Hello, Arjun. Yes, sure. Let's begin.

Please start by introducing your professional journey in academia

OK, I'm Doctor Juhi Dutta. I'm currently working as a postdoctoral fellow at National Taiwan University in Taiwan. Since December 2025.

Okay, so you're a postdoc at National Taiwan University since December 2025. Can you share a bit more about what led you to this position,

Prior to this, I have completed my PhD from Harishchandra Research. So I have been focusing on theoretical high energy particle and astroparticle physics and phenomenology right from my PhD at HRI in 2019. From where I moved to Hamburg. At Daisy and University of Hamburg, and to Oklahoma and to IMSC Chennai and then to NTU. And throughout I have been working on. Various aspects of connecting theoretical particle physics models. To experimental signatures at present and future experimental facilities like the Large Hadron Collider experiment currently at CERN.

You've worked at a range of top research centers, always on the theoretical side of high-energy physics. When you're teaching or mentoring,

And we have also been working on presenting. Future signatures which can be seen at upcoming proposed colliders. So yes, I I actually love working with students and I have been mentoring them right through the start of my post grad, post doctoral experience. And I find it very fascinating to introduce quantum mechanics to them. Especially when we talk about the ultraviolet catastrophe that led flying to propose the. Planck's hypothesis that led Einstein to propose the Planck hypothesis and the quantization of energy. So that's a very basic principle which I think students. It showed me how I transitioned from classical mechanics to quantum mechanics, so. I try to emphasize students to try to think about things more fundamentally. Like in particle physics, we can compute simple processes to find the amplitude of. Cross sections. And. I try to make them workout simple such processes to understand the standard model processes. Like plus C minus going to. ZH or WW These are very standard. Things which we can learn analytically in particle physics. And. Once the student has a handle from the particle theory course, he or she can try to. Now step into the research world where. They can start using different computational tools which are available and try to understand what is. The deeper in these tools because these tools are. Are very necessary for complicated particle collision experiments that we have to simulate. And they are.

Right, so you really emphasize helping students make that leap from classical to quantum thinking, especially using examples like the ultraviolet catastrophe and Planck's hypothesis. When you're guiding students through those first calculations of, say, electron-positron annihilation or production cross-sections, what's a common misconception you see them struggle with, and how do you address it?

So I I try to make them focus on analytical aspects of these kind of processes. I have also worked with students on dark matter pedimentology, so I try. So for electron positron annihilations the processes it's just electron positron and they are fundamental particle itself. I think this is. Quite easy for the students to understand. The problem arises when one has to look at. Proton proton collisions which where protons are a collection of quarks, so. That is where the concept of folding the pattern partonic cross section with. Extra part on distribution functions come in. And, and I think this is something where students learn. The different PDF functions which are there? What are the quark? The valence core contributions, the C core contributions and how these? PDS differ for each of these. The PDF contribution of each of these quirks and C quirks and valence quirk differs so. In my experience, I have found students. Taking a bit more time to grasp these kind. The concepts and. Learning through which of the core collisions are the dominant processes? Or not. So these concepts will be clearer when one starts to. Understand more about these valence quirks or the C quarks which are the dominant contributions at what energies? So these are all energy dependent.

So. I find students are able to relate more to the fundamental ones and then slowly build their understanding. With that information.

Alright, so you notice students often struggle when moving from the simplicity of electron-positron collisions to the complexity of proton-proton collisions, especially grasping parton distribution functions and how energy changes which quarks dominate.

Yeah, starting students with fundamental processes and building up makes sense, especially given the complexity of proton collisions. When you introduce computational tools for particle physics simulations, how do you ensure students don't just treat them as black boxes but actually understand the physics behind the outputs?

Yes, that's a very good question. So let me first start with the proton part of the story. So of course the proton distribution functions are energy dependent because. As the energy increases, suppose the valence core content in a proton is UUD, so the eu quark and the de quark and there are. Huang's also which carry a large part of the energy. So as high energy as the energy increases, the gluons can split into other sea quarks. And so the population of these quarks keep changing as one goes to different energies. So that is why these PDFs are mainly a distribution in in energy and one can see that the C quarks can start being more. When there is high energy.

So. So this is something we learned in the particle physics course. With structure, functions and etc. So students can students have some theoretical knowledge about this and? When coming tools we try. To look at the theoretical validation.

