Good day, Dr. Erukala Kalyan Kumar. I'm Arjun. I've had a look through your background and your professional record. I'm interested in hearing more about your work and how you see yourself fitting in here with our faculty. Ready to start?

Please start by introducing your professional journey in academia.

Yeah, very good morning myself, Doctor Erika Kalyan Kumar. So I have been working in Presidency University, Bangalore from past two years. So earlier to this I did my PhD at night at Orkela and before my PhD I worked as an engineering faculty for four years in a private institution in Warangal. So basically I belong to Warangal. So coming to my research background, I worked on.

Engineering design department in native worker.

So to check the strength recovery and also to improve its stiffness and strength of the composite materials for aerospace structures. So my research aligns with the numerous publications. I have over 30 SEI index publications in my research and also couple of patents out of which one was published and 1 was filed and a few book chapters and as well as.

For under smart materials integration to the composite materials. So basically it deals on computational modeling of a component materials with integration of shared memory alloys.

Yeah, this is about my brief deduction on research and teaching experience.

Conferences as well.

Thanks for sharing that, Professor Kumar.

Yeah.

To begin, Professor, could you describe your experience and approach to computational modeling, particularly in the context of your academic or research work?

Yes, Sir, Sir. Actually during my PhD, I worked as a like on smart metrics integration, specifically shared memory alloys in the laminate composites, OK. In that what I did was I used a computational modeling of the laminate and also I integrated with SME in it. And I also I improved the stiffness of the material which is basically utilized in aerospace structures, OK. In this the computational modeling I used was.

Through MATLAB and also a self developed code in which we have tested with several what we call it as several structural parameters, you analyze several structure parameters like static analysis, dynamic analysis and post buckling analysis and as well as some deflection analysis and we are also verified that these these all. Have been improved with the help of the Smart Material integration.

So computational modeling, we have already succeeded and I just wanted to continue the particular work with the different aspect.

In the vit Vellore in Cimmer Research Center. So that is the main aspect I have applied, yeah.

Thank you, Professor. Could you elaborate specifically on how your self-developed computational code handled structural parameters? For instance, was it primarily numerical simulation, or did it include analytical formulations as well?

Yeah. See the MATLAB code is a customized code for which it is developed for SMA integrated laminated code. So in this basically we have used higher order shared information theory, reduced higher order shared information theory and also we have a utilized Newmark solutions and several Fourier series transformations through experimental modeling. So this is a customized code we have developed by utilizing the analytical approach as well.

And uh, the numerical solutions have been, uh, uh, the numerical solutions have been evaluated through experimentally and, and compared both of them. And we have been validated that particular code experimentally as well as we also used some lower order uh, softwares like Abacus and Ansys, uh, to, uh, like to verify whether our customized code is working properly or not. So with the double factor, we have verified and it is working well and through that only we have.

Uh, like we have taken some of the results and also we have been compared with experimental, uh, results and we have analyzed it, Sir, so.

That's what.

Thank you.

How specifically did your computational modeling account for the unique properties of shape memory alloys, such as their hysteresis or phase transformation behavior?

Yeah.

Basically there will be 4 parameters in shape memorializer like Martin said, Martin said phase and Austin phase, Martensite phase there will be start temperature and temperature and austenite phase there will be start temperature and temperature. So we call basically this whole loop, we call it as a hysteresis loop of the SMA. So we have a change of the phase transformations of SMA in the customized code so that the behavior of the SMA will be implement will be utilized. To the composite parent material like epoxy resin material which you already used.

To eradicate that particular loss and to balance that particular strength in the material. So that is how we used SMLSMA as the integrated element and in the in basically in SMA we used netinol as a the material means nickel data MLR is the base material and we have used the that particular phenomena like shape memory phenomena to adopt it to increase the stiffness and strength of the period material.

So, so that if there is any damages or if there are any like strength reducing strength things in the compared material where we have utilized this SMA to balance that particular strength or stiffness. So that is the how, that is how we use the we utilize the phenomena of phase transformation of SMA to.

Thank you for the clarity, Professor. Could you now discuss your experience in applying AI or ML techniques to materials science or manufacturing? Have you had hands-on applications or projects in this space?

Yeah, Sir, Sir, actually, uh, my plan is during my PhD, I only look for like a strength recovery of the material using SME. So now I am focusing on self filling mechanisms that can be implemented to the aero panels like morphing of wings. And it is very essential for aerospace applications where we can, if you are implementing any self-filling mechanism in aerospace material such that those, those cracks can be detected early. So if the cracks can be.

Early then it can be solved and it can it, it will be implemented for safety of the aerospace materials as well. For that what we are doing is so we need to get some of the experimental data like all the structural parameters data like deflection, static, dynamic and post buckling all this data and we need to we need to like develop a neural network model A and model neural network model so that we need to train that particular data.

