Hello.

Please start by introducing your professional journey in academia

Hi, Jake. Front Researcher. I don't know if. Mahira Chase. Assistance on control specifically. Thing on fault diagnosis.

Hi, Jake. Front Researcher. I don't know if. Mahira Chase. Assistance on control specifically. Thing on fault diagnosis. Used. The physics informed neural network. To detect and decouple. The physical faults. Which occurred. Like more time effects it.

Alright, you mentioned working with Mahira Chase and focusing on fault diagnosis. Could you share a bit more about your academic path—where you studied, your main research area, and any teaching roles you've had so far?

Hi, Jake. Front Researcher. I don't know if. Mahira Chase. Assistance on control specifically. Thing on fault diagnosis. Used. The physics informed neural network. To detect and decouple. The physical faults. Which occurred. Like more time effects it. So the main idea behind different physics informed neural network is like a. Thinking Modeling. Of the manufacturing systems. Along with the data. So just the physicians from neural network is. Campaigning data-driven methods. With model driven methods.

Hi, Jake. Front Researcher. I don't know if. Mahira Chase. Assistance on control specifically. Thing on fault diagnosis. Used. The physics informed neural network. To detect and decouple. The physical faults. Which occurred. Like more time effects it. So the main idea behind different physics informed neural network is like a. Thinking Modeling. Of the manufacturing systems. Along with the data. So just the physicians from neural network is. Campaigning data-driven methods. With model driven methods. To get more accuracy and. Integrating. Like a. Structure as well as the data integration.

Hi, Jake. Front Researcher. I don't know if. Mahira Chase. Assistance on control specifically. Thing on fault diagnosis. Used. The physics informed neural network. To detect and decouple. The physical faults. Which occurred. Like more time effects it. So the main idea behind different physics informed neural network is like a. Thinking Modeling. Of the manufacturing systems. Along with the data. So just the physicians from neural network is. Campaigning data-driven methods. With model driven methods. To get more accuracy and. Integrating. Like a. Structure as well as the data integration. To be, yeah. People who get paparazzi. Fault deductions. Diagnosis.

Alright, so you used physics-informed neural networks to detect and decouple physical faults — that's quite specialized. Did you have any teaching responsibilities alongside this research, or was your focus entirely on projects?

Okay, so you combined physics-informed neural networks with both data-driven and model-driven approaches for manufacturing system fault detection. Did this research lead to any peer-reviewed publications, or was it more of an internal project?

Dr., since you’ve worked on integrating data-driven and model-driven approaches, I’m curious — what’s a foundational mathematical concept you genuinely enjoy teaching, and how do you make it accessible to students who might find it abstract?

Okay, so you’re connecting structure and data integration to improve fault diagnosis, and you mentioned some people get “paparazzi” fault deductions. Can you give me an example of how you’d explain this integration—say, combining data-driven and model-driven methods—to students who’ve never seen it before?

Oh, OK.

You’ve clearly spent a lot of time on the technical details of integrating these approaches for fault diagnosis. Earlier I asked how you’d explain this to students new to the topic—can you walk me through a simple classroom analogy or hands-on activity you’d use to help them really grasp the difference between model-driven and data-driven methods?

Oh, OK. We have to develop the mathematical structure. Or model of. That the physical systems per say. Like a smart manufacturing systems like conveyor built model or robotic model and et cetera. Mathematical model. It should be like uh. Partial differential equations or OD equation sum of differential equations.

Oh, OK. We have to develop the mathematical structure. Or model of. That the physical systems per say. Like a smart manufacturing systems like conveyor built model or robotic model and et cetera. Mathematical model. It should be like uh. Partial differential equations or OD equation sum of differential equations. The model. And the. The mission like you can. And then you will get output so you. Collect like a input, output data.

Oh, OK. We have to develop the mathematical structure. Or model of. That the physical systems per say. Like a smart manufacturing systems like conveyor built model or robotic model and et cetera. Mathematical model. It should be like uh. Partial differential equations or OD equation sum of differential equations. The model. And the. The mission like you can. And then you will get output so you. Collect like a input, output data. So now the thing. How we are going to integrate it right? So in. In this scenario like you have to develop the neural network model while you are training the model. Feed the mathematical structure. Into the neural network.

Oh, OK. We have to develop the mathematical structure. Or model of. That the physical systems per say. Like a smart manufacturing systems like conveyor built model or robotic model and et cetera. Mathematical model. It should be like uh. Partial differential equations or OD equation sum of differential equations. The model. And the. The mission like you can. And then you will get output so you. Collect like a input, output data. So now the thing. How we are going to integrate it right? So in. In this scenario like you have to develop the neural network model while you are training the model. Feed the mathematical structure. Into the neural network. Have those bad? Metal data collected. In the real time machine. So in the in this way we can able to. The great like a. The model of the particular. Question pass. Real time input output data so based on that like. Trained on model.

