Test Systems and Automation with Averna's Hugo Lessnick

Hugo Lesnik: We always want to keep this interaction with the client and make sure that we are discussing often with them just to make sure everything is clear that we're going the same direction. There's no surprise and usually that's the key of the success of a project.

Jim Beretta: Hello everyone and welcome to the podcast. It's my pleasure today to have Hugo Lesnik who's the engineering manager for a company called Averna and they're out of Montreal and they're an automated test company. So uh, welcome to the podcast, Hugo.

Hugo Lesnik: Thank you, thank you for having me.

Jim Beretta: So can you tell me about Averna? And I think it's one of those great companies that is like a big secret in the Internet. And I know you don't mean to be a secret but uh, I'd like to uh, help uh, solve that.

Hugo Lesnik: Yeah, sure. So Urvana was founded in 1999 in Montreal. So over the year uh, we grow with natural, I mean a lot more client and everything. But we also did acquisition uh, to cover different market. So we now have 11 uh, offices in different countries. So in the US Mexico we have in Europe, so Belgium, Poland, uh, Czechia, Germany, Vietnam, India, China and I'm missing one which is Japan. So I mean obviously we grew up a lot uh, with the year and also our expertise grew up. We are tackling a large uh, project now so we are almost 1100 employees and mostly engineer in uh, different field activity, mechanical, electrical, software and others. Averna, the branding. We are a testing uh, company so we are really, our expertise is testing but, but we are also doing automation on production floor for example. Now I was just saying that the big news for us also was that in the beginning of the year we joined with a new company, Sveria. So they are based in France, in uh, Toulouse. So that will help us to expand our technical capacity and we are very complementary to each other. So they are really in the aeronautic and defense market. We are in pretty much everything else. It's a really good match. So we hoping that that will generate uh, also other opportunity for us.

Jim Beretta: It's exciting and I was going to mention when I interrupted you, so sorry about that, is that uh, it's really hard to separate yourself from doing like an automation system and a standalone ah, test system. So with that it's a good segue for me to ask what kind of systems do you build? Are they integrated, are they standalone or are they automated?

Hugo Lesnik: We're doing pretty much everything so it's really based on what the client want. So we're doing integrated Solution or system, uh, that will go to whatever system that they have. If they have a production line and they want for us to work on a specific module or a specific system to address some needs, we are able to do so. But we can also create, I mean the full line, automated line. Now we have the capacity of doing this, uh, but still there's a lot of clients that's asking for standalone system because either the throughput is not that much and when uh, the product is a high value product, low volume, sometimes it's making sense to do a standalone system. It can be automated, semi automated or even manual. So we are already adjusting to whatever the client needs are.

Jim Beretta: So Hugo, how did you get into the industry? I'm curious.

Hugo Lesnik: So I mean it's been now 16 uh, years that I ah, that I work for Averna. It's easy for me because my, my son like was born two days after I started at Averna. So it's always easier to know exactly the, the number of the time that I was uh, I'm here. So I started as a software engineer. So it was my second job after university. And then I, with the time, um, I did a lot of software obviously for you know, test sequence. But also uh, we worked on software uh, to manage uh, the production floor. So more the data side of uh, the test, uh, because testing is generating a lot of data. And then so I did that for a couple of years. Then I migrated a bit more to the hardware side so getting exposed to the mechanical, electrical and also did system engineer. So system engineering is uh, what we call a technical lead on project. So someone that know a bit, everything that can take the call and make sure uh, the project is going in the right direction technically. And now I'm the responsible for the engineering group in Montreal. So I have under my team mechanical, electrical, software and also project manager which, which are the key resources doing an actual project.

Jim Beretta: You know, project management is so important. We could probably have a separate uh, podcast about that all on its own. So but and you must see products, right? And you're like I wonder how they tested that. Do you do that?

Hugo Lesnik: Totally. I mean uh, each time we we looking at uh, a new product we always okay, how they, how they did that. You can imagine the solution that was designed in order to test this specific product. So yes, it happened a lot.

Jim Beretta: So what is testing to you?