With it and with the intact means without any crack panel data as well as with some of the different different crack panels data. So after implementing that particular setup in the training, then if you are going to give any responses like frequency responses or uh, deflection responses of that particular panel in the real time panel, which is not yet trained. So obviously this based on this training data, it will be uh, it will be uh.

And this damage detection can be done through this AML neural network. So I'm working on it presently. So we have filed a patent on this pre training of SMA and to implement for the self filling materials as well. So if you are going for this, we need the computational data of the experimental setup. For that purpose I I require a high computational requirement and I hope VID will be the one which I can utilize all these facilities to implement this.

And which can be a great project and good research if you can adopt AML to the.

To this particular setup. So this is how my plan is, Sir, yeah.

Yeah. Just give me a second to read the question, Sir. Yeah.

Thank you, Professor. Could you explain how you plan to train your neural network model? Specifically, what structure do you envision for the model, and how do you intend to handle variations in material properties or environmental factors across different data sets?

Huh. OK, so yeah.

So basically, Sir, we need an experimental data for intact plates means without any detection intact plates and as well as we need some experimental data for cracked plate. So initially what we'll do, we'll take, we will pick a composite in which with SM and without SMA or which smart metals and without smart materials, with damage and without damage. All this data will be extracted with the help of the. This computational modeling code where we have already integrated SMA in it.

And without SMA, with SMA we already have developed these MATLAB codes. So with that we will take out the data by using this HSD formulations. And also we use the setup called CDAP and Fourier transform series and lab use software. With the help of all these integrations of this experimental, we'll get the data. These data will be trained, 80% of the data will be trained. With this code and also 20% will keep it for testing. So when we got to know this?

Uh, neural code has been trained enough. Then what we will do, we will take a foreign plate with with some random plate and we will test it with for, for example, we tested for in a frequency analysis. OK, Then we will get some sort of frequency response of it. These frequency responses will be given to that particular neural network model. So based on that response, it will recognize whether is there any internal damage or micro cracks based on the responses. So it will do a reverse modeling kind of thing. So this is how we are going to plan and.

Also we need a lot of experimental data to train it for that. We need high computational system for that. And I, I do have some of the data we are doing in minor level, but I wanted to execute in a major level where I can explicit that particular model into a greater thing. So this is called AML assisted structure health monitoring of a self filling SMA error panels. So this is what we thought about.

In higher level.

Thank you for the detailed explanation, Professor. Now, shifting gears slightly, can you describe your ability to teach courses that balance both theoretical and practical components, perhaps drawing from your experience with computational methodologies?

Sir, I I already informed you that I have 5 years of teaching experience. It's been a lot of experience. I've been teaching several subjects in mechanical engineering. So basically I I prefer to use a laboratory and theoretical subject, both integrated kind of subjects very much. So that is, for example, if I'm considering a finite element analysis subject, FEA subject in this I will teach theoretically, mathematically, analytically to the students. How this going how this engineering problem can be solved in in the class. So when we discuss the fundamentals to the students first after discussing the fundamentals, we just trying to go through the analytical part and we we we will explain the students to solve it manually. So after that particular class, the practical component can also be discussed in the lab where the same problem can be taken over in the lab. And in the lab we are going to solve it in a simulation software like ANSYS or Abacus which will be relevant to that FES software.

So this shows the student theoretical as well as the practical knowledge of that particular problem. So I love teaching in that way. So most probably not all the subjects will be having the same, but I will try to implement with the different teaching pedagogies so where the student get interested and also get some practical knowledge out of it. So this is and also I've taken this Fe example because you asked about how this computational modeling knowledge has been implemented. So in this way I will use this computational modeling.

If to the students with the help of particular software, yeah.

Yes, Sir, Sir, basically I am an outcome based person, Sir. Like how how I'm going to teach at home, I'm going to implement the teaching. It will be minimum to me how the student is going to get the outcome. So I used to see always I used to analyze with the help of few things like for example, I will give a sort of a problem in the class to solve by them like like an assignment or else in the lab if I'm going to the teaching, if I'm going to for a lab, I'm going to give a different modeling technique.

Thank you, Professor. Following up, how do you ensure that students understand and engage with these complex subjects, especially when dealing with computational or analytical models?

To solve if they're going to do it, I will analyze the class how the how the class is going to do. If they're doing in a better way, I do confirm that the students are understanding quite well. If they're not doing so, I will analyze where the particular where the student have been structured and up to what level the students have been doing it. Yeah, I do agree that few students might do it and few students might not do it That that too I can agree but but on the case to case basis I will interact with them and also. If if I'm the mentor to them, I will be specifically teaching them personally.

