Yes, good morning. Hope you can listen to me.

Please start by introducing your professional journey in academia

Yeah, so. I'm doctor Amit Kumar Das.

Alright, Dr. Das. When you walk into your classroom on the first day of a new course, what’s the one foundational concept or topic you most look forward to teaching, and why does it matter to you?

In the area of intelligent optimization. I develop several. Optimization frameworks.

Okay, so your background is in intelligent optimization and you've built several optimization frameworks. Can you give me a concrete example of a classroom or lab activity you use to help students actually grasp the fundamentals of optimization, especially when they’re struggling to connect theory with practice?

We reported through several. Go to your review journal papers. Those results? He applied those things.

Okay, so you're saying you developed novel optimization frameworks, compared their performance with state-of-the-art algorithms, and published in peer-reviewed journals. When you introduce these frameworks to students—either in class or in a lab—what's a hands-on activity or example you use to help them really grasp how your optimization method works in practice?

Solving several. Operations and manufacturing problems. Industry problems. We had reported those results also. After that. I did my postdoctoral. Research from. Technological University. Singapore. There I was in court. In several. Love you life projects related to. Artificial intelligence and. Machine learning.

Currently I am serving as. A faculty member. At IIM Mumbai. Inadio analytics and data.

Alright, so you mentioned applying your optimization frameworks to real industry problems in operations and manufacturing, as well as collaborating on artificial intelligence and machine learning projects during your postdoc in Singapore. When you're teaching students about these advanced optimization techniques, what's one specific way you make those abstract concepts tangible for them—do you use simulations, case studies, or hands-on coding exercises?

Alright, so you're teaching analytics and data at IIM Mumbai now, after your work on optimization frameworks and artificial intelligence projects. When you introduce intelligent optimization to students who may not have a strong math or programming background, how do you make that topic accessible and engaging for them?

So. I have more than 10. Experience in the idea of.

Alright, so over ten years of experience in analytics and data at IIM Mumbai — that's quite a foundation. When you first introduce new students to intelligent optimization, what's one misconception or stumbling block you see them struggle with, and how do you address it in class?

Yes. Great question. Uh, if the student so. In the class. I'm not start. Uh, initiate the discussion. Using the equations. I'll avoid. In the initial phase. What I'll do? I will try. Intuitive best. Learn first I will set the context. What is the problem? Try to set the. Align with. And those industry problem quite is important to understand. What are the real need? Challenges. In industry. I'll explain that. Then I will engage them. What can be the possible solutions? So they will come up with different different. Uh, perspective tips. Ideas. Correct SO. My job. After that. Will synchronize those diverse type of. Perspective. And after that? Align them. With analytical framework. So it will be a journey from intuitive. Then some diverse. Perspective Generation. Discussions. And then? I'll try to. Align and integrate with some analytical frameworks. So they will first understand. Why part of that problem? Then they will try to understand. How? It is going to happen. How we can? Solve it. So first if they understand the why part. For them it will be very easy. After that. To understand. How to do it? So in this manner I'm going to. Teach if some students has. No prior background in optimization. Call. Yes.

Okay, so you emphasized starting with intuition and real industry problems before getting into the equations — that’s a thoughtful approach. Can you give me a concrete example of a classroom or lab activity you use to help students connect those intuitive discussions to the more formal analytical methods?

Let's take one example.

So you were about to walk me through an example — go ahead, I'd like to hear how you’d connect the intuitive part to the formal methods in an actual class session.

Let's say you are a production planner. Working as in that role. You're you have that responsibility to plan. Optimal. Optimally. The production. Schedule. Now you have different resources. We know. In industry we have finite amount of resources we don't have. Unlimited. So. How to? How to? Optimize or optimally? Schedule. The production. And that planning?

So you're putting the student in the role of a production planner with limited resources — that's a great way to make the challenge real. Once they've brainstormed approaches, how do you help them bridge from that open-ended discussion to actually formulating the optimization problem mathematically?

Let's say. That, uh. And does the other company? Is producing 2 types. Product Product A and Product B. And they have.

Okay, so you’re setting up a scenario where students have to plan production for two products with limited resources — that’s a very real-world setup. When you move from that intuitive scenario into actual analytical modeling, what’s the first formal tool or framework you introduce to help them structure the problem?

And how can you come up? With a good solution. Which will be. The optimal, the best one. Are you getting my fault? There can be a. How many solutions? As all of you are saying. Because in the class. Many students are going to give different different answers. So all of them are. Possible solutions?

So you have to. Consider that solution. Which will be. Considering all those limited resources. So in this way I will drive the discussion. Then I'll try to. Introduce.

That's an interesting point. Could you elaborate a bit more on that?

Okay, so you walk the students through a real production planning scenario, have them brainstorm and debate different ideas, and then focus on feasibility and optimality under constraints. When you run this kind of session, how do you actually check if your approach is working—are you looking for certain student responses, or do you use something like a formative assessment during the activity?