Hugo Lesnik: Well, testing if we, I mean the short answer for this is making sure that the product perform the way it was designed for. So I mean we can resume testing as this Sentence but it's a bit more complex. We can also say it's making sure that the end client will receive a product that will work. So that's why the company want uh, they want to make sure that when the project, the product leave the factory and going to the actual client everything work without issue. So testing is often, I mean it's part of the assembly process at different stages. So I mean you want to make sure to capture defects as soon as possible because the more you wait in the assembly, I mean the more time it's going to take to find it and, and to resolve it also. So this is where it's super important to have testing. Usually we test sub assembly first and then it got it get assembled to one product and then we're doing like a final, we're testing a different stage of the uh, of the production assembly just to make sure that well we have a better access to all the equipment before it's, it's, it's assembled and to make sure that we found the problem as soon as possible.

Jim Beretta: Are there some big like overall trends in automated test?

Hugo Lesnik: I would say the big one right now is automation. The reason is it's fairly simple. I mean company want to reduce the uh, cost producing product. There was also an issue in the past two, three years about resources availability which is still there. But I would say that they want to see how uh, can we produce trying to reduce the cost. So I would say that's the big push that we're seeing right now in I mean for the past 23 years the big focus was about EV so electrical vehicle which it slowed down a bit right now with different decision that was taken. We can see that the slowdown on that area but uh, we also seeing that every, all the product regarding the data center this are also very hot right now. So that's another trend that uh. It's more market trend that we're seeing that we are getting more requests from customer with product going to data center. And there's also the, I mean currently with the situation in the world, aerospace, defense, it's something that we see that uh will probably come like in the next month year that we're going to see investment on that, on that market.

Jim Beretta: Hugo, in our warmup we talked about uh budgeting for test equipment and it kind of seems that people or companies they tend to underestimate the costs for uh test equipment. And so I guess uh. Do you have any comment on that?

Hugo Lesnik: Some companies are really good at designing products. This is the interesting part I Mean this is where you want to put money, you want to build something new, something innovative. So that's really um, it's easier to get budget for this. But at the end if you're not doing production and, and production include testing, I mean that's what's bringing money. So this is where it's super important because I mean we saw in the past good product that had bad quality and at the end the product was not a success because the quality was not there. And quality, to assure quality we need to do testing. Testing can be expensive. We are testing the uh, limits of the product. So let's say uh, I'm an example. We're testing electrical breaker. We want to make sure that when there's an issue the breaker will do the job. So that means we need to create a short and we don't want to destroy the test station because we want to be able to, you know, it could happen that a product has an issue with it, it will maybe explode, maybe not working. So I mean we need to build those testations to withstand the worst case. And also there's expensive equipment based on what type of test that we need to do. I mean we need to use instruments and they are expensive. So I mean it's like you said, it's always a bit underestimated. We're used to talk with clients, they understand usually pretty well. It's just that maybe the budget needs to be readjust uh, a bit to match a bit more what they really want to have in term of quality.

Jim Beretta: Aligning those expectations for sure.

Hugo Lesnik: Exactly.

Jim Beretta: So how do you work with customers?

Hugo Lesnik: Well it really depends. We are pretty flexible. So I mean we have different mode of engagement with clients. It's really depend where they are with their product development cycle. Uh, if it's a new product that they are building, we probably will be like to be involved a bit more in nvents so that we can help them to think about, okay, it's good that you have a product but how it's going to be tested. So we can work either with a spec phase, uh, when there's an amount of worry that we have with having how we're going to test this, we can go with the spec phase to okay, let's try to see and understand what can be done. And we're going to do something like uh, a prototype and try to see if the solution that we are proposing are working correctly so that can secure a bit more the client because he doesn't want to commit to something that he's not sure it's going to work. Uh, so we can have a spec phase. If we're doing repeat, uh, it's a bit different because we take uh, whatever the client have. It's a bit more safe because they already have the solution on the production floor. So we can repeat it so there's less risk, so we can go more in a turnkey type of engagement. But for our project, no matter what the mode of engagement is, we always want to keep this interaction with the client and make sure that we are discussing often with them just to make sure everything is clear that we're going the same direction. There's no surprise. And usually that's the key of the success of a project.