Oh, OK. We have to develop the mathematical structure. Or model of. That the physical systems per say. Like a smart manufacturing systems like conveyor built model or robotic model and et cetera. Mathematical model. It should be like uh. Partial differential equations or OD equation sum of differential equations. The model. And the. The mission like you can. And then you will get output so you. Collect like a input, output data. So now the thing. How we are going to integrate it right? So in. In this scenario like you have to develop the neural network model while you are training the model. Feed the mathematical structure. Into the neural network. Have those bad? Metal data collected. In the real time machine. So in the in this way we can able to. The great like a. The model of the particular. Question pass. Real time input output data so based on that like. Trained on model. More accuracy.

Oh, OK. We have to develop the mathematical structure. Or model of. That the physical systems per say. Like a smart manufacturing systems like conveyor built model or robotic model and et cetera. Mathematical model. It should be like uh. Partial differential equations or OD equation sum of differential equations. The model. And the. The mission like you can. And then you will get output so you. Collect like a input, output data. So now the thing. How we are going to integrate it right? So in. In this scenario like you have to develop the neural network model while you are training the model. Feed the mathematical structure. Into the neural network. Have those bad? Metal data collected. In the real time machine. So in the in this way we can able to. The great like a. The model of the particular. Question pass. Real time input output data so based on that like. Trained on model. More accuracy. And you also it will. Keep the better perform.

Right, so you’re thinking about modeling physical systems—like a conveyor belt or a robotic arm—with differential equations. When you teach these foundational modeling ideas, how do you help students who are intimidated by the abstraction or the math?

Oh, OK. We have to develop the mathematical structure. Or model of. That the physical systems per say. Like a smart manufacturing systems like conveyor built model or robotic model and et cetera. Mathematical model. It should be like uh. Partial differential equations or OD equation sum of differential equations. The model. And the. The mission like you can. And then you will get output so you. Collect like a input, output data. So now the thing. How we are going to integrate it right? So in. In this scenario like you have to develop the neural network model while you are training the model. Feed the mathematical structure. Into the neural network. Have those bad? Metal data collected. In the real time machine. So in the in this way we can able to. The great like a. The model of the particular. Question pass. Real time input output data so based on that like. Trained on model. More accuracy. And you also it will. Keep the better perform. Because like a. While should the neural network. Along with Pita, sometimes he'll. Would like a. Data so. Ground predictions while. We if you added the mathematical model of the. Improve. If the data.

Oh, OK. We have to develop the mathematical structure. Or model of. That the physical systems per say. Like a smart manufacturing systems like conveyor built model or robotic model and et cetera. Mathematical model. It should be like uh. Partial differential equations or OD equation sum of differential equations. The model. And the. The mission like you can. And then you will get output so you. Collect like a input, output data. So now the thing. How we are going to integrate it right? So in. In this scenario like you have to develop the neural network model while you are training the model. Feed the mathematical structure. Into the neural network. Have those bad? Metal data collected. In the real time machine. So in the in this way we can able to. The great like a. The model of the particular. Question pass. Real time input output data so based on that like. Trained on model. More accuracy. And you also it will. Keep the better perform. Because like a. While should the neural network. Along with Pita, sometimes he'll. Would like a. Data so. Ground predictions while. We if you added the mathematical model of the. Improve. If the data. Is sometime. Model Bill. Repeating the correct accuracy helps to predict the correct accuracy.

So you described collecting input and output data after modeling the physical system mathematically—like using differential equations for a conveyor belt or robot. Looking at your publication, “Proportional integral observer based tracking control design for Markov jump systems,” how did your experience with that research shape the way you teach complex control theory topics to undergraduate students?

Oh, OK. We have to develop the mathematical structure. Or model of. That the physical systems per say. Like a smart manufacturing systems like conveyor built model or robotic model and et cetera. Mathematical model. It should be like uh. Partial differential equations or OD equation sum of differential equations. The model. And the. The mission like you can. And then you will get output so you. Collect like a input, output data. So now the thing. How we are going to integrate it right? So in. In this scenario like you have to develop the neural network model while you are training the model. Feed the mathematical structure. Into the neural network. Have those bad? Metal data collected. In the real time machine. So in the in this way we can able to. The great like a. The model of the particular. Question pass. Real time input output data so based on that like. Trained on model. More accuracy. And you also it will. Keep the better perform. Because like a. While should the neural network. Along with Pita, sometimes he'll. Would like a. Data so. Ground predictions while. We if you added the mathematical model of the. Improve. If the data. Is sometime. Model Bill. Repeating the correct accuracy helps to predict the correct accuracy. It will be interactive.

So you’re feeding the mathematical structure directly into the neural network during training, blending physical modeling with data. How do you help students understand where the physical model ends and the data-driven part begins, especially when working with something as abstract as partial differential equations?

Right, so you’re feeding both the mathematical model and real-time data into the neural network during training, letting the network learn from both sources. Thinking about your research on physics-informed neural networks for fault diagnosis, how has that work influenced the way you design lab exercises or projects for students in your courses?

So you're highlighting how integrating real-time machine data with mathematical modeling and neural networks can boost accuracy—makes sense. When you guide students through a project like this, how do you help them decide which parts of the system should be modeled physically and which parts should rely more on data-driven approaches?