Based on the results like how they are writing their exams, based on their assignment, how they are doing their approach and based on their confidence level in the class, I will come to know how how they can like how they can achieve the things. And also based on that I can take up some remedial classes or else I can take up some improvement classes to them and to improve their.

During the non class hour times and as well as apart from these things.

To their technique and as well as to improve their subject knowledge as well. Yeah, that is how I will approach. I will take the approach.

Thank you.

Yes, Sir. See when I'm going to evaluate any kind of script or any kind of designing or any kind of modelling. So I'm not going to check individual student history earlier like how they're going to perform and see the sheet to sheet. Like for example, if I'm going to see any designing like any modeling in the system, for example, if they're done any component, any design component, I will just check the history of the component, how many steps he has did see few, few students might do.

your experience in evaluating students, such as designing examinations or assessing their projects? How do you ensure fair and consistent evaluation?

In very few steps and few students might do very longer steps. So based on that, the evaluation can be done and I will identify the errors, how this in the process, how the student is going to get the mistake, how they he's doing the mistake. So I will correct it and fair. I will be fair every time because it's not I don't get anything. If I wanted to favor any student, I wanted to. My personal feeling is to improve the candidate, to improve the students. It's my personal goal. So always I will be fair to the student as well as.

Evaluation, So I, I will step necessary, I will take necessary steps to improve my evaluation as well. Like I will frame a particular rubrics like uh, for the evaluation scheme, for example, if they have done this particular part of step this, this will be the marking criteria. So based on that particular rubrics only I will go into mark the student and based on the number of steps they are going to take to solve, for example, few students might be bright, they might be using very minimal steps to solve the problem. Few students might take a longer step. So I will try to improve.

These people more and I will try to improve like I will try to encourage those people more. So in the case to case bits I will evaluate, but it will be fair according to the rubber itself only.

Thank you, Professor. Could you also outline your experience in guiding student research or projects, particularly how you mentor them through complex, self-driven initiatives?

Yeah, Sir. I just wanted to explain with an example here during before my PhD, like when I'm working in a private institute where the students where the students are too much interested to do some automobile kind of thing and some some part some sort of practical experience for them. So what they what I have did is I picked 7 to 8 students out of the class who are interested and they might given divided them into teams and I've given some part of software part.

And few people have given from some, uh, uh, like a component fabrication type of card and uh, within three to four months of time, uh, we with a guided environment with some, some support from the management as well. So they have fabricated A2 Wheeler electric vehicle.

OK, further final layer project. I am the guide for it and we have implemented it in a good way and that has been giving a good efficiency, good mileage and all. So we have participated in some competitions and we got a good rank out of it. So this is how I will take up a students team. I will motivate them to get into the and those people are settled in a good places. Now with the help of the project like by showing by by getting that exposure they will set few students have settled in good automobile industry as well.

Their path like which path may might be suitable to them and what are the hurdles they may face. We will caution them. We will try to solve their problems and also we will implement our particular subject, whatever the subject they are learning in Btech, how they can be implemented, how they can be utilized in their jobs as well for the interview as well. So these kind of things as a teaching, as a teaching assistant professor, I'm very well known and I've been doing from past four to five years of time so on mentoring and. Students in project and lot of projects you did during my PhD as well. I deal with a lot of Mtech students in Nadir, Orkela, where many students have been doing their PhD right now and be in contact with me. They have been in collaborative work with me. So it's been a good experience with them. And I love teaching and I love collaborating with students and I love doing different things and I love being and I feel very happy and satisfied with the teacher when the student gets a good placement or like a good kind of position to him. Yeah, that is what.

So in the same way mentoring the students, even I do mentoring Sir, I have 30 students in the current institution where I'm going to work here. So depend on the particular student requirement. Like for example, a few students might be interested in higher education, Few students might be interested in like doing some sort of job, Few students might be interested in going to some other jobs, few students might be interested in starting a startup or entrepreneurship. So based on the case to case base, we will mentor them like this. We will mentor them according.

I'm bit emotional with students, yeah.

Thank you, Professor, for sharing such a passionate and insightful perspective on your mentorship experience. Could you discuss your approach toward communicating complex concepts effectively to students, ensuring clarity while maintaining rigor?

Sir, can you repeat the question because?

I think that was some sort of.

Internet issue? Can you repeat the question Sir? I didn't get it.

Yeah, Sir. See, it depends on the type of students. For example, if I'm going for a second semester student or first year student. So as they're beginning to the engineering, we cannot directly deviate into the complex problems or complex things. We we need to make them understand what are the basic things like what are the basic understanding of the engineering and what are the basic concepts of engineering in that. For example, if you are going for the higher semesters students or like final year students, they already been to this all basics of engineering and everything.