Jim Beretta: I was thinking about that. Thank you for that answer. I was thinking about that in really automation and test you should really err on over communication, right? Like you know, make that one extra phone call because that's really going to help you.

Hugo Lesnik: Don't assume anything assumption is the worst that can happen. Sometimes we're trying to say, oh, it must work like this, but after you get surprised. So it's better to like you said, over communicate multiple sync and make sure that we are on the same line. Usually we prevent a lot of problem after doing that.

Jim Beretta: With a big experience in machine builders and uh, robot integrators, I know that test specification documents can always be better. And how does that work with you guys? Because you're probably the same exact way. It's like yeah, machine, uh, builders, not great test specification generators.

Hugo Lesnik: Again, it depends with client and depend of what market they are, uh, some. I mean if you are in medical, I mean oem manufacturer, product designer, I mean the spec will be a bit more complete usually and it will be like easier to work with, uh, that if you're working something off the shelf that customer can buy, which is no matter the product. So it really depends. It can always be better. What we're seeing often is that we start with a first version, but there is uh, often iteration on that version based on the feedback that we give. Uh, because we're missing that. Are you sure you think about this? We can see that during the project we can see revision going like, okay, revision 6, 7, 8. So again it's working together to make sure that at the end they have the test coverage that they want. And that's what is really important.

Jim Beretta: Let's talk a little bit about machine vision and AI because these are probably two things that you talk about every day.

Hugo Lesnik: AI, it's everywhere right now. I mean it's a big trend. Uh, no matter what you're doing in life, AI is there, I mean everyone is looking to see how they can improve their uh, the work and be more efficient to what they do. Uh, Averna. And, and I can say in general the testing industry are doing the same on the test station which is super important for us because we are generating a lot of data matrix log. If we're talking about a test that take two, three hours, I mean this will generate a ton of, of result. So what is super important for AI Because AI is a manifest tool to analyze and to find issue on a production floor. What, you know, garbage in, garbage out. Yeah, it was, I mean this, the sentence was uh, initially for database but I mean AI it's the same thing. Uh, if you're provided garbage to the AI, it will give you garbage. So this is why uh, the test station that we're doing every metric result, everything needs to be organized in a way that we can after that use AI to find production issue. If you have an issue with the low yield, for instance, is it the station, the operator, an instrument, is it a new component that we put on the, on the uh, on the product. So if everything is clean, you're going to have like successful results with the AI. And this is the thing that we try to do right now. If we're talking more on the, on the station, make sure the data uh, is clean, make sure the data is organized.

Jim Beretta: Yeah, those are exactly great recommendations. So do you ever work with uh, companies to help them with their test specifications? Like if, let's say it's a bit weak, it's like okay guys, we got to work together to create a better effect.

Hugo Lesnik: Yeah, I mean it happened a lot. Either there's some missing uh, information in the document and they know it's just that they don't have the information. Sometimes it's hard because doing ah, at test specification you need to be in line with the R and D group. Uh, so this we don't have access to the R and D group. So it's really the test engineer on the client side that will talk with them. Because I mean engineer, ah, the R, uh and D group knows exactly what they, and they know exactly what they need to test test engineer. It needs to translate this into a spec that is useful for us that we can build a system that will do that. But that's a, I mean again uh, R and D are good at designing product. They are not like test specialists. So this is where we can help. And if we're seeing some gap in the uh, test subscription. Uh, we can have a discussion uh, and try to see is it normal, are you sure that you're. Because you're missing this, this and this. And we can work with them to try to improve the test coverage of the product that they want to test.

Jim Beretta: And you work with um, like when you're not doing direct work for maybe a manufacturer, you're working with automation integrators and machine builders. Because these guys as I mentioned earlier, they don't really have a specialization around testing or instrumentation or so. And so that just means extra time probably in your schedule does it?

