From Student to Senseye Educator: Predictive Maintenance in the Classroom with Jasleen Kaur
Show notes
Welcome to the Trend Detection podcast, brought to you by Senseye Predictive Maintenance – which gives you visibility and insights into all your assets, from single machines to full plants to help you reduce downtime, increase knowledge sharing and accelerate digital transformation across your organization.
In this episode, we're joined by Jasleen Kaur, a graduate electrical and electronics engineer at Siemens Digital Industries, who shares her firsthand experience delivering industry‑led education through the Connected Curriculum initiative.
How industry and academia can work together effectively through initiatives like Connected Curriculum to close the skills gap between university education and real-world engineering roles.
The practical difference between condition monitoring and predictive maintenance, and why predictive maintenance adds real value by anticipating failures and reducing unplanned downtime.
How AI-powered tools like Senseye (and its Copilot) help beginners and professionals alike interpret machine data, troubleshoot issues, and make informed maintenance decisions using natural language.
What a real-world, hands-on predictive maintenance course looks like, including the use of synthetic data, staged learning over several weeks, and practical platform experience rather than theory alone.
Why human judgment still matters in an AI-driven workplace, and how students are taught to combine critical thinking with AI insights rather than relying blindly on automated recommendations.
You can find out more about how Senseye Predictive Maintenance can reduce unplanned downtime and contribute towards improved sustainability within your manufacturing plants, by visiting: www.siemens.com/senseye-predictive-maintenance
Connect with Jasleen on LinkedIn:
https://www.linkedin.com/in/jasleen-kaur-907391268/
Previous episodes about Connected Curriculum
https://podcasts.apple.com/ca/podcast/hands-on-with-ai-bringing-senseye-to-the-classroom/id1589803102?i=1000724963690
Transcript
And alongside um my graduate scheme um at Siemens, I got this opportunity to, you know, work with Steve Jones through the connected curriculum um to deliver um essentially a course to students regarding SensNSI, predictive maintenance, and, um, yeah, that's just a quick brief about me. Welcome to the trend detection podcast powered by Siemens. And now your host, Nel Sullivan. Hello and welcome to the latest episode of the Trend detection podcast. Um, I'm really pleased to be joined, um, by my colleague, um, Jasle Corr. Um, and we're, we're sort of carrying on a topic we've, we've covered a couple of times on the, on the podcast, and it's one that's I'm really sort of passionate about cos it's so fascinating. Um, and I, I won't go into the details cos I'm gonna let Jasleen sort of give us a bit of a reminder of it, but, um, we've done a, a few podcasts in the past around, um, an initiative called Connected Curriculum. We've done an interview with Steve Jones, who is also based in S Stevens UK who's, who's sort of running the, um, the initiative. And we, we also did a very special episode, um, from Manchester Metropolitan University as well where, um, where students are actually, in this case, um, Sensai are actually being taught, you know, the, the skills and AI and we're gonna go into that in more detail today, um, but they're being taught these skills using Sensai, which is just a fantastic initiative from my point of view and getting. People really interested in the topic, uh, as well. But, um, before we go into, in, into why we're, we're talking today, um, I'll hand over to you, Jasmine, to just give a bit of background about yourself and your involvement in, in the project as well. Sean Ile, thank you so much for the opportunity. Uh, yeah, just to introduce myself a little bit. Uh, my name's Jasleen. I'm a graduate electrical and electronics engineer, uh, working at Siemens Digital-Industries. And alongside, um, my graduate scheme, um, at Siemens, I got this opportunity to, you know, work with Steve Jones through the connected curriculum, um, to deliver, um, essentially a course to students regarding Sensai, predictive maintenance, and, um, yeah, that's just a quick brief about me. And it, and, and really, I guess at the heart of this, the connected curriculum, Jazzine from, you know, from my, my experience as well. So it really is about connecting academia and industry leaders like Siemens, of course, and educating students on, you know, new upcoming technologies. AI is obviously at the heart of that, but not just AI of course as well. Yeah. Um, so, the course that I was, have been running is the Center Predictive maintenance, which, of course, involves a range of technologies, um, including like ITOT, sorry, uh, information technology and operational technology, um, you know, integration and, you know, data handling. And this is one part of Connected curriculum, um, which is the Sensi predictive maintenance course. But of course, connected curriculum covers a wide range of, you know, bespoke, um, leading industrial technologies, um, that, you know, help students bridge this, you know, skill shortage and the gap between industry and the skills that you learn at university. Fantastic, yeah, very well summarized, I think, and yeah, it's a, it's, it's a very good reason why I'm, I'm sort of very intrigued by this topic and the 3rd podcast