sharp and lucid as I was then. But the difference is life experiences and experience in controls and experience in what works and what doesn’t work, and I’ll maybe lead you to make less mistakes than you would’ve at that age, but I don’t think you’re quite as lucid either.
Tomas Guillen:
Yeah, I can hope that I’ll dial in all the fine tuning in these late twenties, early thirties, and then I’ll just be set up for success.
Sam Hoff:
There you go. So tell us a little bit about yourself. Where are you from?
Tomas Guillen:
So everywhere is what my normal response is to that. I went to three different middle schools, three different high schools. While I was growing up, my dad was in the army, so we moved quite frequently. I never lived in Texas before going to college at Texas A&M and I love it. I love it. So now I think I have earned the right to say I’m a Texan.
Sam Hoff:
Nice.
Tomas Guillen:
Yes.
Sam Hoff:
Besides Texas, where is the place that you lived the longest and where was your favorite place to live?
Tomas Guillen:
Longest was Colorado. It was also my younger years. That was all of elementary school, but I still think back and it’s hard. That could be my favorite place, but again, I was just so young, so it’s not like more prime memories were made there. Beautiful, I love the weather, love everything about it, but maybe favorite place was surprisingly Columbus, Georgia. I lived there for one year and the high school that I went to, the people that I met, it was great. I made a lot of great friends there.
Sam Hoff:
Nice. Tell us a little bit about your dad’s army career. I find it quite interesting, and he’s still in the army, right?
Tomas Guillen:
He retired from active duty, but now he still remains active working with the National Guard. But yeah, his whole rundown is he enlisted in the army straight out of high school and served the allotted five years, six years to get school up to whatever he would like to do. So then he went back and did undergrad and law school and then hopped back into the army and did his officer career in the JAG Corps, and so that brought various locations around. That’s why we moved, well, I guess everyone moves right? But we moved pretty frequently doing mostly JAG jobs and then when we were in Columbus, Georgia, he was actually in the 75th Ranger Battalion. That was also a career-long thing. He did ranger school and then also carried through and continued to stay in ranger positions, so that was related to his deployments as well.
Sam Hoff:
What a badass.
Tomas Guillen:
Yeah, it’s pretty, there’s not much on top of officer, JAG, ranger. It’s kind of his word is law type thing.
Sam Hoff:
Out of curiosity, where did he do his undergrad and law degree from?
Tomas Guillen:
Both University of New Mexico. Both my parents are from New Mexico, born and raised. They both went to UNM.
Sam Hoff:
Nice. So I have a nephew that graduated West Point in 2010, and he’s actually in law school right now. He wants to be a patent attorney. Interestingly enough, he was not accepted to West Point, his first try, and they sent him to NMMI. Have you ever heard of NMMI?
Tomas Guillen:
No.
Sam Hoff:
Yeah, it’s New Mexico Military Institute. It’s kind of like almost like a junior college and it teaches you how to march and teaches you a little bit more academic and all. And then the next year he went to West Point, so neither here nor there.
Tomas Guillen:
Very cool.
Sam Hoff:
And tell us a little bit about your education.
Tomas Guillen:
So Texas A&M University, I did the industrial and systems engineering program, which has high level applications for some of the things that we do as a systems integrator, but the world of controls is really its own thing. There were some ideas of, I remember I had moderate exposure to PLCs, but nothing really deep enough. And so yes, I graduated with my Bachelor of Science from Texas A&M and came straight to here.
Sam Hoff:
Now you had some internships, so you interned in the summers. Why don’t you tell us about your internships?
Tomas Guillen:
Yes, I am so grateful for those. They kept my eyes open and looking forward to what comes after college. So I interned with Patti Engineering the summer of 2020, the summer of ’21, and then when I graduated in spring ’22, I came full-time. So there was some momentum with that and I was very beyond blessed really, to have such a smooth transition into this job.
Sam Hoff:
As I understand it, you did another internship at Georgia Pacific the year before you did the Patti Engineering internship?
Tomas Guillen:
Yes. That is how I got my foot in the door of all of this. I mean, people talk about the term canon events, and this was my canon event. I guess I was wandering around the career fair at Texas A&M, which is amazing by the way. And I made eye contact with this guy at a Georgia Pacific booth and no one was standing in line at this booth. And so I walked up, introduced myself, and I talked on and on as much as I could about this marble sorting project that we did with these Lego robotic kits. And little did I know, it was basically the fundamentals of controls engineering and an automated line in a factory, just Popsicle sticks and Lego version and that I was talking to the controls engineer. So it all just, it happened and I talked about it and I had my first summer there at Georgia Pacific. It was great.