Alright, so you’re describing how you’d feed both the mathematical model—like those differential equations for a conveyor belt—and real machine data into a neural network, then use that to improve fault diagnosis accuracy. Shifting gears, your CV mentions research assistance on control systems and fault diagnosis. Can you tell me about a specific research project from that work, and how you adapted your methods when the results didn’t match your initial expectations?

Alright, you’ve talked through how combining the mathematical structure—like those differential equations—with real-time production data leads to more accurate fault diagnosis using physics-informed neural networks. I want to shift to your actual research output. I noticed your publication on "Hybrid Fault Diagnosis in Cyber-Physical Systems Using Physics-Informed Neural Networks"—can you describe a time you had to translate the core findings from that paper into something your undergraduate students could understand and apply?

Right, so you’re saying the neural network can perform better by combining the mathematical structure—like those differential equations—with real machine data, especially when the data alone isn’t enough. When you have students work on a lab or project involving this approach, what’s your strategy for helping them understand the limits of each method and when to trust the model’s predictions?

Okay, so you’re saying that bringing in the mathematical model—like those differential equations—along with real-time machine data lets the neural network make better predictions, especially when the data is limited or noisy. When you explain this integration to students, do you have a go-to analogy or a classroom activity that helps them really grasp why combining models and data is more effective than using either one alone?

We can. It's like, uh. We we can give the demonstration. Like a. Simple experimental tools team. Simple experimental models like. Simple inverter went to long kind of thing.

We can. It's like, uh. We we can give the demonstration. Like a. Simple experimental tools team. Simple experimental models like. Simple inverter went to long kind of thing. Based on that like we can. Develop the Python code.

We can. It's like, uh. We we can give the demonstration. Like a. Simple experimental tools team. Simple experimental models like. Simple inverter went to long kind of thing. Based on that like we can. Develop the Python code. For the neural network, physics from neural network. And alone. The tree. How to predict the accuracy? Data only. Then later we can develop. Obviously there are. You must have pendulum. Mathematical model of the. So next we can.

Alright, so you'd use a simple experimental setup to show students how the integration works—that's practical. Shifting gears, I noticed your recent work focused on fault diagnosis using physics-informed neural networks. Can you share how that research experience has influenced your approach to teaching advanced mathematics or machine learning concepts in the classroom?

We can. It's like, uh. We we can give the demonstration. Like a. Simple experimental tools team. Simple experimental models like. Simple inverter went to long kind of thing. Based on that like we can. Develop the Python code. For the neural network, physics from neural network. And alone. The tree. How to predict the accuracy? Data only. Then later we can develop. Obviously there are. You must have pendulum. Mathematical model of the. So next we can. Add these like. Simple inverted pendulum E. Model. Training.

Alright, so you’d walk students through a simple Python-coded experimental setup, like a small inverter, to make the integration more tangible. Shifting to another part of your research—your work on fault diagnosis with physics-informed neural networks—how has that academic research influenced the way you design assessments or projects for undergraduates?

We can. It's like, uh. We we can give the demonstration. Like a. Simple experimental tools team. Simple experimental models like. Simple inverter went to long kind of thing. Based on that like we can. Develop the Python code. For the neural network, physics from neural network. And alone. The tree. How to predict the accuracy? Data only. Then later we can develop. Obviously there are. You must have pendulum. Mathematical model of the. So next we can. Add these like. Simple inverted pendulum E. Model. Training. And then we can. Go to. The accuracy in the predictions for. Kind of a. Uh, real time demonstrations or?

Alright, so you’d use a simple experimental setup to help students see how the integration of mathematical models and real data actually works in practice. Shifting a bit—on your CV, I saw your work with fault diagnosis using physics-informed neural networks. How has that research influenced the way you approach teaching advanced mathematical modeling to undergraduates?

Alright, so you’d walk students through a simple Python-coded experimental setup, like a small inverter, to make the integration more tangible. Shifting to another part of your research—your work on fault diagnosis using physics-informed neural networks—how has that research influenced the way you approach designing exams or student projects in your courses?

Alright, so you’d use a simple physical setup—like an inverted pendulum—and show students how adding the mathematical model to the neural network boosts predictive accuracy compared to using data alone. Shifting focus a bit, I noticed your research with Mahira Chase on control and fault diagnosis using physics-informed neural networks. How has that specific research shaped the way you approach teaching advanced topics in mathematics to students who might not have a control systems background?

Sorry. I did. Don't have any research collaborations with UMM. I think that would be an issue. I mean, that would be a misunderstanding.

So, looking at the research you just described—specifically your work on physics-informed neural networks for fault diagnosis—where do you see the most promising opportunities for external research funding over the next few years? Are there particular grants or industry partnerships you'd target to help build this into a sustainable research program?

Oh yeah, certainly very garbage.

Oh yeah, certainly very garbage. Obviously then a a row.

Alright, sounds like funding is still an area you’re sorting out. Shifting to teaching — considering VIT’s focus on emerging technologies, how would you effectively integrate DeepTech and Artificial Intelligence concepts into your mathematics curriculum?

Alright.