on the topic, but I guess we're coming at a slightly different angle. I mean, you mentioned first of all, you know, you were given the opportunity by Steve, so how did that come about first of all? Uh, yeah, so, um, as I mentioned earlier, um, I recently started my graduate scheme. And of course, at Siemens, the whole, um, there's this whole thing about, you know, uh, growth mindset, growing your skills, you know, soft skills, technical skills. And, um, I was introduced to Steve by another graduate, uh, themselves. And, you know, Steve, um, came up and was like, uh, hi, Jasleen, this is an opportunity, you know, this is an exciting opportunity where you get. Present, uh, you know, to a wide range of students, and you get to learn this, um, uh, like this whole thing about, you know, Senseii, AI, and how predictive maintenance, um, actually happens. So, yeah, um, I met Steve and then I got to know about Sensai, learned about Sensai, delved into it, you know, how, um, essentially it works, and then, of course, um, created this course essentially um about Sensai predictive maintenance and how essentially we would. Deliver these course to, you know, students, how they can learn what Sensii is, and, you know, essentially, how it's used in the industry and what predictive maintenance is. So that's how I came about, you know, um, with this opportunity and presenting to students. And I guess just to start with, I mean, not, not to make this all about, um, Sensei, but in terms of what sort of intrigued you about Sensi, you know, to, to, you know, got you sort of excited for the project, even if, you know, just hearing about, you know, being described to you and that kind of thing, what, what, what were your feelings around that? So, um. In the industry, of course you have like things such as condition monitoring. And, you know, looking at the factory floor. And, um, I knew a bit about this, um, just from, you know, reading about certain, uh, situations on the factory floor. And as I was introduced to Sensai, um, I understood that this was completely, like, the, it is the forefront of, you know, leading industry technologies, and essentially, like, what predictive maintenance is. So, and how it differs from condition monitoring. So, I learned about That and I was like, well, this is great, uh, you know, this is where I want to grow my skills, you know, learn more about um essentially how this works. So yeah, um, from that, um, I got into Sensai and then I researched it a little bit on my own, you know, essentially, um, what it does, where it has been used, any, you know, user stories and things like that. And then that's from exactly where my interest um delved more within this software itself. No, that's fantastic, that's fantastic, and what, and I guess what intrigued me about your story as well, Justine, I should say actually to, for the audience, it was actually um a, a LinkedIn post, I think you did, Jasine, that sort of drew, um, my attention to, to what you were doing, which was fantastic, but, um, in terms of that transition from a student to, um, let's say, an, an educator, you know, what, you know, what, what were your sort of motivations behind that? I think you described it a little bit there, but was there anything that, you know, that very distinct shift from learning to, you know, the learner, as it were. Um, and then you, you also described some of the, you know, creating the module, so maybe you could talk a little bit about how you put the module together, or, you know, in collaboration with others, I'm sure as well. Uh yeah. So, um, yeah, this was an exciting opportunity because the same year I graduated from university and stepped into the industry. And then, um, a few months into my placement, I am, you know, suddenly becoming an educator. Um, so, yeah, um, I think why, um, essentially, I chose this opportunity was because coming from university myself, um, and entering this industry, I realized that, The skills that you learn at university are not essentially the same that you would need in the industry, and this is something, um, the, you know, the initiative of connected curriculum covers, but something I would also like to contribute back, um, you know, after learning this sort of, um, that more skills are needed in terms of when you're entering the industry. So that's where I was like, yes, I will take this opportunity, you know, um, helping students learn more about industry relevant. Um, technologies. So, when I took this project, um, that Steve presented with, um, essentially, the students that we were going to be presenting to were, um, new to predictive maintenance. Um, essentially, they did, they didn't realize essentially what condi condition monitoring is, what would predictive maintenance be, and exactly what SensSII is. So, um, keeping that in mind, the course that we designed essentially runs from Um, it's 2 hour sessions over the course of the 3 weeks, and that was designed due to technical, um. Technical implications such as, you know, uh, taking in the data, which would take a bit of time. So, we designed the course over the 3 weeks, um, and, you know, within the first week, we kept it relevant as the students would be learning about what condition monitoring is, what predictive maintenance is, and all the, you know, the background information, you know, setting the essential background that the students need to learn and take on this, uh, course. So, from that, they were given