Professor, your recent publication “Pseudoscalar contributions to Zh production at the LHC at 95 GeV and above” caught my eye. Can you walk me through how you’d explain the core ideas of that work to an advanced undergraduate class who may not have much background in Higgs physics?

So you’re tying the changing quark content at high energies back to what students see in theory versus simulation. For your paper on pseudoscalar contributions to Zh production, how do you help students distinguish between theoretical predictions and the uncertainties or limitations of experimental data when discussing results like those?

Yes. So that's a very nice question. So the chapter production paper that we have. In Standard model, we know that from the at LHC, when the proton collision phase, the standard model contribution will only come from. The the standard model Ch contribution which will have the. Contribution from which are suppressed contributions like 1 contribution would be the QQ bar to ZH. There's channel process and there'll be also contributions from the glue, glue which will be by a box diagram, which is much more suppressed compared to QQ. So these are the expected diagrams. What we are trying to see is suppose SCMS or Atlas has measured a number for this cross section and there is some uncertainty with this number. So right now the number is quite. Close to the standard model, but there are still some errors. So our idea is in the new model like 2 HDM extension, there is something called the pseudoscalar which is an extra CPU part. And this particle can also be produced via Buon fusion and BK to a ZH. So we are trying to see whether this. Added contribution of the pseudoscalar. Is within the error with which CNS or Atlas has measured. So if if our contributions are. Already very large, then we know that the pseudoscalar of a certain mass is already excluded from data.

So we try to look for the prediction of this kind of a pseudoscalar contribution and see if there is any new physics in this. So our goal is to understand. How much of the new physics contribution can 1 accommodate in the ZH production construction when it comes from a pseudoscalar? I would try to help the students learn first about the. How the standard model cross section is? Computed. What are the diagrams? It's flowing and then. Talk about. Confidence level. So I think right now in high energy physics. Yes. So VIT is a very premier institute with a lot of focus on research and teaching. So I think this is a very good opportunity for me as well, so. So I plan to. Plan to apply for the anrf research grant, especially the Prime Minister early career research grants and there's a lot of push in high energy theory right now in. The Collider Physics using machine learning. And I have been a Collider physicist and also working on dark matter and Collider interplay. So I think my current studies which I am doing, I also use machine learning now. To improve several of these dark matter searches, so I will push towards applying for funding from an RF using. Parent research and implementing machine learning to improve those research directions in terms of Collider prospects. I will also. Also, uh, try to work more on. Interdisciplinary aspects.

Makes sense. Now, thinking about your research in theoretical high-energy physics and phenomenology, where do you see the most promising opportunities for securing external funding over the next few years? Which grants or funding bodies would you target,

Alright, Professor, let’s turn to pedagogy. Suppose you’re teaching a large undergraduate physics class—maybe 200 students—and you can’t use slides.

Such as impact of quantum computation and high energy, which is also a very upcoming and highly funded area. Of research so. So I would. I actually would prefer more of blackboard teaching. In this situation and in that case, I would want to engage students by asking them questions or making it a more interactive session. I have done this in my tutorials when I was a teaching assistant in my PhD. I try to give them more problems. Or we start a small, teach something of a concept, and then try to give a hands on calculation to them. And try to ask them to work something out on the board. Call a student up and also maybe others can join in in. Contributing ideas.

Okay, so you like to use blackboard sessions and get students working out problems in front of their peers, which keeps things interactive. How do you gauge if that approach is actually helping the quieter students engage with the material, especially in a large group?

And uh. And I would want to make it more interactive by having more quiz sessions in the class to test our understanding continuously. So I will. So I will try to call. So if there is a large group, I will not only. Be asking the most engaged students are the ones or the most vocal. I will try to call up. A mix of students, it can be even the quieter student and they can try to maybe not on the board they can try to work it out on. Own puppies, but in the. Point is not to put someone on the board and. Make them an exam. Start an interactive session, so. He or she can get started and his friends can join in. We can all start thinking about how to proceed with this problem.

Okay, Professor, you've talked a lot about your research and teaching, but there's another important aspect to this role.

Or the quietest student can. We try to solve it and I'm always available for questions. So we can always discuss. What do you mean by outcome assessment data? Is it research outcome?