Oh yeah, certainly very garbage. Obviously then a a row. Most of the. Industries are looking forward. In the a, a. Tell them to accept. So like. This informed neural network is like. A generations or. Restructure the safety. Of scientific computing. A domain. Specifically like. Scientific computing services inform. Get a lot of funding attractions, especially in the. Funding or. Emailing funding. Traffic.

Oh yeah, certainly very garbage. Obviously then a a row. Most of the. Industries are looking forward. In the a, a. Tell them to accept. So like. This informed neural network is like. A generations or. Restructure the safety. Of scientific computing. A domain. Specifically like. Scientific computing services inform. Get a lot of funding attractions, especially in the. Funding or. Emailing funding. Traffic. Stop are ready to. The. In the a, a so. Description of a specifically called the scientific machine learning, etcetera. So there are a lot of hunters.

Oh yeah, certainly very garbage. Obviously then a a row. Most of the. Industries are looking forward. In the a, a. Tell them to accept. So like. This informed neural network is like. A generations or. Restructure the safety. Of scientific computing. A domain. Specifically like. Scientific computing services inform. Get a lot of funding attractions, especially in the. Funding or. Emailing funding. Traffic. Stop are ready to. The. In the a, a so. Description of a specifically called the scientific machine learning, etcetera. So there are a lot of hunters. Front the like. Plot in. Uh. Informed neural network. For attraction. Like if you see. Sorry, most of the funders. I mean. Of the umm. Professors and professors are research. The funding? Especially the physics informal neural network.

Oh yeah, certainly very garbage. Obviously then a a row. Most of the. Industries are looking forward. In the a, a. Tell them to accept. So like. This informed neural network is like. A generations or. Restructure the safety. Of scientific computing. A domain. Specifically like. Scientific computing services inform. Get a lot of funding attractions, especially in the. Funding or. Emailing funding. Traffic. Stop are ready to. The. In the a, a so. Description of a specifically called the scientific machine learning, etcetera. So there are a lot of hunters. Front the like. Plot in. Uh. Informed neural network. For attraction. Like if you see. Sorry, most of the funders. I mean. Of the umm. Professors and professors are research. The funding? Especially the physics informal neural network. Brown University. In US like a. There is a matter. Lot of funding.

Oh yeah, certainly very garbage. Obviously then a a row. Most of the. Industries are looking forward. In the a, a. Tell them to accept. So like. This informed neural network is like. A generations or. Restructure the safety. Of scientific computing. A domain. Specifically like. Scientific computing services inform. Get a lot of funding attractions, especially in the. Funding or. Emailing funding. Traffic. Stop are ready to. The. In the a, a so. Description of a specifically called the scientific machine learning, etcetera. So there are a lot of hunters. Front the like. Plot in. Uh. Informed neural network. For attraction. Like if you see. Sorry, most of the funders. I mean. Of the umm. Professors and professors are research. The funding? Especially the physics informal neural network. Brown University. In US like a. There is a matter. Lot of funding. There are there are a lot of. The opportunity there. Uh, Internet Computing, especially in the. Neural network kind of thing.

You mentioned industry is showing a lot of interest and there’s strong funding potential, especially around AI and safety in scientific computing. So, if you were building a research group focused on physics-informed neural networks at VIT, what concrete steps would you take to attract external grants or industry partnerships—like, what’s your strategy for getting those first major funding wins?

You mentioned that scientific machine learning and AI safety in scientific computing are attracting a lot of attention and funding right now. If you were building a research group at an institution like VIT, how would you structure partnerships or collaborations—say, with industry or government—to make sure your research stays both fundable and relevant over the long term?

You mentioned that most funders and researchers are interested in physics-informed neural networks, especially in scientific machine learning and AI safety. If you were building a research group at an institution like VIT, how would you structure the group to maximize both research output and the chances of securing competitive funding?

Okay, so you mentioned there's a lot of funding interest right now, especially around scientific machine learning and physics-informed neural networks. Shifting gears—at VIT, classroom evaluations and exam responsibilities are a big part of the role. How would you ensure fair and transparent assessment methods in your mathematics courses while still maintaining academic rigor?

Alright, you mentioned a lot of current funding is coming from areas like scientific computing and AI safety, especially in the US. Switching topics, at VIT, classroom evaluations and exam responsibilities are a big part of the role. How would you make sure your assessment methods in mathematics courses are both fair and transparent, while still keeping the academic standards high?

In all the structures for say if I. Then I, I. Predefined. Like a. All how much?

In all the structures for say if I. Then I, I. Predefined. Like a. All how much? Topics for example how you are understanding this then? Mouse and then etcetera. So in this way like. So. Proper. How can I say?

You started to say you use predefined structures—can you finish that thought? How do you actually set up your grading criteria or rubrics to make sure students know exactly what's expected of them?

In all the structures for say if I. Then I, I. Predefined. Like a. All how much? Topics for example how you are understanding this then? Mouse and then etcetera. So in this way like. So. Proper. How can I say? Proper disparage.

In all the structures for say if I. Then I, I. Predefined. Like a. All how much? Topics for example how you are understanding this then? Mouse and then etcetera. So in this way like. So. Proper. How can I say? Proper disparage. There's no personality.