scenarios, you know, what happens on the factory floor, um, you know, using diagrams and visuals, um, explaining to them, um, what, uh, predictive maintenance is and how it differs from condition monitoring, and giving all of that, you know, background, then delving into uh where Sensi sits, and essentially what it provides to customers. And, of course, um. Or as it runs over the 3 weeks, um, within the first week, the students were able to, you know, uh, navigate the Sensi platform, add different assets, uh, to, uh, the Sensi platform to allow for data ingestion, um, over the upcoming weeks. And within the second week, they delved more into, you know, data handling, the technical expertise of the platform. And then by the end of the 3rd week, they were able to understand uh what Sensii does, um, how, essentially, what are the benefits of the platform, and what you can get out of it, um, as a Sensii user. So that's how the 3 week course was essentially delivered. Oh, fantastic, and what, what a fantastic overview that was as well. That's incredibly clear, that's fantastic. Um, and I, I guess in that period, is it, one thing I picked out, you mentioned at the start was, you know, really identifying from the very start, the differences between condition monitoring and predictive maintenance, which, again, I think is still, uh, as predictive maintenance has become a, a a bigger and bigger term over the past 4 years I've been sort of involved in it, there's still this confusion between the two. So it's interesting that you decided to start, maybe start there and compare the two. Was there? You know, was that, you know, was that the reason why you sort of start in that area and were the students, you know, mixing up the two to start with? Maybe that was why you decided to sort of set the groundwork at the, at the beginning. Uh yeah, so, essentially, students realized that, um, of course, there are faults at the factory, you know, there's like unplanned downtimes, machines break, and then some students did know that, you know, condition monitoring is where you monitor the states of the machines. However, of course, something that they needed to learn within the course or the background is essentially that predictive maintenance is where um. The benefit where the drawbacks of condition monitoring essentially are covered. So wearing condition monitoring, you can monitor the condition of the machine and, you know, um, essentially find out how the machine, how the machine's health is. You're not really preventing any any unplanned downtimes. You're not really preventing, you know, where. The machines are broke before the machines break down. And this is where the students learn that predictive maintenance is where you're predicting, you know, how the machine is going to behave, or, you know, any failures that are going to be upcoming within, you know, the next few weeks for a machine. Um, and that's something that the students, um, had to learn as the first, you know, learning objective, essentially. And I guess to, to continue sort of, you know, those, those sort of learning outcomes, you know, as, as the sort of course progressed, what were, you know, if you think from a student point of view, what were their sort of, you know, oh my God, that's amazing moments, you know, or, you know, maybe it wasn't that exaggerated, but you know, moments where they're like really sort of like, oh wow, this, this is completely different to what, what I thought it was, or where their mindset sort of changed. Can you identify those kind of. Yes, um, so exactly, the amazing moment was, you know, how when they were using the platform, they realized how AI can, you know, help them, you know, understand these machines, you know, um, using natural language, they are understanding. How the machines are behaving, troubleshooting, you know, the data or the machine health that is currently happening. And that's exactly like the, at the end of the, um, so the 3rd week within the course, they were, you know, analyzing the data and the machines that they had on the platform. And from that, when they, you know, integrated with the copilot Sensai copilot, they were like, wow, this is amazing, you know, I can finally understand what's happening, like, They were shocked that even as beginners into this field, they were like, wow, I can, you know, understand what's happening. I can troubleshoot, and, you know, this amazing technology as it is. And that's interesting, you, it's interesting you bring up, um, the co-pilot as well, cos, you know, even, even from my perspective, I mean, Sensai's been a, a brilliant product, but for, you know, pre, um, implementation of that technology, it's always obviously been there, it's been with the growth of LLMs, but for someone like me, it's helped to understand and use the product much easier. So I guess from a, let's say a, a teaching perspective, did it assist with that kind of thing, help them understand the concept, uh, a lot more, would you say, Jesse? Yeah, so, um, of course, um, as an educator myself, I set the background for them and then helping, so as much as I can explain to them the benefits of SENT, it's great that they get to see this um live, that, you know, rather And me just, you know, going on about, uh, going on at them, like, this is the benefit, this is Sensai, this is great. Um, they get to see it in person and, you know, use this copilot to like, help them even like delve further into the problems, or