Those different framework. Optimization framework. Listen me a programming framework. Then how to? Phone letter. Objective function. What is my variables? What are my constraints? How we can that so? Then I will link it. When they will understand. Why part then? They will, I will. I'll try to. Make them exposed to how? What in this way? After formulating that problem. I'll show them. How we can also solve graphically because we have? Two variables here. And also using excel solver we can use. Then I will also explain. The results? Optimal results. Which are the binding constraints? Which would be the critical resource? Which will be. The fully utilized, underutilized. All these concepts. Step by step, gradually, I'll explain. So that they can understand fully. I'm very much expert in this. Particular to me. Yes. That's all.

Yes.

Given your experience, Professor, let’s shift focus a bit. Accreditation cycles can be pretty demanding — if you joined our department and found that outcome assessment data was inconsistent across courses, how would you approach fixing that?

Alright, so you walk students from real-world intuition into formal frameworks, and you mentioned you're quite experienced in this area. Let me pivot to your potential department responsibilities: Professor, accreditation cycles can be pretty demanding — if you joined our department and found that outcome assessment data was inconsistent across courses, how would you approach fixing that?

OK. Can you? Can you? Please come again that question.

Of course, Professor. The question I asked was: accreditation cycles can be pretty demanding — if you joined our department and found that outcome assessment data was inconsistent across courses, how would you approach fixing that?

That's my answer.

Okay, so you’d bring together the course instructors to review and adapt learning outcomes, making sure any updates reflect both student needs and industry relevance. What’s your understanding of the day-to-day documentation and evidence-gathering that accreditation bodies typically expect from faculty during these cycles?

Yeah, that's my answer.

Thanks for sharing that, Professor. Let me shift gears to something a bit different—imagine a student formally complains that your grading is biased, but at the same time your department head is pushing you to improve overall pass rates. How would you handle that situation?

OK. So in this uh. Tuitions, uh, it is not that. Under some pressure, we need to. Buckle down. Which should not be. It should not happen. Like some students are complaining. And then what we can do, we can. We can. Check our papers. Normally it happens in. Calling me cities also. That uh, checking is done by some other. Professors all some. Research assistants. OK. Or maybe academic associates? So they can check those papers so that this. Problem of biasness can be eliminated. Yes, human being are sometimes become biased but. If someone is complaining, then we can take that kind of. Ways to evaluate the papers. Answer skips. That using the academic associates or. Maybe some? You're gonna suffer alone. Between the professors also. OK, who's not teaching? Maybe, but. From the same area we can again. Do that. Uh, in that way also you can evaluate so that we can eliminate this biasness. And about that? Your department is. Pressing into. Increase the pass level. In that case we need to understand. That. What? Uh yeah, maybe for. By doing that. We can, uh. In temporary basis you can get the game. But in. Long term basis, it is not going to help us today. Because ultimately, we have to see. The side of the student. Student outcome. They are understanding their learning level. If that is not proper. They are going to fail in industry. So industry people will point they will come for placement or hiring. They're not going to like those, uh, you know, not necessarily develop those, uh, students and ultimately it is going to. The. Image of that institute. So we need to see this long term. Effect and impact. And accordingly. We should not. Uh, if some of the students are really. Doing uh. Very bad. OK then. We have to. We have to give more opportunities. To for example V exams. RVV exams you need to contact. And uh. If it is needed. We can personally. In any individual VCs we can. But just giving the pass marks. Without having their capability or without developing the capability into that student. It is not a good way to. Go with him for a long time. It is going to. Ultimately impact the institute image. And plant refuge. So my thought process is we should not do. Do that instead. We should. Father may be increased. Like personal mentoring? And whatever they need, we can can. Ask them what? Why? What is the issue? Why? You're facing difficulty in understanding. OK, so personally, those things are. Yeah, beyond the class we can can. Stress that that is possible. So that ultimately. Their learning should be maximized. OK. And. During hiring process. The industry people should not see that. Students are not. Knowledgeable. It should not happen. So. We have to prioritize the student outlook. Always they're learning latest. We should not compromise with that. Next meeting.

So you’d prioritize fairness by involving external graders or colleagues and push back if pressured to lower standards, focusing on long-term student competence instead. But if your department head insisted on raising pass rates as a metric for faculty evaluation, how would you defend your approach in that meeting?

Yes, uh. That we can can. Very nicely politely, you can. Discuss on that. Debate on that. And, uh, jointly we can discuss that. Our privatization should be ours. Today's outcome? They're learning level. And uh, that's we should have the focus for our long term. Sustainability. Long term growth. Of that institute. We should not compromise at all in that part. Otherwise, that institute is able to. Go into the Norton. Proper direction? No, it is going to go. In wrong direction. OK, so. We can. We can set. We can, we can. Talk, uh, or. You can bring in different examples. OK. From other institutes. And jointly. We can have a good discussions on that. And. Without having. Any personal attacks? So in that discussion, we should not. Uh, personal talk like we should. Make the topic D personally. Be personalized. And. Amiable. Kind of things should be there. It should be a very healthy discussion. And ultimately, you should. Prioritize the split up. I'm aligning and. For giving that. For doing that, whatever. And the ways we need to take. We should go with that.