Yeah, fair question. By outcome assessment data,

So in that case I will try to focus on what are the places where. Where there are gaps. So I would first try to check what are the gaps where exactly the students are not doing well, whether it's some particular area of work which needs more emphasis or if there is some. Some exams which are too tough for them, so I will try to also look for maybe implementing continuous assessments. With assignments or quizzes intermittently so that students can really test their day-to-day knowledge. And exams are not really a huge, huge shock. Where? One really does not know how to cope up with it. So I will try to do these kind of continuous assessments. And have more tutorials Make students or a group of students try to. Do some work together so that they can help each other out.

So in that case I will try to focus on what are the places where. Where there are gaps. So I would first try to check what are the gaps where exactly the students are not doing well, whether it's some particular area of work which needs more emphasis or if there is some. Some exams which are too tough for them, so I will try to also look for maybe implementing continuous assessments. With assignments or quizzes intermittently so that students can really test their day-to-day knowledge. And exams are not really a huge, huge shock. Where? One really does not know how to cope up with it. So I will try to do these kind of continuous assessments.

So you'd start by pinpointing exactly where students aren't meeting learning outcomes, maybe by analyzing which areas or assessments show the biggest gaps, and then use more frequent assignments or quizzes to smooth out the pressure of big exams.

Alright, Professor, let me give you a scenario. Imagine a student files a formal complaint alleging your grading is biased, but at the same time, your department head is pressuring you to improve overall pass rates. How would you navigate that situation and balance academic integrity with institutional expectations?

So, uh. Firstly I would ask the student what is the bias because I I have invigilated and I'm very particular about why and how. In the Max I will try to explain to the students where exactly. The marks they are deducted and why they are deducted. I would be pointing that out exactly.

Right, you’ve covered both fairness to students and holding your ground on academic standards. Now, let’s look at industry alignment. Given your background in high-energy physics and quantum computation, do you have any collaborations or contacts in industry that could translate into internship or placement opportunities for students?

So that's one way and I will also. Try to of course be very transparent so I I can explain where exactly why I have deducted Marks and try to help them learn. A lot of particle physicists. Who are also among my collaborators and my communities. So a lot of my seniors and juniors, some of them also transition to finance roles like data scientists. Computational scientists. In the finance sector or even in other industrial sectors. So I do have friends and. My colleagues who work in such places, and I think they will also be. Happy to have more. Interest if if the students want to, they can. Apply to these places I can help. Connect to them. Or we also have industry seminars from these. These particular professionals who can come and visit our institute at bit. Give the students an understanding of what work beyond academia looks like so that they can. Make a conscious choice of. If it aligns with them. But yes, I think particle physics has has. Potential in in other sectors as well besides academics and.

Lots of people also join for. Other cases like Google and. Many other big data companies. So data scientists, R&D, these are common.

Alright, Professor, let's shift gears and talk about some core skills. Suppose you're designing a course module on semiconductor device physics. If you had to demonstrate band structure effects using a hands-on experiment or lab,

Right, so you’ve seen students from particle physics move into places like Google, big data, and finance, thanks to their computational skills. When you teach machine learning concepts to physics students, what practical project or assignment would you give them to bridge the gap between theory and real-world application?

Yes. So that's a nice question. So I will mainly try to. Focus on fundamentals. So even in machine learning, it's it's the kind of linear regression. Problem or classification problem? I can try to give them a small data set of of something they are working on me. If they're in particle physics, If they're studying particle physics, one can try to just have a simple. A simple process like D plus E minus going to say Z8 or. Whatever so they can start with that process and generate the kind of signal data for this. With say just the cross section and some PT of the Z or PT of the H. So I will try to. Guide them to taking a small problem and take some of these. Events as data set and then try to build a machine learning algorithm using a simple linear regression or classification. 1st we have to teach them the concept of loss function. And how one sets up a small network. So I will try to cover these. And as test data we can use some simple data sets. Which either are physically motivated, then it will be. Much easier for students to connect.

2 So that is the way I will try to. Get started on this kind of computational techniques.

Okay,

So you’d start with fundamentals and a physics-based dataset, like a simple event classification from particle collisions, and walk students through implementing basic models and loss functions using that real data. When you introduce quantum computation to students who’ve never seen it before, how do you make those abstract concepts accessible without overwhelming them?