Sam Hoff:
That is cool when you think back to moments like that, pretty awesome. I got to tell you one quick story, and I don’t know if you’ve heard this before, but Dave Foster, who’s the right hand man here, kind of runs operations at Patti Engineering and a whole lot smarter than me. We were at the auto show in Detroit in 1994, and at that time there were three people in Patti Engineering, my alma mater Kettering University had a hybrid electric car that they built out of a Saturn, and we went up to the car, they had a booth at the auto show in the basement. We went up to the car and I said, who are the double Es on this project? And there were two of them, and one of them was Dave Foster and that led him to apply to work at Patti Engineering.
Tomas Guillen:
There it is.
Sam Hoff:
Yeah, so it’s amazing where you meet people and what happens and that kind of stuff.
So with the perspective of being a Gen Z, how do you see the future of controls?
Tomas Guillen:
One – there are a few different things floating around. I guess it’s funny because this experience with Patti Engineering, all I’ve known is the cutting edge, right? Like TIA portal is immensely powerful compared to, I guess, the whole history of controls and what it stems from. So we already live, in my eyes, we live in the future right now compared to what people, I guess the fundamentals of what it was like beforehand. But the immediate things that come to my mind that while using, for example, TIA portal, because that’s what we touch the most. There have been times where it does feel very behind, I guess, and that is particularly with this thing called SCL, Structured Control Language, which is a different way to write controls logic instead of ladder logic. You can use more written text and we were using it for more simple applications, just HMI animations, and I remember it was frustrating, bothersome, because the development environment didn’t have any debug, any amount of error messages of why the code you wrote wasn’t behaving as you wanted.
So it was a very tedious process to basically remove code that you thought could be problematic and then guess and check, and it was just red light, green light of does it compile or does it not compile? So the future of controls, I can see it opening up in this particular example to written SEL and written code being more applicable, more widespread, more used just because that is a platform, that’s a format that Gen Z and I guess younger controls generations are familiar with and will use more widely compared to – again, the fundamentals of controls and how back in the day, it was just relays wired together so that transition could be more prominent. And it already is like written controls, language is widespread, but I still see just these corners, these little areas that still feel old.
Sam Hoff:
How do you see AI and machine learning going into actually developing code in the future?
Tomas Guillen:
That has crossed my mind, especially when we had a project where we were doing a control switchover. It was the first big project that I was on, and I asked myself that. I was like, how far is AI going to go and will it ever get to the point where it writes logic? And it’s hard to imagine, at least right now, or my understanding of it. I would think it’s really cool for AI to be applied for troubleshooting, like watching the behavior of all of your inputs and outputs and probably the mechanisms of your logic itself and assessing circumstances when bad behavior or good behavior is happening. I think AI can, at least that is the lowest hanging fruit, but walking into a factory and somehow the AI taking a large workload off of developing brand new control software? That is far out in my eyes, only because it would have to obviously not just know the fundamentals of the logic and how to write the software, but it needs to have a deep understanding of the process. It’s trying to control the equipment, how, you know what I mean? If AI can pull that off, we’re in trouble. It is very, very aware, very, very aware.
Sam Hoff:
And I think we’re not necessarily in trouble. I think like, okay, if you take a look at AI, and most of it’s like a good assistant. You know what I mean? And AI learns, and I think in order to use AI to really develop code that’s going to match our standards and our customers standards, you’re going to have to break it down into a lot of small chunks, give it simple problems to do, let it do the code, kind of review it. We’ve used AI some to develop some of our marketing and this kind of stuff, some of the sub pages on our website, and it’s very much when you see it, it’s just not ready to go. It needs to be reviewed two or three times, edit it and this kind of stuff.
Tomas Guillen:
In this sense of it being a good assistant, I can totally agree with that then, because the particular example that I was thinking of with the first project that I worked on, it was a large coffee extractor where you have water getting pumped in, steam flowing in as well, to heat the water. So all of this is PID madness. And the heat exchangers are bringing the water up to temp, and then you have the coffee grinds coming in. All of this is happening at once. So as far as AI helping with the development of that code, and like you said, it being a good assistant, it would be awesome. It would be cool if you just say, write me the skeleton, the framework of I need four analog inputs and I’m going to be considering this amount of information. You give descriptive text on asking the fundamentals of what you want, and then it generates that, and then you bring it to the finish line, I guess when you’re actually on site.
Sam Hoff:
Yeah, I think AI is going to make our future a lot more efficient. It’s going to change our jobs a little bit, but I really think it’s something you can’t ignore, right? It’s coming. It’s like the guy back in the early seventies that was making these large relay panels and was ignoring this thing called a PLC, right? Because that can never replace relays, huh?