Right, so you’d focus on keeping the conversation constructive and depersonalized, using examples from other institutions to make your case for student learning as the core metric. Before we wrap this section, Professor, can you briefly outline how your experience with industry consultancy or projects could create impactful collaborations for both students and research at VIT?

Yes, yes. That's a great point actually. And that's a great strength of any faculty. Right. The ultimate. Goal, uh. Is to make that student a whole person. Not only a for particular skill. In different different scales they should be equipped with. And ultimate goal is to make. Incorporate ready. Or industry ready. They should understand different industrial challenge. This would not be limited to only the theoretical things. They should understand theoretical things. No doubt. But how to? Bridge the gap between. Theory and practice. Now if that person. The Minter. That's Professor. If he or she has. Search industry experience. Our consultancy experience. He or she is involved. Is solving actively. Those real world. Industry problems? He or she can bring in the class. During the discuss. Those experience those examples. That. How? The industry problems all look like. Their depth and breed. All these things. All those things. Those examples. Yoshi can bring. In the class discussion. So that the students can be aligned with that one. They should not understand only the theory. They should also relate with their. In the practice. How those are relevant? In industry. What are the? Different challenges. You can have. In industry. And whatever. The topics, whatever they are learning. Those methods. Who is frameworks? Can help them? For framing or offering solution.

So that there should be good balance between this theory and practice. This would this would get that. Uh, elephants. Those subject matters. Those topics, those concepts. How they are relevant? In the domain of. So if that. Officer has good understanding. Doing actively. Such industry problem solving. Then he or she can bring those. Instant says examples. And giving those they can discuss also. Those issues problems. In the class. And students will be highly benefited. But ultimately, they need that. Their thinking process should be channelized in the direction. So they are critical things. Ultimately, they should. Develop. A critical thinking process. That will be the ultimate goal. So. Through that, their thinking process will be aligned with that. Industry practices. So that will be. The good bridge between this. As I said. Theory. Thank you.

Alright, Professor. Now let's talk specifically about bridging academia and industry. Given your background, do you have any concrete industry connections or experiences that could help create internship or placement opportunities for VIT students?

So you’d use your own industry and consultancy experience to give students practical exposure, showing how theories map to real-world problems and helping them develop critical thinking aligned with industry needs. Can you give one concrete example—maybe the name of a company or project—where you actually facilitated this kind of collaboration for students or integrated a live industry problem into your teaching?

OK. So. A few days back we are. Working with regime pack. Where they had. Given US1 challenging problem. To it's related to predictive analytics. OK. How you're going to predict? The next. Uh, maintenance schedule. Four different aircrafts. It is related to the aircrafts. So they have given. Very big data. So how the data processing is to be done? Data clinic process and. What are the? Feature selection. How to do that feature selection? And in that features. What are the different ways methods are there? How to find out the significant insignificant features? And after finding out those features, what kind of kind of. Model you should select. What are you this? Predictive modeling purpose. I have to get that prediction. This year cut is going to be. Uh, need that. The maintenance. At this time or that time. After one month, after two months or three months, whatever it is. OK, so. Those things are. Very much important having that reload. Data and I. Uh, though I have not. Because that data was confidential. So I mask the data. But I told that this is. I did not mention the name of that. Company or. Web specific things. So. That data was already marked by them. That Jim had given us. During that time, it was. Must already. So. That marks data I. Give to the stories. To do the whole process to repeat. Whatever we had done during the project. And they did it while. And come up with a very good way. Katie Seal, uh, that prediction. And in this process. They they learnt actually the whole. Predictive part. Starting from that data cleaning, data preprocessing features. Model selection. Model building. Hyperparameter tuning. Everything. Their own they did the coding part. And in this way. They did a wonderful job. And. It it they enjoyed. Actually, they learnt very. Uh, in a better way with that. So. Thank you.

So you’ve shown how you connect industry projects to student learning. Now I want to check a few core areas for this role. Starting with multimedia or artificial intelligence in media—when you’re designing a curriculum module on, say, media content analysis using artificial intelligence, what’s your process for balancing foundational concepts with hands-on tool exposure for students?

All these three things like business context, the technological knowledge and this. Machine learning knowledge. If you have this combined one. Composite one. Then only you will be. It will be you will successfully. Build that model. So business context, technological knowledge and also the missing learning. Are required for doing. For solving that. Model building or. Uh, solving such? Complex problems.

Alright, you emphasized how blending domain knowledge, technology, and machine learning is key for students to build real solutions, not just run tools blindly. Quick check on another core area—can you describe how you structure your student evaluations and exam duties to ensure they're fair and truly measure understanding, not just memorization?