Interviewer 0:00:00
Hello Dr., thank you for taking the time today. I'm Arjun. I've reviewed your background and would like to go deeper—shall we begin?
Dr. Yogeshwaran Krishnan 0:00:14
Yes.
Interviewer 0:00:14.1
Please start by introducing your professional journey in academia
Dr. Yogeshwaran Krishnan 0:00:35
Thank you for giving the time. To introduce myself. I did my bachelors in physics and after that I did my masters in material science.
Interviewer 0:00:36
Alright, so you started with physics and then moved into material science for your master's. What drew you to materials science after physics?
Dr. Yogeshwaran Krishnan 0:01:48
And after that I worked as a junior research fellow at. Yeah, actually I like to walk and material science because I like. To do research. So I chose from physics to. Material Science. I like to do. Research. So I did. Shift from physics to bus. The material science because. Physics is a general subject and. I I want to move towards uh. Reset towards materials. So I went. To study material science. In PhD College of Technology. Climb the door And uh, this is the best university or College in And uh, this is the best university or College in. Yeah, when I was doing the Junior research fellow. The time, uh, there are so many juniors for me.
Dr. Yogeshwaran Krishnan 0:01:43
And after that I worked as a junior research fellow at. Yeah, actually I like to walk and material science because I like. To do research. So I chose from physics to. Material Science. I like to do. Research. So I did. Shift from physics to bus. The material science because. Physics is a general subject and. I I want to move towards uh. Reset towards materials. So I went. To study material science. In PhD College of Technology.
Interviewer 0:01:47
Dr., since you mentioned your experience as a junior research fellow, can you describe a teaching moment or a lab session that was particularly rewarding for you?
Interviewer 0:02:10
So you had a group of juniors while you were a junior research fellow. Can you share a specific situation where you had to explain a complex material science concept to one of them, and how you made sure they understood?
Dr. Yogeshwaran Krishnan 0:02:15
I I thought them. To do. Research.
Interviewer 0:02:16
Okay, so you focused on teaching your juniors how to do research. Was there a particular experiment or project where you saw a student really grasp the material and take ownership?
Dr. Yogeshwaran Krishnan 0:02:43
Yeah, actually I used to teach them. A Molecular dynamics.
Dr. Yogeshwaran Krishnan 0:03:00.0
Yeah, actually I used to teach them. A Molecular dynamics. How to do molecular dynamics for research of material science? So I taught them every basic things and asked them some questions.
Dr. Yogeshwaran Krishnan 0:02:57
Yeah, actually I used to teach them. A Molecular dynamics. How to do molecular dynamics for research of material science?
Dr. Yogeshwaran Krishnan 0:03:03.0
So when I asked question they answer. From that answer I. Understand whether they understood or not I used uh. To teach them using laptops.
Interviewer 0:02:56
Given your research on machine learning potentials for solid-liquid interface materials, where do you see the most fundable directions for the next few years? Which grant agencies or programs would you target to support this line of work?
Interviewer 0:03:00
Alright, so you taught your juniors how to use molecular dynamics in material science research. When you were mentoring them, what strategies did you find most effective to help them really engage with the theoretical side, especially when the concepts were complex or abstract?
Interviewer 0:03:03
Alright, you guided your juniors through the basics of molecular dynamics for material science research. When you were mentoring them, what strategies did you find most effective to help them really engage with the material and develop independent research skills?
Interviewer 0:03:57
So you made use of laptops to teach molecular dynamics—that’s pretty hands-on. If you were to scale up your research group in solid-liquid interface materials,
Dr. Yogeshwaran Krishnan 0:04:42.0
So it's like trial and error method. They tried to do some. Uh, simulation along with me. Do get a research funding. We have to write or read too much. Research articles that is people are working on. And what they're working on so from that. We can find this problem, a small problem and we can try to solve that It was all that we can write a proposal. To BST or Prime Minister Early Grant.
Dr. Yogeshwaran Krishnan 0:04:34
So it's like trial and error method. They tried to do some. Uh, simulation along with me. Do get a research funding. We have to write or read too much. Research articles that is people are working on. And what they're working on so from that.
Interviewer 0:04:42
Alright, let’s shift focus. Suppose you’re teaching quantum chemistry to a large undergraduate class, but you can’t use slides or traditional lecturing.
Interviewer 0:04:52
You mentioned writing proposals for DST and Prime Minister Early Career grants. When you're mentoring undergraduate or postgraduate students at VIT, how would you help them identify a research problem that's both academically significant and fundable?
Dr. Yogeshwaran Krishnan 0:05:25
Try to ask them to read.
Dr. Yogeshwaran Krishnan 0:05:33
Some research problems. Some research papers went uh, read the research papers they. Will find a research problem.