Hugo Lesnik: I mean it's two different things. So automation is one thing, testing is something else. So yes we had it in the past project where we automation uh company reached to us to say okay, uh, I mean we have this, this and this. We need to do that. Can you take care of the probing aspect of the dut and we will make sure that uh, the product will be there and everything. So yes we did do those kind of project and we're still doing it now. The thing with Everna is we involve and we got also different capacity now that we can also tackle the full line of automatization so we can do testing plus everything around it uh, to be able to move the product at the right place and make sure it's going to whatever other section that it needs to go after. So Everna is now capable of doing everything. Uh we have a lot of experience in automation and uh, on different complex system that include test or not. Sometimes it's only to automation to build the actual product. So we want to make sure that uh, when we doing the assembly everything it's, when it's super precise we can build system that will help using vision and other technique to make sure that the uh, assembly is super precise. We're capable of everything now but we still have clients that are automation uh, integrator that reach us mostly for the testing side uh of the line.

Jim Beretta: You go. When you're working in some industries like medical device it's tricky because the companies may be trying to concurrently build the automation for the device and the test system at the same time. So what happens when you have to build in that extra flexibility for your test or fixtring or whatever?

Hugo Lesnik: Yeah, I would say on the medical, medical it's pretty uh, cycle uh, uh, it's not super fast. So medical usually I mean they have time to build a product and, and we come at some point more at the end. But if you're talking about, I mean fast paced. This industry or company that are developing a new product each two years. This is where it's becoming so important to be involved as early as possible because they don't want to wait until the product is done before thinking oh yeah, uh, we need to do production so we need to do testing because I mean the product needs to be on the shelf by Christmas or whatever date it is. So they involve us right from the when during the product conception. And I mean we try to work in parallel. Uh, for sure there will be time where we need to go back because the product went back a bit. But if we can build um, the solution in a flexible way, we try to minimize how much the delta will be uh, if something has to change. And the big benefits on this is that as soon as the product is ready the test system will be close also to be ready. That's what company like this want. They want to be able uh, to be close to the market as soon as possible when uh, the product is uh, ready.

Jim Beretta: Hugo, you probably sign NDAs all day long but do you have some projects that you could talk about?

Hugo Lesnik: Yeah, that's a bit always tricky to talk about project uh and we have so much variety of project that we're doing in different market, field of activity, military. Uh, we have some product that can find at Christmas time for the kids for instance. I mean, I mean it's hard to select one. I mean I can talk with one I did lately was uh, uh, on a uh, surgeon robot. So it's super interesting because obviously it's technology is great on this. So we developed two systems that will help to well that will perform tests to make sure that the arm is able to withstand some uh, force on each direction and also perform other additional tests. But it's super fun to work on project that uh, you can see that it's going to be used, I mean it's used in hospital, it's used uh, to do surgery. So it's cutting edge technology. So that's kind of project that was always fun to work on and when you have, you know exactly what's going to be the purpose for it.

Jim Beretta: I think it's also exciting too to work in healthcare. And you know I think of your engineers and project managers and they're helping people or helping you know, uh, uh, in people's times of need. So I think that's really interesting and exciting part of working in healthcare industry. And I bet you I know who that project was for. Hugo, is there a really difficult test project that sticks out in your mind.

Hugo Lesnik: I ah, would say some time. Uh, because we discussed about AI, we didn't talk too much about vision, but vision is a big part of what we're doing. Vision, uh, is used to inspect product. Uh, that's the general definition but you can go as deep as needed on this. So uh, a vision. One type of project that we often do is to inspect like product. And sometimes it's product that doesn't have any electronic on it, it's just like a tube, plastic tube for instance. It looks so easy to say, okay, let's look at defect. But the question is what is a defect? What is acceptable, what is not? And try to. Because they can have multiple, can be a scratch, it can be uh, a um, defect in fabrication. Uh, it can be anything. So this is a bit complex often to define with the client. Okay, what are you looking for? Because often we realize that client, they want to, yeah, we want to make sure that the product is good. What each defect you want to look at when you want to do everything. This is where it's becoming complex and we talk about AI. AI is part of vision now, uh, machine learning also uh, that help us to you know, to try to make the model, learn what is a defect. But again the client must provide us uh, the information. What is a defect? So often what type of project, it's always harder to sync the expectation of the client with actually what we need to do in term of performing the vision inspection test.