Tomas Guillen:
And now we have safety over – PROFIsafe, right? It’s just crazy how far we come from relays.
Sam Hoff:
Yeah, for sure, for sure.
Tomas Guillen:
The other future, I guess future dreaming… if you think of Tony Stark in his cave and all the holograms floating around that are in the air – that type of augmented reality that’s just floating out in front of you would be so cool if whenever you cross-reference different bits in controls logic, you can see the before and after of what’s talking to it and what it’s talking to. That’s less applicable and less of the real AI industry, what’s going to push us forward. But that’s my dream. That’s what I think the future of controls is jokingly, that if we have that, and that comes to reality and we use it; versus – compared to when they look back at us and they’re like, “wow, they had to look at their computer monitors.”
Sam Hoff:
And you think about it, I know that we’ve really seen the benefit over the last two to four years, two to five years of virtual commissioning. And I know Texas, you guys did a project last year where there were 15 different installations and we had to do it on weekends, while the system was still running. And we really took advantage of the virtual commissioning and simulation to be able to test our code to make sure that when we were doing installations, at least we knew our code was correct. Right?
Tomas Guillen:
Absolutely. Yeah. Arian was the man leading the charge on that, and it was great. They’re exactly what you said. We would simulate each section of the install that we were doing just so that 80% of the debug logic kinks can be resolved before actually getting on site. Because yeah, that customer was actively running and using the line. So the max downtime we could get was four, day install windows.
Sam Hoff:
And that simulation that you do in the virtual commissioning, that’s the basis for the digital twin going down the road, right?
Tomas Guillen:
Yes.
Sam Hoff:
It’s so funny. My son is big into iRacing. I don’t know if you know what iRacing is?
Tomas Guillen:
Is it more simulated accurate track racing?
Sam Hoff:
Yeah, a lot of the drivers will use it to prep for a road course or whatever. I’m sure a bunch of the NASCAR guys all this week we’re driving COTA on the simulator just to get a feel for the track.
Tomas Guillen:
That’s such a funny application of digital twin because absolutely. I got a steering wheel and pedals for my computer, and I did F1 with my friends for a bit and it’s quite accurate. And then, that was a few years ago, but just doing that and then for the very first time two weeks ago, I went go karting and I realized that I had the oversteer reflex. As soon as my rear end started sliding out, I did an instant steering wheel correction and I was like, oh, wow.
Sam Hoff:
You did learn something.
Tomas Guillen:
Yeah, that definitely came from the F1.
Sam Hoff:
Yeah, it’s cool. It’s cool. Now I understand that you, in your spare time, you and I share a similar type thing that we like to do. We both play poker a little bit, huh?
Tomas Guillen:
Oh yes. It had its peak maybe a few years ago, but I still stay involved whenever I have the opportunity for a house game with some friends, and I’ll watch here and there. But yeah, it’s great. I love it.
Sam Hoff:
And playing online is one thing, and it’s very mathematical and whatever, but there’s nothing like sitting at a table and looking at a guy across from the table or a lady across the table and trying to figure out whether they’re bluffing or not. So tell us about your favorite project at Patti Engineering.
Tomas Guillen:
Let’s see, I’m going between two, but the one that is more on my mind, it was a short and sweet project, it wasn’t anything crazy like a war story of the time endured, but just a simple solution where we went out to a customer who had a garment sorter, and they were having poor performance on this sorter. And we initially diagnosed it as the chain being stretched uniformly, and that was causing this issue. So pure mechanical, as soon as you have something like that, there’s no really amount of software that can resolve it. So we gave them the bad news that that was the case, and they put a chain on order. I went back like eight, nine weeks later. So it was the chain, because then the behavior that we were seeing was no longer there. It was just so fluid. It was my favorite project because it felt like almost a lab in college or something where you go through all these steps. It felt like so orchestrated, I guess, because the changes that I made and the result going from how poor it was performing to 100% accuracy, it felt it was such a slam dunk and it was very satisfying is probably why it’s my favorite project, just because everything flowed perfectly. It was great. So it was very satisfying.
Sam Hoff:
And that is cool. Yeah, you kind of always liked in your career to have those type of wins like that. And one of the things that I think sometimes our clients or people, end users, fail to realize is that in solving any problem, I mean, we’ll go in and people will just be changing stuff. We’re changing this board, we’re changing this board, we’re changing this, and they’re just trying to search for solutions and by trying a bunch of stuff, and a lot of times they make it worse. 95% of any issue or problem is finding the root cause of that problem. When you do that, it’s fairly easy to fix. So you guys found the root cause. The root cause was not throwing a whole bunch of code at it or trying to make the drop off points on this chain more accurate. The root cause was the chain, right? The chain itself was longer in some sections, shorter in other sections, which was causing the hook to hook distance to be different throughout the chain. And you guys found the root cause, they replaced the chain, and this system has been fixed, right, until the next time they have uneven chain stretch.