Of course. How do you approach communicating complex concepts effectively to students, ensuring clarity while still maintaining academic rigor?

For them, uh, normally we will revise a bit of, uh, like basic understanding to them and we'll delve into the complex problems through different pedagogies. For example, if the students are, uh, not understanding the concept or the students are not getting that particular concept in a better way. So we, we generally use different teaching pedagogies like, uh, some sort of videos. I will show some sort of practical, uh, uh, videos I will show them or else if possible, if time permits, then I will plan for an industrial visit where that particular, uh.

System will work practically over there or there are a lot of virtual apps are there there we we we kind of virtual apps are using in online we can show through the virtual apps as well. And if the if the if the problems are complex in nature, So what we can also use this a tools like ChatGPT Gemini to get it simple and to make the notes very simpler to identify the solutions in a simpler way. And also there are few textbooks where.

For beginners, there will be the problematic, the problems will be in a simple way. So we will recommend the students to use those books for reading those complex theories. And also as a teacher, I, I'm always a continuous learner. If still student, if he is not able to, uh, like understand my point of concept, I will also take help of my senior professors in my university to explain them in a better way. Or else I will learn from them and I will execute it the same in the classroom. So this is what, uh, I, I feel.

Uh, to make the student understand in a better way, Sir, Yeah.

Yeah.

overview of your research publications? How do they align with advancements in computational modeling or materials science?

So if you if you go through my research publications through my resume, so you will find a lot of applications on computation modeling, damage detection, share memory allies, smart materials. So we have been even I worked on some mechanical biological kind of things and but majorly I worked on computational modeling of this laminates integrated with shape memorial materials. So here what I wanted to tell you is.

My research point outs only for one application like for any spatial or aerospace applications, if you wanted to improve the strength of that particular composite. So how it can be utilized if we can use the SMA as a material and then if you're using the SMS material, so how to improve it and also to detect the damages wherever the damages are there. So to balance that particular damage, how you are going to use the SME. So this is my core point. So coming to the my whole publications relevant to CIMR, Yeah, absolutely.

So basically.

Because CMR is the Center for innovation research and manufacturing. So even I do the same during my PhD for computational modeling as well and also the experimental things as well. So I wanted to elaborate my experimental findings in a higher level through with the help of this particular university. So I am looking forward to join over here and to explore my research capability over here to benefit my personal as well as the organization as well. Thank you. This is what I wanted to.

Thank you, Professor Kalyan Kumar. Based on your detailed responses, your strong expertise in computational modeling, shape memory alloys, and your dedication to teaching and mentoring are evident. This concludes the skills verification portion of our discussion. If there’s anything further you'd like to add or ask, feel free to share.

Yeah, if you have any information regarding to the role and committee, please let me know because even I have not contacted anyone from the university. If whatever the information you have regarding the role, could you please tell me it will be beneficial for me for further wrongs as well.

Our scheduled time for the interview is now up. Before we conclude, do you have any questions related to the role or the company?

No, no, I know. I know the qualifications and responsibilities because I already read in the interview. So I just wanted to ask one final question. So based on our interaction, could you please give me my feedback where I can improve myself? Like is there any constructive feedback from your side so that it will be beneficial for me in any other aspect as well? Yeah.

The position entails delivering high-quality teaching in theoretical and laboratory courses, evaluating students, managing examinations, guiding student projects and research, and contributing to publications in international journals. Additionally, significant emphasis is placed on engaging in research development, providing consultancy services, promoting industry–institution interaction, and participating in R&D initiatives. The preferred qualifications align closely with your background, as the role seeks expertise in computational modeling, AI/ML applications in materials science, and digital twin technologies. There’s an emphasis on proficiency in computer programming, computational analysis, and holding a Ph.D. in a relevant specialization. Notably, experience in curriculum development, accreditation, and exposure to high-value funded projects or patents is advantageous. Do you have any specific questions about the responsibilities or qualifications for the role?

Thank you so much, Sir. Thank you so much for your valuable suggestions and you can conclude the interview. Sir, thank you so much.

Professor Kalyan Kumar, you've demonstrated exceptional depth in computational modeling, materials science, and practical teaching strategies. Your passion for mentorship and your innovative approach to research, such as integrating AI/ML with structural health monitoring, is certainly commendable. One area to consider refining might be the structuring of answers. While your responses are very detailed and rich in content, presenting them in a more concise and logically segmented way could enhance clarity further, especially in professional discussions or interviews. Your enthusiasm and dedication stand out. Keep that focus on research innovation and clarity in communication, and you'll continue making strong impressions in academic and professional settings.