Interviewer 0:05:30
Alright, Dr., shifting gears a bit. Accreditation cycles can be demanding — if you joined our department and found that outcome assessment data was inconsistent across courses, how would you approach fixing that?
Interviewer 0:05:35
Okay, noted.
Dr. Yogeshwaran Krishnan 0:05:57
Can you repeat the question please?
Interviewer 0:05:58
Of course. Accreditation cycles can be demanding — if you joined our department and found that outcome assessment data was inconsistent across courses, how would you approach fixing that?
Dr. Yogeshwaran Krishnan 0:06:20
I'll try to analyze the problem first. Why it is there is inconsistence?
Dr. Yogeshwaran Krishnan 0:06:23
I'll try to analyze the problem first. Why it is there is inconsistence? See the outcome.
Interviewer 0:06:22
Right, so first you'd analyze why the data's inconsistent. Once you spot the root cause—maybe it's different assessment methods or record-keeping—how would you actually get faculty to align and implement a consistent process going forward?
Dr. Yogeshwaran Krishnan 0:06:30
So, uh. After finding the problem. Then I try to solve.
Interviewer 0:06:26
Okay, so you'd start by analyzing the source of inconsistency.
Interviewer 0:06:34
Let’s consider a real situation, Professor. Imagine a student files a formal complaint, alleging your grading is biased. At the same time, your department head is pressuring you to improve overall pass rates. How would you handle this tension between maintaining your grading standards and responding to institutional pressure?
Dr. Yogeshwaran Krishnan 0:07:33
To make a pass percentage I would not go. Degrade by grading system. But I will in instead. I will. Improve my teaching so everyone will get a good grading.
Interviewer 0:07:40
Alright, Dr., let’s talk about industry connections. Given your research in solid-liquid interface materials, do you have any collaborations or contacts in industry that could help students with internships or exposure to real-world projects?
Dr. Yogeshwaran Krishnan 0:08:04
Oh yes, I do have. Contacts uh, where I did my post doc.
Interviewer 0:08:08
Okay, so you have industry contacts from your postdoc days. Now, let’s shift to a different area—student evaluation and exams.
Dr. Yogeshwaran Krishnan 0:09:06
And also some, uh, private. Or risk private research. Tutes like Toyota. I just use the same method for everyone so it will be consistent but. Before that, when I teach, I. Analyze the student. Ability to learn. So according to that I will try tool. Teach different people differently. But I will make. Bring the outcome, but that. The everyone is understanding the problem. Or everyone understanding the. Science behind that?
Interviewer 0:09:08
So you adjust your teaching methods to bring everyone up to speed, but keep grading standards consistent. When you have to explain a tough concept—say, electron delocalization—to students with very different backgrounds, what's your go-to strategy for making sure everyone walks away understanding it?
Dr. Yogeshwaran Krishnan 0:09:39
I will go very fundamental to teach uh. About the subject.
Dr. Yogeshwaran Krishnan 0:09:43.0
I will go very fundamental to teach uh. About the subject. So I will teach like you don't know anything.
Dr. Yogeshwaran Krishnan 0:09:47.0
Sorry, I don't know anything about the subject.
Interviewer 0:09:43
Alright, you mentioned going very fundamental to teach the subject. Can you give me a concrete example—maybe a topic from your PhD work—where you had to break down a really advanced concept for students with little background, and how you actually did that in practice?
Interviewer 0:09:47
Alright,
Interviewer 0:09:49
So, you’d teach from the absolute basics if someone has no subject background. Can you give me an example of how you might introduce a core chemistry concept—like acid-base behavior—to a class where some students say they know nothing at all?
Dr. Yogeshwaran Krishnan 0:10:55
So I. Will teach like. Yeah, the research problem is date was. Machine learning. Active learning. Things active learning is like or will get a data. One after that we just. With the different uh. Ensembles, it's like average everything. I went from that handsome Bell will do the molecular dynamics.
Dr. Yogeshwaran Krishnan 0:11:01
From molecular dynamics will. Take a data again. Using. Active learning.
Interviewer 0:10:58
Okay, so you tackled a machine learning problem, used active learning with ensembles,
Interviewer 0:11:03
So you're looping between molecular dynamics simulations and active learning to refine your data set. What was the key insight or result that came out of that approach?
Dr. Yogeshwaran Krishnan 0:11:34
In the beginning we had a really big problem. Like we have very less data.
Interviewer 0:11:49
Okay,
Dr. Yogeshwaran Krishnan 0:12:26
But when we give more data. The active learning. From Macdulonic we took a very good. Uh, ensembles and we got a very good potential from that. Ensembles. Yeah, when we did her research. Uh, when we do, uh, very low. Data after that using active learning. Created more data, more and more and more data. From that more data we got a more data.