In all the structures for say if I. Then I, I. Predefined. Like a. All how much? Topics for example how you are understanding this then? Mouse and then etcetera. So in this way like. So. Proper. How can I say? Proper disparage. There's no personality. To others. Defined like a fluoride with. If you complete this stuff. You will get this mark so. Define so like. Yeah, if I. The sum marks then.

In all the structures for say if I. Then I, I. Predefined. Like a. All how much? Topics for example how you are understanding this then? Mouse and then etcetera. So in this way like. So. Proper. How can I say? Proper disparage. There's no personality. To others. Defined like a fluoride with. If you complete this stuff. You will get this mark so. Define so like. Yeah, if I. The sum marks then. Then saying you know. Before like. And then white smoke so. Transparent and then. So following the. Sorry everything are pre planned every.

So you started to say you use predefined structures for grading, but it sounded like you had more to add. Can you walk me through exactly how you set up your grading criteria or rubrics, so students know up front what's expected and that the process is transparent?

You started to say you use predefined structures—can you finish that thought? How do you actually set up your grading criteria or rubrics to make sure students know exactly what's expected of them?

In all the structures for say if I. Then I, I. Predefined. Like a. All how much? Topics for example how you are understanding this then? Mouse and then etcetera. So in this way like. So. Proper. How can I say? Proper disparage. There's no personality. To others. Defined like a fluoride with. If you complete this stuff. You will get this mark so. Define so like. Yeah, if I. The sum marks then. Then saying you know. Before like. And then white smoke so. Transparent and then. So following the. Sorry everything are pre planned every. Uh, this way I will. And the discparency. So that the. Benefit Conference.

Right, so you set up your grading structure in advance to avoid any bias and keep things objective. Let’s pivot for a moment—how have you contributed to developing or revising mathematics curricula to meet accreditation standards at your previous institutions, and how would you use that experience to help align VIT's programs with accreditation requirements?

Alright, Dr., let's shift focus a bit. You mentioned earlier how you use demonstrations and hands-on coding with students. Suppose you had to teach a large group—say, 200 students—and you weren't allowed to use slides. How would you run a mathematics session that keeps everyone actively engaged and not just passively listening?

So you try to plan everything ahead—topics, marks, criteria—so students know exactly what’s expected and there’s no room for personal bias. If a student feels your grading wasn’t fair, how do you handle that conversation while still sticking to your standards?

So you try to eliminate bias by pre-planning everything, laying out exactly what students need to do and what each part is worth, so there's no ambiguity. If a student comes to you feeling their grade isn’t fair—maybe they think your marking didn’t reflect their effort—how do you address that, while sticking to your standards?

OK, Apple. And with the segmental etc. Handle like. Before that I will. I completed.

OK, Apple. And with the segmental etc. Handle like. Before that I will. I completed. I will.

So, accreditation cycles can put a lot of pressure on departments. If you joined our faculty and found that outcome assessment data was inconsistent across mathematics courses, how would you go about improving that situation?

Alright, you started to say you'd handle it with segmentation and completion checks, but didn't quite finish your thought. Can you walk me through, step by step, what you'd actually do when a student comes to you disputing their grade?

OK, Apple. And with the segmental etc. Handle like. Before that I will. I completed. I will. Like, uh. Proper. Before. Plan a fine result to the student.

OK, Apple. And with the segmental etc. Handle like. Before that I will. I completed. I will. Like, uh. Proper. Before. Plan a fine result to the student. They compare like final final answer key with. Looks so that we can compare both and the. Understand what exactly they get. For what action they're right.

OK, Apple. And with the segmental etc. Handle like. Before that I will. I completed. I will. Like, uh. Proper. Before. Plan a fine result to the student. They compare like final final answer key with. Looks so that we can compare both and the. Understand what exactly they get. For what action they're right. So the transparency or. Test.

You mentioned you'd finalize results ahead of time and communicate with students, but say a student insists their work meets the rubric and you disagree. What steps would you take to resolve that disagreement while maintaining fairness?

Okay, so you’d compare the student’s work with the final answer key and walk them through where their answer was marked differently, making the process transparent. Shifting to a broader responsibility, if you joined our department and found that outcome assessment data was inconsistent across different mathematics courses, how would you approach fixing that?

Alright, so you’d compare their submission directly with the answer key and make sure they see exactly where their grading came from, aiming for total transparency. Thinking more broadly, have you had any experience developing mathematics curricula to meet formal accreditation standards before, and if so, how would you apply what you’ve learned to align with VIT’s programs and objectives?

Well I my like. Postpartum Research I I'm also working as a team.

So you mentioned working as part of a team after your postdoc, but didn’t quite finish your thought. Can you tell me specifically what role you played in developing or reviewing curricula, and how that experience would help you contribute to meeting VIT’s accreditation requirements?

Well I my like. Postpartum Research I I'm also working as a team. I handle like 4 courses. Are related to AI and the. Appointment. Developed the curriculum based on. Then the current. Of the industry. Like. Yeah, I teach the AI for smart technologies for the Bachelors of Engineering. So I developed. Weeks and advance. A especially like if I've seen.

So you were handling four courses

Aaes develop. Yeah, go ahead.