So quantum computation and high energy physics is a rather emerging new field. But there are many physical processes where Atlas and CMS for instance have. Have put. Some limits on the on using entanglement and observables which show entanglement. So one of these variables are like polarization variables of processes like H2Z star. And from the Z there can be a dialect turn and one can try to construct the different different. Polarizations of the Z boson and construct observables using these decays. And. More specifically, angular observables and using that they can construct the polarization matrix. So this is a slightly more advanced, but I would try to again. Maybe connect to such a physically motivated process which has been studied and then. Try to. Because these kind of polarization matrices or these density matrices, they are taught in quantum mechanics early on. But uh, yeah, I think those concepts. Are slightly dot in quantum mechanics? We have to build on that theoretically, but maybe. Connecting to this kind of a physical process and measuring. Observable which sort of. Encapsulates the information about entanglement.

So you’d connect the abstract math to real-world measurements, like polarization in actual particle detector data, and have students use density matrices they’ve seen before to measure entanglement observables. That’s a pretty advanced bridge. For your teaching in general, what’s one specific thing you do to help struggling students catch up without slowing down the rest of the class?

And understanding how. About that could be a good way to motivate them to look at the physics of it instead of just. Exactly. OK, so. I would uh, I would not jump to the density meter. I will try to 1st go through the quantum mechanics and introduce density matrix as a teaching concept. But yeah in case students are struggling. So I try to. I try to help them through then so I can. The setup tutorials where they can or office hours where they can come and talk to me about any specific part they have a problem with or I can set up a tutorial where. Everyone can discuss. Which is outside the class R so all the students can have different problems. So it will also be a fairground for everyone to ask questions which will also make the student more comfortable. In sharing their problems if they are more quiet, so I think a more interactive session with everyone. Or if we have tutors at KIT, then one can also have more tutorial sessions with them. But yeah, they're always free to come and talk to me and discuss problems.

Wherever they have in understanding and I will try to have more of these quizzes so that they can immediately test.

So you’d hold extra tutorials and office hours, making sure even the quiet students have a chance to catch up, and encourage open discussion so everyone’s comfortable asking questions. For your own research, Professor, can you give me an example of an industry project or consultancy you’ve been involved in, and how you brought that experience back to enrich your teaching?

So you use immediate quizzes to help students see where they're struggling right away, not just after a big test or assignment. Now, thinking about semiconductor device physics,

So I have not yet been a part of industry. I have been in academia. MMM. So. To get actually involved, of course I need access to the lab where they can test.

Junction. They have several experiments where one can test reverse bias and. All the way things.

Okay, so you haven't had direct industry experience yet, but you'd want students to get hands-on in a lab if possible.

Right, so you’d have them experiment with different biasing conditions on the p-n junction to see real effects. Now thinking about industry collaborations—even if you haven’t done consultancy yet, how would you set up a project or partnership to give students exposure to real-world physics applications?

Yes. So in that case I think I would attempt to. I would want them to connect to. My colleagues are who have. Transitioned into the industry so I would try to set up lectures by them, maybe on video conference or zoom. Given that some of them are abroad and I would try to. Get them to give lectures so that students actually know how their particle physics skills from academia can transition and how they use those skills so. The model building skills, the statistical skills, the computational skills, How you think, not just the computational part, how you analytically approach to solve, approach a problem and solve it. These are the kind of techniques which are useful in industry. Based on whatever problems they give.

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

So there are people who are into banking or into. Into other fine fintech organizations so are in computational centers. I would want them to. So if possible, can you tell me what are your expectations and? What is the typical role?

Of what kind of how much teaching and how much research am I expected to perform and. Yeah. Am I allowed access to labs?

You're right, a lot of our students end up in fields like banking, fintech,

Okay, so you're wondering about the balance of teaching versus research and whether you'll have lab access. Teaching load for assistant professors here is typically two courses per semester, with flexibility to pursue research throughout the year, and yes, lab access is available subject to safety and scheduling protocols. Is there anything else you'd like to know about the department or expectations?

So since I'm a theoretical physicist and some part of my studies require unavoidable high scale simulations, so. Are am I also? Allowed to use the cluster access or. Yeah, such kind of facilities at PIT. And I would love to offer bridge courses for students interested in research, say in the dark matter or particle physics or the. The work I do or how to transition, so if students are interested, I would be happy to mentor students. With reading projects. And maybe schedule such courses, design such new courses. So is that something that? Is also possible. Hello.