Interviewer 0:12:29
So by ramping up the data with active learning and molecular dynamics, you improved your ensemble's predictive power. Let’s go back to guiding student research—when supervising a student on a project like this,
Dr. Yogeshwaran Krishnan 0:13:11
So the basic initial training data we got more. The potential work more. Yeah, it's a very good question. Actually when I do the teach. The students. First I will understand. What problem I gave? And I can't judge their ability to do. Ho ho ho I I?
Dr. Yogeshwaran Krishnan 0:13:17
Investigator. I know which one they can do. Or which one I can assess?
Interviewer 0:13:13
So you try not to judge a student's ability at the start and instead focus on understanding the problem you've assigned. When you're mentoring students, how do you encourage them to take ownership of their research while making sure they stay on track and meet deadlines?
Interviewer 0:13:18
Alright.
Dr. Yogeshwaran Krishnan 0:13:40
Yeah, actually. I will keep every weekly update. And monthly update.
Dr. Yogeshwaran Krishnan 0:13:48
So I will understand whether they're going on the track.
Interviewer 0:13:43
Weekly and monthly updates are a good way to keep students accountable.
Dr. Yogeshwaran Krishnan 0:14:08
Yeah. We that try to do. Metal air battery.
Interviewer 0:14:04
Right, so those regular check-ins help you see if students are making progress. Thinking about your own PhD specialization,
Interviewer 0:14:10
Alright, so you tried to work on metal-air batteries. What was the main obstacle you faced in that research, and how did you address it?
Dr. Yogeshwaran Krishnan 0:14:41
Yeah, in metal air battery we need to do. Too much solvent and salt.
Dr. Yogeshwaran Krishnan 0:14:54.0
Yeah, in metal air battery we need to do. Too much solvent and salt. So it is very hard to train. Machine learning potential. For that, first we try to do only with the solvent.
Interviewer 0:14:54
So you found that solvent and salt optimization was a big issue in your metal-air battery work. How did you go about selecting or testing different solvent-salt combinations, and what did you learn from those experiments that shaped your next steps?
Interviewer 0:14:56
So you found it was too complex to train a machine learning model on both solvent and salt at once, so you started just with solvent. Shifting gears a bit,
Dr. Yogeshwaran Krishnan 0:15:50
Then we introduced a salt. Then we introduced the. Surface in that way we reduce the time. 2. Oh, we when we do theoretical and we know. Exactly in uh. Experimental result also. We can collaborate with the experimentalist and we can compare.
Dr. Yogeshwaran Krishnan 0:15:55
With the error. Like ±.1 or .2.
Interviewer 0:15:53
Alright, you mentioned collaborating with experimentalists to compare your theoretical results with experimental outcomes—that's a solid integration.
Interviewer 0:15:57
So you’re saying the error is within about point one or point two—pretty tight agreement. Thinking about your research publications, which paper are you most proud of, and what made it particularly significant in your field?
Dr. Yogeshwaran Krishnan 0:16:37
We published. Uh, with a curator package.
Dr. Yogeshwaran Krishnan 0:16:51.0
We published. Uh, with a curator package. That package is the one, uh, we did, uh, all the group used that, uh, package.
Dr. Yogeshwaran Krishnan 0:17:12
Also, we published that, so we are so proud of that because. All the. Uh, people in my department used that package. So Yes, please tell me about the lab. I am going to work with Yes, please tell me about the lab. I am going to work with. Please tell me.
Interviewer 0:16:51
You published work using a curator package—that's a concrete achievement. Thinking back, what was the most challenging aspect of preparing that research for publication, especially in terms of ensuring it met the standards of a reputed journal?
Dr. Yogeshwaran Krishnan 0:17:02.0
Also, we published that, so we are so proud of that because. All the. Uh, people in my department used that package.
Interviewer 0:17:02
So your group developed that curator package and it’s now widely used by others—definitely something to be proud of. What was the process for deciding which features or functions to include in the package, and how did you balance your own research needs against the broader needs of the community?
Interviewer 0:17:11
Alright, Professor, we're wrapping up. Is there anything you'd like to ask about the role or the team before we finish?
Interviewer 0:18:05
You'd be joining a lab that's focused on both classroom teaching and hands-on research—so you'd have access to facilities for experimental chemistry, plus opportunities to guide student projects and collaborate on emerging technology specializations.
Dr. Yogeshwaran Krishnan 0:18:43
About the lab I'm going to work with. For the department I'm going to work with, please explain about the. No, I'm good with that. Thank you.