The future is DeepTech | L Tulasi Gandikota

L Tulasi Gandikota (Guest): So DeepTech is basically, if there is a science-based innovation, then it's called DeepTech.

In traditional engineering, let's take solar as an example, because I come from the chemical engineering background, let's say solar energy. If you really look at it, what is a fundamental element there that is called the crystalline silicon cell.

So what does it do? It converts light energy into electricity and if you really look at the solar system that we have on our rooftop, what is called a rooftop solar, what it has is it has the solar panel, which is nothing but an area of these cells, which will convert light into electricity using photoelectric effect, and then through them you have a system that will connect all these multiple cells.

Then there is inverter because it produces DC current and what we use in our daily life is AC current. So there is an inverter and then there is a battery to save, during the day you will have solar.

If there is sunlight, then you will have sunlight. If you need uninterrupted power, you need a battery to store the energy and use the battery when there is no sunshine. Or otherwise if you're connected to the grid, then the grid will supply the current whenever there is no solar in it.

If you really look at the fundamental cell, crystalline silicon cell that was in the lab was in the 1950s. So the solar PV as a thing was based on some scientific discoveries or the innovations happened in the 1950s, but today if you really look, there's so much development happened and both efficiency has improved from whatever, 2%, 3% that it was in the fifties to today, close to 30%, so that it is competitive to other power technologies.

Then if you look at it what is today, it has become like a more like an engineering problem in the sense you, everything is established. There is no unknown here. Today, if you have to put a solar rooftop, there are a lot of suppliers and there is lot of standardization that happened and there are knobs, etc., so that somebody who needs to set up this solar power on your rooftop today is a simple engineering problem. That is somebody will have, you have some rules that follow those rules and there are problems. You solve those problems and put it.

But where is DeepTech today? If you really look at this whole solar thing, so the conventional solar, which is based on the crystalline silicon has some defects, for example, is too expensive although the cost of making it has come down, it's still expensive and it is like a rigid plate.

So you want to put it on a flexible solar so we have a startup called ABX3 PV, which is working on flexible solar, which is based on a new material, which is perovskite.

So there are advances in the materials that are used from, let's say, crystalline silicon in the case of solar PV to perovskite.

So the DeepTech today, flexible solar cells are DeepTech. Maybe down the line that becomes an engineering solution.

Similarly to really look at the software, just doing simple software with certain to solve a problem or to provide a service is software. But there are advanced software like you probably we hear every day about AI ML (Artificial Intelligence and Machine Learning).

So how do you write code to bring about efficiency or to bring about something that is a little bit of unknown is requires a lot of research and it requires you everywhere.

You need to solve problems, whether it in software or engineering or anything, you have to have the software. But the amount of unknown and the amount of research that is required to bring a solution or a product is DeepTech.

L Tulasi Gandikota (Guest): I think today there is a lot of awareness, thankfully about DeepTech. But I think just being whatever you are studying in your textbooks. So these days the textbooks are quite nice for the school students especially.

There are a lot of activities which is basically called as experiential learning. So students should not read them as another theory or a concept, but actually use the facilities available to them. It doesn't have to be very fancy. They don't need very fancy labs. Use whatever you have at home, whether it is in the kitchen or is in the small tool set, or go to a lab that allows you to tinkering labs.

 Some of them are quite well equipped. So just basically experiment with your hands, whatever activities such, for example, I use a photoelectric effect, how to make a PN junction? How to make, how to convert, how to prove photoelectric effect or Newton's laws or anything that you'll study.

Just don't read them as a theoretical or memorizing idea, but actually try to do with your hands and learn through that experiential learning.

That is the first and the starting point. If you're a school student, if as you grow, you become an engineer or you take a science career, look around, and see the problems that are around you.

Whether it is water, for example, what are the basic needs of human being?

It is water, clothes, shelter. So we have certain way of doing things, but there could be challenges in that way.

For example, water is a problem.

But there is abundant water in which I see. How can we convert? There are reverse osmosis technologies that are available.

Are there challenges with those? How can we solve this problem? So just being aware and curious about your surroundings, knowing, identifying the problems that we as a society face and trying to use science to solve the problems is a great way to get into DeepTech.

L Tulasi Gandikota (Guest): I think curiosity, just being curious to know, to learn to ask and asking questions and having the problem-solving mindset, as I told you before and exposure definitely matters.

For example, places like Indian Institute of Science (IISc) and other institutes open up their campus to everybody on one day there are whatever and there's a lot of information available on the internet. There are a lot of online educational videos through different channels. So just being curious to learn and not afraid to ask questions, whether it is to the teachers, to your parents, to whomever you come across and not have any fear.

Just try and be safe. Of course, nobody wants anybody not to try, but you have to be courageous enough to just try with your hands to do and in addition to this, as I said, the schools have labs. The colleges have the incubation centers and places like IISc have these incubation cell and innovation cells.