like, you know, the fictitious scenario that we used, um, essentially. And actually that was something we didn't, we didn't mention because something we've mentioned on the podcast with, um, with Steve and, and the other podcasts is, is the use of sort of synthetic data, right? That's an important thing to sort of maybe point out. I mean, what, why was that? I mean, maybe we can remind the audience of why, um, the project was sort of approached in that way using synthetic instead of live data. Uh, yeah, so, of course, using live data can have, you know, security issues and it's not the ideal solution. So, the best way we as a team decided to deliver this course was to use synthetic data based on, you know, rep that replicates the faults that actually happened on the factory floor. So that Way students can understand, you know, get a perspective into what troubles that can happen on a factory floor, but also, you know, safeguarding the use of, you know, factory data or, you know, using rather than using live data, of course, using this synthetic data. Yeah, and, and maybe, you know, cos we talk, talk about the security aspect, but I think it's actually the, you know, the, cos you, cos obviously you're not on a factory floor, right, it's, it's something that's been set, you know, cos if you wanted to go down that route, it'd have to be running constantly, um, in, you know, which isn't practical even if you stop it and start it, that's not, you know, it doesn't work how Sensi would work in real life cos it would be. Etc. So, so I, I guess that's an important point to point out as well. Um, yeah, exactly. So, um, even with the factory, like if we use live data, we would of course have to, you know, collaborate with the factory, get the sensors running, and of course, have some sort of confident confidentiality or cybersecurity set up, um, in that sense. And of course, if we wanted. To deliver this, um, course to a wide range, um, of audiences. So this could be, you know, students or at a certain university or a different university, um, having synthetic data essentially helps, you know, safeguard or, you know, makes it a bit, uh, much easier to deliver the course itself. And I guess we go back to, you know, the learnings of the students. So, of course, you know, and I think you, you sort of said it right a bit earlier as well that there's so, there's only so much you can learn by reading and, I mean, uh, and of course lectures and stuff, you do learn a lot, but until you, like, you, the practicality of the course, I guess is where its real strength, um, lies. But, uh, from your point of view, Jasmine, how does that translate into those sort of first roles in, in manufacturing environments, on shop floor environments? What, what are the key sort of takeaways for students? Uh, yeah. So, um, I think, as I mentioned, from the first week, the students essentially are able to navigate the platform, and they've got the whole session is designed to be as much, uh, as practical as possible. And from this, um, over the course of the three weeks, they learn a wide range of skills, um, that correlates. You know, entry-level roles, such as, you know, they're learning about data handling, you know, how the data connects from an operational technology level to a, you know, information technology level. And then connecting that into Sensii and how, um, you know, to handle that data. So it's optimized for the LLM. And, you know, from that, then, troubleshooting to make decisions. And of course, um, with the rise of AI they can use AI but they need to learn that, you know, making these decisions um for maintenance engineers, or as uh interpreting the machine health, you can rely on the AI outputs, but at the end of the day, it depends on you as the user to, you know, decide which is the most relevant, um. You know, relevant suggestion of the AI or whether the AI suggestions are accurate, or, you know, essentially that. And along with that, um, they also do a bit of block programming, so. Not only just, you know, being handling data in terms of like, inspecting it, where, how it's behaving, but also filtering, uh, the data to, you know, um, essentially be optimized, and, um, and then from that block programming, they then, of course, um, are able to learn how to expand, expand those data handling skills, not only just on the Sensa platform, but which comes used in other, um, scenarios as well. And it, and it's interesting that you, I mean, you're, you're, you're so right, even in real life projects with, you know, global manufacturers, it's the same thing, right? You know, that you can't, the AI can, you know, the, the, the platform can produce insights and guidance, but if no one takes action for or believes in it, then it, it's not gonna prove its value, which, uh, which again, I think is quite important. And I guess on that sort of note, Jasle, would you say that it, you know, maybe at the early stage of the course or during the course, was there sort of misconceptions about predicting maintenance that it would be able to provide, you know, exact analysis on the exact day that it, something's going to fail, for example. I mean, that's something we often hear, um, with customers sometimes, but, um, is there anything like that you can, you can sort of remember? Um, so, in terms of the course itself, um. I think we have to keep in mind these students are not experts, and what they have is a basic understanding of the platform itself. …