Jim Beretta: That would be a good job for AI, getting that out of the, out of the client actually. And I just wrote that down on my pad. What is a scratch? Because I've been asked that before many years ago, um, is there uh, are cycle times getting faster or slower or is it all over the map?

Hugo Lesnik: It's all over the map. It's really based on what type of tests we're doing. Uh, we had tests that takes 10 seconds, we had tests that takes hours. I mean if you want a good example of long test is when you're doing a burning test, which is performing a test over and over and over again in a temperature chamber. So you want to cycle in different temperature, so minus 10 to 40 degree. And, and you want to make sure that your product can withstand six hours of going cycling in temperature performing exactly the same test. Uh, so it's, I mean automation can make the test time uh, faster because if you want you can remove the human interaction that you have to do to do connection and stuff like this. But at the end there's tests that take times because, because time is an important variant in test that uh and I talk about temperature chamber but we can do like uh, vibration also uh, so doing tests, uh, doing vibration. So there's many, many example of tests that take time. So there's not like, I mean it's a bit all around uh, in term of what we are seeing really depend of the type of test that we're doing.

Jim Beretta: Is there like an ROI to a test uh solution? Like do your customers ever say yeah this is about a two year ROI or a ten minute roi?

Hugo Lesnik: Yes, uh, mostly when we're talking about automation. So because you have you know the matrix you have. If let's say you have an example of what you're doing right now with either a manual test or a semi manual system that you have, you want to go to fully automated solution. You add uh, it's easier to compare that if you have nothing. So that would say it's the easiest one. Then you're talking more about quality. What's the price of quality? What's the price of having issue in the field, sending people or bringing back the uh, product. So that is where it's, you can have data if you're already doing the product. If it's a brand new product. I mean you need to run an evaluation and that will help you to invest the right budget so that the return uh, on the investment makes sense for you and the management.

Jim Beretta: You've got so many facilities in so many countries and so many customers. Uh graphical user interfaces and ergonomics and, and safety. Is that coming into like every convers conversation as well?

Hugo Lesnik: Yeah, it's super important, ergonomic super important because they have some rules and guidance. Each client have safety uh group that make sure that all stations are complying with standard and norm that they have. So we must know uh, in advance because that will affect the design safety. Like I said we are testing the product sometime at the highest point. So, so you want to make sure that. So sometimes it's happened that it's it break. So we need to make sure that safety on the operator is protected. Super important. I mean we're doing electrical tests, sometime mechanical tests. So we must make sure the system comply with all the um. I mean there's some you know, basic rules or guidance that we're using but often the client also send us uh, what it's important for, for them and what we must be, we must take care in consideration when doing the design because they are pretty much requirement for the, for the system and if it's not passing their inspection and all the proper documentation is not there, they will simply refuse the system, saying that this doesn't go on the production floor. Super important. Ui, ui, it's always the first thing that you're seeing. So I would say you can have a big system. If the UI is, is like a bit low in feature, people will egg the system even no matter what the system is able to do. I mean, UI is the first thing, it's the interaction with the operator. So I mean we must be careful and put some effort and feature on the UI to make sure that it's doing the job. Information is there, it's clear and it's easy to uh, be usable by uh, the operator.

Jim Beretta: So you've written the spec or you've helped write the spec, you've designed the equipment, you've built the equipment. The now you have to uh, and you've tested it, now you have to commission it all around the world to the different cultures and in multiple languages. That must be a challenge.

Hugo Lesnik: It's a challenge. I mean we can have a client in North America, but he has a factory in Vietnam or in Asia somewhere. When we're doing the commissioning, we will go on site to make sure that the system can, is working the same way it was working at our facility. So we perform tests and everything. But again, like you said, it's different culture, different language. I would say that usually, I mean people are super, uh, willing to help and usually when we have a new system, it's going to be beneficial for them also to be part of the. Either they have a new line so it's part of the new line, or the system replace something that is old and was breaking often. So people are super willing to make sure that we are well, they welcome, uh, well us and also that they will help us in any way to make sure that we are able to do the uh, the acceptance of the system on site.