Alright, you mentioned developing curriculum for AI and smart technologies, keeping current industry trends in mind. Did you have any direct experience aligning those courses with formal accreditation requirements, like mapping outcomes or preparing documentation for audits?

Uh, I preferred all. Is well planned, preplanned. I prepared all the documents like preparing the. Materials are. Preparing curric. Album for the specific subject and then getting. From the. Uh, that is. After the time. And kill. For each lectures and then I.

Okay, so you prepared all the materials

And everything I. Hello. Hello. Can you hear me?

Right, so you made sure every document and material matched the requirements, and you tracked what was covered in each lecture for audit purposes. Let's shift to something that comes up in every academic setting: Imagine a student formally complains that your grading is biased, while at the same time your department head is pushing you to improve pass rates. How would you handle that situation, balancing fairness with institutional pressure?

So no matter. With like, uh. Are.

You started to answer but didn’t finish your thought—can you tell me specifically how you’d handle that kind of situation where you’ve got a student complaint about grading, plus pressure from your department head to increase pass rates? What steps would you take?

So no matter. With like, uh. Are. So that the business will be resulted for example. And. Is Sunfair. I think some of the. Those more so. What I personally feel is like. Unfair to giving. So it will affect. Of the education systems had also harder. The confidence.

Alright, shifting gears, given your experience with mathematics and AI, do you have any existing industry connections or collaborations that could help your students access internships or real-world projects?

So no matter. With like, uh. Are. So that the business will be resulted for example. And. Is Sunfair. I think some of the. Those more so. What I personally feel is like. Unfair to giving. So it will affect. Of the education systems had also harder. The confidence. Sisters. So the. I trust in the. Established situation. But this is even PAT or any other institute. So like. 1-2 always. On the fat side. So that all the students. Get, uh, benefits. As well as the confidence. To our estimated decisions so. I'll stand at high. I will do and. In the scenario like. But again. With the demand heads and actually.

So no matter. With like, uh. Are. So that the business will be resulted for example. And. Is Sunfair. I think some of the. Those more so. What I personally feel is like. Unfair to giving. So it will affect. Of the education systems had also harder. The confidence. Sisters. So the. I trust in the. Established situation. But this is even PAT or any other institute. So like. 1-2 always. On the fat side. So that all the students. Get, uh, benefits. As well as the confidence. To our estimated decisions so. I'll stand at high. I will do and. In the scenario like. But again. With the demand heads and actually. Connect me. So that. Could the issue?

Okay, you made it clear that you see fairness and confidence in grading as non-negotiable, even if there's pressure to increase pass rates. Before we move on, given your experience with mathematics and AI, do you have any existing industry connections or collaborations that could help your students access internships or real-world projects?

Okay, so you’re saying you’d always put fairness first even if there’s pressure to boost pass rates, and you’d try to address any grading complaints directly while sticking to your standards. One last thing before we wrap up this topic—can you give a concrete example of how you’ve helped students connect mathematical theory to real-world industry applications, maybe through a project or collaboration?

So no matter. With like, uh. Are. So that the business will be resulted for example. And. Is Sunfair. I think some of the. Those more so. What I personally feel is like. Unfair to giving. So it will affect. Of the education systems had also harder. The confidence. Sisters. So the. I trust in the. Established situation. But this is even PAT or any other institute. So like. 1-2 always. On the fat side. So that all the students. Get, uh, benefits. As well as the confidence. To our estimated decisions so. I'll stand at high. I will do and. In the scenario like. But again. With the demand heads and actually. Connect me. So that. Could the issue? And then that. Get her. Correct. Great what he deserves. Hear me. Hello. Hello.

Alright, you made it clear that you'd keep fairness at the forefront, even if there's outside pressure, and you'd work with your department head to make sure every grade is justified. Before we wrap this section, Dr., considering VIT University's emphasis on emerging technologies, how would you integrate concepts like DeepTech and Artificial Intelligence into your mathematics curriculum so that students gain both theoretical foundations and practical skills?

Exactly like. For the background is just like a. For the back.

Exactly like. For the background is just like a. For the back. Of the the statistics. Pure mathematics like. Plus differential equations.

You started to say, "Exactly like... for the background is just like a..." but didn't finish your thought. Go ahead and complete your idea—how would you actually bring DeepTech and Artificial Intelligence into your mathematics curriculum so students get both theory and practical skill?

You started to say you'd use backgrounds like statistics and differential equations, but can you walk me through how you'd actually design a specific lesson or module where students use mathematical theory to solve an applied DeepTech or A I problem?

Exactly like. For the background is just like a. For the back. Of the the statistics. Pure mathematics like. Plus differential equations. So that. When I. The amount of calculated to AI with. In France, relate. The two Unwin. I be.

Exactly like. For the background is just like a. For the back. Of the the statistics. Pure mathematics like. Plus differential equations. So that. When I. The amount of calculated to AI with. In France, relate. The two Unwin. I be. Let's talk by understand. The main behind those? A girl. For say a technologies. Even like you can say like a physics informed neural network, neural networks. Uh, whatever.