Tomas Guillen:
Yeah, another 20 years.
Sam Hoff:
Yeah, 20 years down the road.
Tomas Guillen:
It was hard to imagine it being anything else truly. But I did breathe a sigh of relief when the new chain got put on, and then I saw that the behavior was no longer there because I was like, okay, yeah, imagine, right? Imagine they put the new chain on and then it is still something else. But thankfully we were on the right track.
Sam Hoff:
I had one, I think this is back when I was writing code in the ’90s, but I had one where they were having an issue at a paint shop that was happening for two years. It didn’t happen all the time. It happened once a day or whatever, but it would pretty much mean that the vehicle that they were painting would need to be repaired. It would not spray, it would just drop out of the gun. And they had a couple different PLCs communicating with each other. And I realized that there was, because back in those days, things didn’t communicate real fast. There was a spot where the one PLC could miss the other PLC signal to turn on the gun and all that kind of good stuff. We put some hooks in the logic to fix that and took care of the problem forever. The fact that there were a bunch of people that had been there before and tried to fix it and had unsuccessfully done it and I did it, I kind of put that as a feather in my cap. And I know a lot of you young guys don’t believe I ever coded, but yeah, back when my mind was a little bit more lucid and I didn’t get a lobotomy, because most of what I do now is kind of sales and meeting with clients. I was a pretty good programmer.
Tomas Guillen:
I believe it. I believe it. I’ve heard other war stories of you getting your hands dirty.
Sam Hoff:
Yeah, I did come up in an era though, where memory was precious and you could do a lot with indexed addresses or I wrote a lot of code that nobody could follow, and documentation was not my strong suit.
Tomas Guillen:
That’s just job security.
Sam Hoff:
There you go. So tell us a little bit more about yourself. How do you like living in Austin, Texas?
Tomas Guillen:
I love it. It’s funny, I always say that I have an unbiased opinion of the Texas triangle just because I’m out of state. So I have spent time in Dallas, I have been to Houston, I guess this is all before graduating college. I was evaluating the three cities, and Austin is definitely my favorite. I am happy to be here, it’s great, yeah. The hill country out west is beautiful, whenever you want to explore. I don’t know, it’s the sweet spot, honestly. It’s not as humid and muggy as Houston, and it’s more lively than Dallas.
Sam Hoff:
I’ll tell you what, my first trip ever to Austin was 2009 when we were thinking of opening an office down there, and I just can’t believe the amount of growth in that city. I mean, I remember when I went down there originally, the Frost State Bank, the highrise, kind of dominated the city line. Now nobody even notices it because there’s so many high rises around it. Yeah, Austin is a cool town, and I know you’re supporting one of our local companies there by driving a Tesla, which you gave me a ride in. I got to tell you, the instant torque of a Tesla is pretty damn cool.
Tomas Guillen:
It is amazing. I have never had a performance vehicle, it’s always been like a 150 horsepower Subaru or a full size truck. So jumping to this was great. It’s funny, I made the mistake, I guess, of putting all-season tires on it, and you can feel the difference in acceleration. So once those get chewed through, I’ll go back to the summer compound.
Sam Hoff:
Oh, they’re just a little bit slower?
Tomas Guillen:
Yeah, it’s noticeable. It’s not a little bit, it’s not even the same car. So it was a bit of a mistake, but lesson learned I guess.
Sam Hoff:
There you go. All right. So any parting words or parting advice for advice? Maybe somebody’s listening to this and just starting their college career or they’re late in high school, and any parting words for them on the benefits of getting into controls and manufacturing?
Tomas Guillen:
Stay curious, stay active, definitely. If you’re thinking about controls or you’ve just heard about it, invest. It is… probably why I’m so excited for it to be my career path is it’s never ending. You get as much out of it as you put into it. So the deeper you ground yourself and the wider of mastery that you achieve, that directly feeds back to your capability and what you’re able to do. So I personally love it. And if you have any amount of interest or curiosity, I would definitely recommend pursuing it.
Sam Hoff:
Awesome. All right, man. Thank you for your time. Appreciate it.
Tomas Guillen:
Thank you. Thank you for having me. This has been great.
Related categories: Blog Podcast