Jim Beretta: Hugo, I'm going to switch a conversation a little bit and we're going to talk a little bit about talent, about bringing in new talent into Averna. So uh, you know, when you're recruiting for new people, which I'm assuming you're growing as much as you are, you're doing this all the time. What makes a good hire? Is it curiosity? Is it technical? Uh, is it uh, people skills?

Hugo Lesnik: Uh, I would say everything now but the DNA. For, for me, what is super important, it's flexibility and curiosity. We are doing a ton of project. It's not well it can happen. That project is over a year, two years but usually we're doing more project that you can calculate in months. So that means that you can work on different project during the whole year. So and sometime at this, sometime at the same time. Uh, so you have two, three projects that you work because I mean you are working for a specific amount of time like 100%. But after you need to support there's integration, they have validation. So we need people that are flexible uh, and they are able to work on multiple projects. Since we're doing different technology because the client asked us to do can you work, can you do this or use that. So it's really important that they are comfortable to even if they don't know they have the baggage and they will receive the help to make sure that they will ramp up uh easily even if it's something that they are maybe a bit less comfortable. So I need to see this. When I'm meeting with a candidate I need to see that he has this willingness to learn and flexibility and he likes to see different things because testing can be, I mean so much different Testing we can do, can be, can do like electrical testing of pcb but we can do also super mechanical station that will turn stuff and press on button and other stuff. So it's really large in the term of uh, what we can do. So this is what we I'm looking for on a candidate. Flexibility, curiosity and having fun working. I mean at the end we want to have fun during a project. Super uh, important for me uh, and all the project that I did that the team, it's interesting what they are working on and they have fun and

Jim Beretta: you learn so much right. Like it's just every industry, every mechanical or physical attribute, every scratch like you get to learn so much. It's such an exciting uh industry. Hugo, last question. Like what's the future of test equipment and automation from your perspective?

Hugo Lesnik: Well this, this is always a bit hard uh to see or to, to think what it's going to be in 10, 20 years. Uh obviously AI will be part of it in some way. Is it clear? Probably not at this point. Uh, we, I think each uh Avrina is looking at this right now to see how can we introduce this more and more in the system. I mean Averna, uh we're used to work with new product that are cutting edge product on different field of activity. So I mean we need to stay aware of what's coming uh, what are the new technology that we need to learn because at the end, the product will have those and we will need to test those. So it's super important for us to keep like an edge on what's coming so that we are ready when the client comes and have a specific needs for a specific technology going to be ready. Uh, so I would say, I mean for the rest we'll see, I guess, uh, we will evaluate also with how the market will go. Uh, but I would say for sure AI is one thing that we're going to see a bit more. And probably like the rest of many, many other companies, they're looking to see how AI will, uh, change a bit the industry.

Jim Beretta: Yeah, it's exciting times for sure. Hugo Lesnick, thank you so much for um, joining me today. And how can people get in touch with you? They want to learn more, they want to buy some test equipment or they want to join Averna.

Hugo Lesnik: Thank you, Jim. It's been a pleasure. So I mean all the information you can look up at the Averna website, uh, super easy. You have a ton of information about what we are capable of doing, example about client that we did, solution for them. So everything is there. It's probably the best way to know a bit more about what we were doing and also, I mean to reach the right person. If ever you have an opportunity or question, uh, we will be happy to talk to you and have a discussion about your needs.

Jim Beretta: And I will work with Amy and we'll put some of these, uh, links into uh, our show notes. So thank you again, Hugo. You might know that I run a marketing consultancy called Customer Attraction where we focus on marketing, branding strategy and content creation. We do a lot of project work and help you with your marketing challenges. And we're focused on the automation integration and robotics industries. We welcome new customers and projects and help can help you get your marketing back on track, fix your website or perform a marketing audit. I'd like to acknowledge a 3 the association for Advancing Automation. They're the leading automation trade association for robotics, vision and imaging, motion control and motors, and the industrial artificial intelligence technologies. Visit automate. Org to learn more. And if you'd like to get in touch with us at the Robot Industry podcast, you can find me Jim Beretta on LinkedIn. Today's podcast was produced by Customer Attraction Industrial Marketing. And I'd like to thank my team, Chris Gray for the music, Jeffrey Bremner for the auto production, and my business partner, Be safe out there.