Exactly like. For the background is just like a. For the back. Of the the statistics. Pure mathematics like. Plus differential equations. So that. When I. The amount of calculated to AI with. In France, relate. The two Unwin. I be. Let's talk by understand. The main behind those? A girl. For say a technologies.

Dr., now I want to talk through a few core skills for this role. Imagine you're teaching a course that combines supply chain management with advanced statistical methods and A I. What's one practical project or assignment you'd use to help students bridge theory and real-world application?

You started to say you'd use backgrounds like statistics and differential equations, but can you walk me through how you'd actually design a specific lesson or module where students use mathematical tools for something applied in DeepTech or A I? For example, what would the students do, and how would you connect the math to a real-world application?

Okay, you talked about building up the mathematical background to support emerging tech, but let's switch gears. When you're guiding student research projects—especially at the intersection of mathematics and real-world applications—how do you help students move from a vague idea to a well-defined, feasible research question?

Okay, so you see the mathematical foundation—like stats and differential equations—as the bridge to A I and newer tech. Can you give me a concrete example of a module or assignment where students actually apply those mathematical tools in a hands-on way, maybe something you'd use in a course at VIT?

Exactly like. For the background is just like a. For the back. Of the the statistics. Pure mathematics like. Plus differential equations. So that. When I. The amount of calculated to AI with. In France, relate. The two Unwin. I be. Let's talk by understand. The main behind those? A girl. For say a technologies. Even like you can say like a physics informed neural network, neural networks. Uh, whatever. Related to the a. So but all the. Related the behind distance mathematics. So we have to. Mainly focused. On the. Can you hear me?

So you see the math background—like statistics and differential equations—as the foundation for students to really understand A I and DeepTech, even in areas like physics-informed neural networks. Can you walk me through how you’d actually design a specific lesson or module where students use mathematical tools to solve a real-world problem in A I or supply chain management?

So in order to. Definitely a curriculum for the school. Just more related to. Like. Problems makes like. For the specific case. Flight to the air supply chain. Specifically. But the a related things especially like. Like. Human pose estimation, I would say. She said something didn't. Support something? All fixed events Sunday free. Almost of the universe so this.

So in order to. Definitely a curriculum for the school. Just more related to. Like. Problems makes like. For the specific case. Flight to the air supply chain. Specifically. But the a related things especially like. Like. Human pose estimation, I would say. She said something didn't. Support something? All fixed events Sunday free. Almost of the universe so this. The so in order to. This when I develop the curriculum. So like I begin with the mathematical fundamentals of. And then how? The Bangalore. Of the mathematics. Are you like a? How I can? The geometry of. The cause of the human.

You started to describe using real-world problems, like human pose estimation or supply chain scenarios, to make math more applied—but your answer trailed off. Can you give me one concrete example of a lesson or project where students would use mathematical techniques to solve a specific A I or supply chain challenge?

So in order to. Definitely a curriculum for the school. Just more related to. Like. Problems makes like. For the specific case. Flight to the air supply chain. Specifically. But the a related things especially like. Like. Human pose estimation, I would say. She said something didn't. Support something? All fixed events Sunday free. Almost of the universe so this. The so in order to. This when I develop the curriculum. So like I begin with the mathematical fundamentals of. And then how? The Bangalore. Of the mathematics. Are you like a? How I can? The geometry of. The cause of the human. I. AAA model. Like there are several. For human position. Some of the credit card estimate. Yeah, it's. Training the model given by the. It's to predict the human position. Process. With almost. Conduct. Humans so this acute. Yeah, he actually.

Okay, you started to lay out how you'd introduce the math fundamentals and then apply them, but your example was a bit hard to follow. Let's change gears for a minute. VIT expects faculty to guide student research projects, especially around mathematics, AI, or supply chain. Can you describe a time you supervised a student project and how you balanced giving direction with encouraging independent work?

So in order to. Definitely a curriculum for the school. Just more related to. Like. Problems makes like. For the specific case. Flight to the air supply chain. Specifically. But the a related things especially like. Like. Human pose estimation, I would say. She said something didn't. Support something? All fixed events Sunday free. Almost of the universe so this. The so in order to. This when I develop the curriculum. So like I begin with the mathematical fundamentals of. And then how? The Bangalore. Of the mathematics. Are you like a? How I can? The geometry of. The cause of the human. I. AAA model. Like there are several. For human position. Some of the credit card estimate. Yeah, it's. Training the model given by the. It's to predict the human position. Process. With almost. Conduct. Humans so this acute. Yeah, he actually. Some data so that the algorithm is good around 90%. And then I will give. Media Pipe. I give the math behind the media. This is the party's life. Understand how it's. I recommend this. To work on the data sets. Tomorrow, so you need. Because they could. Data task which is available in. Huge, large number of different types. If it's possible then. The difference on the. Jin or if if the related to the football. It's different so they can. The datasets are. Then explain the. Modern. Politically began. Well, I'm in. They weren't like. Startups, for example this. Working at the gym. Train the persons who. Work in the gym, so yeah.

Alright, you mentioned using real-world scenarios like supply chain and human pose estimation, but let's change gears. At VIT, faculty are expected to guide student research, sometimes even interdisciplinary projects at the intersection of math, A I, and industry needs. Can you give an example of a student project you guided in this space—what was your role, and how did you help the student move from an idea to a concrete research outcome?

So in order to. Definitely a curriculum for the school. Just more related to. Like. Problems makes like. For the specific case. Flight to the air supply chain. Specifically. But the a related things especially like. Like. Human pose estimation, I would say. She said something didn't. Support something? All fixed events Sunday free. Almost of the universe so this. The so in order to. This when I develop the curriculum. So like I begin with the mathematical fundamentals of. And then how? The Bangalore. Of the mathematics. Are you like a? How I can? The geometry of. The cause of the human. I. AAA model. Like there are several. For human position. Some of the credit card estimate. Yeah, it's. Training the model given by the. It's to predict the human position. Process. With almost. Conduct. Humans so this acute. Yeah, he actually. Some data so that the algorithm is good around 90%. And then I will give. Media Pipe. I give the math behind the media. This is the party's life. Understand how it's. I recommend this. To work on the data sets. Tomorrow, so you need. Because they could. Data task which is available in. Huge, large number of different types. If it's possible then. The difference on the. Jin or if if the related to the football. It's different so they can. The datasets are. Then explain the. Modern. Politically began. Well, I'm in. They weren't like. Startups, for example this. Working at the gym. Train the persons who. Work in the gym, so yeah. Applied to. The real world update. It's not only the gym and etc. There are several things. For example like even in the Tesla they are using. Several. Cameras to be the. The humans and the Natura object. In front of Inter. In front of the Tesla automated Tesla vehicles. So in this case the. The the test line is also. Kind of, uh. Only the. First predictions, but things like.

Alright, you described tying the math directly to hands-on data problems—like having students work with datasets for human pose estimation and using tools like MediaPipe—so they see the real-world impact. Shifting gears, at VIT you'll also be expected to guide student research and projects, sometimes even in interdisciplinary areas. Can you give me an example of how you’ve supervised a student project that bridged mathematics with another field, and what your approach was to mentoring them?

Okay,

I'm fat, right? I remember so that they. To the railroad location. Very big companies and Excel.

Right, you mentioned working with really large organizations and using tools like Excel on big data problems. When you're teaching, how do you help students bridge the gap between the theory they learn in class and these kinds of real industry applications, especially when the data or tools are much more complex than what they see in textbooks?

I completely agree that the most of. Principal may get confused.

I completely agree that the most of. Principal may get confused. Comparing. Real time applications are industrial applications.

Right, you pointed out that students can get confused when the problems shift from textbook style to real-world scale, especially with big data or messy spreadsheets. When you're guiding a student project—for example, if they're using advanced statistical methods on a real industry dataset—how do you help them structure their analysis and present clear, actionable results?

I completely agree that the most of. Principal may get confused. Comparing. Real time applications are industrial applications. So what? That should be. Like a classroom teaching. To the industries like. Espec uh yeah, you got the. One kind of thing.

I completely agree that the most of. Principal may get confused. Comparing. Real time applications are industrial applications. So what? That should be. Like a classroom teaching. To the industries like. Espec uh yeah, you got the. One kind of thing. So that like I strongly recommend. The students are. I give some. 1000 I developed some. Like a class? Teaching or the. In the textbook. Industrial applications in between like. I give the brief.

Yeah, bridging the gap between textbook theory and what students see in a real company can definitely be confusing. When you're guiding a student project with messy or incomplete data—something you might get from industry—how do you help them decide what mathematical tools or approaches to use to get meaningful results?

Alright, we're almost at the end. Is there anything you'd like to ask about the role or the team before we wrap up?

I completely agree that the most of. Principal may get confused. Comparing. Real time applications are industrial applications. So what? That should be. Like a classroom teaching. To the industries like. Espec uh yeah, you got the. One kind of thing. So that like I strongly recommend. The students are. I give some. 1000 I developed some. Like a class? Teaching or the. In the textbook. Industrial applications in between like. I give the brief. Comments. Uh. Problems. With single problems kind of toy models. Very simple, but it's not real. Quickly, uh. For the Investigl applications so. Time model kinases.

I completely agree that the most of. Principal may get confused. Comparing. Real time applications are industrial applications. So what? That should be. Like a classroom teaching. To the industries like. Espec uh yeah, you got the. One kind of thing. So that like I strongly recommend. The students are. I give some. 1000 I developed some. Like a class? Teaching or the. In the textbook. Industrial applications in between like. I give the brief. Comments. Uh. Problems. With single problems kind of toy models. Very simple, but it's not real. Quickly, uh. For the Investigl applications so. Time model kinases. The timer and etc. So *** ****. The relation between. What they're doing? Two, actually. Applications are needed. So this. In between. This give the So far I don't have all the questions. Thank you so much.

Alright, Dr., we're wrapping up. Is there anything you'd like to ask about the role or the team before we finish?

You’re right, toy models help students connect abstract math to real-world industrial applications, so they’re not overwhelmed by complexity at first. Alright, Dr., we’re wrapping up. Is there anything you’d like to ask about the role or the team before we finish?