So take advantage of these kind of things that are available to you, participate in these hackathons. For example, Government of India, especially for the science engineering students, they kind of put these hackathons, which are the problems faced by different government departments and the private companies.

Participate and learn teamwork. Don't have to solve everything by yourself. Form up with others. Learn from each other. And this is not just for the science students because end of the day to take a research into the market of the lab to market takes everybody to come together. As I said about, again, I take the solar cell example.

So when it was, initially it was only a scientific problem to be able to get the photoelectric effect and be able to make the silicon cell

Then but then you really need a whole infrastructure in terms of developing to markets, you need creative people to make the right products and you need accounting people to make sure that the right accounting metrics are followed and you need the whole, as they say, a lot of different skills to take any research into the market and not and that is not unique to DeepTech in that sense.

L Tulasi Gandikota (Guest): Yeah, I think that maybe because India has been doing a lot of research. We have many research output universities. Some are really good, some are not. But there is a lot of research happens. So mostly it measured through publications, but the real value to the society only comes when we translate this research into products.

As I said, let's take the semiconductors, some other industry where I had some exposure to. So what can students do?

They can design the semiconductor. Designs are what they call as integrated circuits, and there's a lot of innovation that is required, but that doesn't require you to have high-end infrastructure like fabrication plants etc.

But if you really want to make a real good DeepTech science-based product, you need engineering skill, you need research infrastructure, and you also need an entrepreneurial mindset.

In the end, it is not just enough to know the theory, but you also need to know who needs it. For example, the simple questions that we ask students is, what problem are you solving?

For whom is it a problem? Are you the right person to solve the problem? And if you solve this problem, who would be the person or the entity that outing and will they pay money for you? Because end of the day, entrepreneurship will not serve if you don't make money.

So these are the simple questions, but requires a deep thinking and the commitment, perseverance you asked about the skills I mentioned is perseverance, as you might have heard, the famous thing that I failed a thousand times before I succeeded. So that fear of failure should not be there. It's okay. It's part of the learning.

L Tulasi Gandikota (Guest): Yeah, most of the students that we get in my IISc are the science students, science and engineering students. So the kind of at least what I think they should be really excelling, whatever they're doing, whatever field they're doing, because we have wide variety of what they saying, the science and engineering fields.

Ranging from the pure science in physics, biology, chemistry to the all the engineerings thing, which covers the applications in all the DeepTech areas, be it AI ML.

Everybody talks about it, or be it the space or the advanced materials or bio and MedTech. It covers entire applications can be in any of these domains.

So what I think what we are seeing is there is a lot more interest now.

From even from places like IISc, which was predominantly a research institute. Basically making academicians or the faculty to now there is lot more interest, both from students and the faculty to see how their expertise can be translated into products.

Although IISc has a very huge history of setting up several institutions are enabling, setting up several institutions. In India, but still, I think there is a renewed focus predominantly driven by the ecosystem as a whole with government driving a lot of initiatives to promote DeepTech entrepreneurship. So we see the same thing here as well!

L Tulasi Gandikota (Guest): What we hear is the emerging tech is the quantum technology is one thing that is very much in the news. The general AI ML as one of the emerging in this space, Generative AI or whatever you call the underlying sub themes and thought and there is lot of emphasis on the new materials as well because the climate tech, climate tech is very generic good.

But anything that will reduce the footprint because of our usage, be it on the energy, is it on the environment, or any of these things generally being more benign, I would say.

L Tulasi Gandikota (Guest): I don't think there is an easy answer, but it depends on the individual, what their interests are, what do they want to solve? Because as I see, some of these are more of application domains.

For example, the fundamentals like still the basics. The physics, chemistry, biology, and interplay between all this. There is no real boundary between any of these if you really see it.

For example, whatever we call today as AI ML or they're all based on math. The underlying thing is algorithms and how these algorithms really do is a fundamental to it is based in math, but application on this AI could be into healthcare, could be into energy, could be into space, could be anything.

So I think it could really help for students to develop that clarity by reading, by themselves, getting exposure to the research labs and the research infrastructure in the country, as well as getting exposed to the startup ecosystem as well and there are a lot of programs that government is doing to promote this awareness as well as promote a culture of innovation and entrepreneurship across the academic institute and for the startup ecosystem.

L Tulasi Gandikota (Guest): I said, you see what is deep? It's based on science innovation. The fundamental science. So you have to be really expert if you want to be the technical lead in any of these DeepTech areas. So you have to have an expertise.

That expertise could be, as I said, in materials. The expertise could be in writing code. The expertise could be in understanding the fundamental physics. Like in quantum technologies, you really need to understand the underlying physics of these materials. The communication part of it. So that is a fundamental that developing an expertise, if you want on the technical side, if you want to be on the business side, which is what DeepTech entrepreneurship is all about, then you also need to have an understanding of, as we talked about, the customers.

The customers doesn't always have to be individuals. The customers can be companies as well. What problems they are facing. How can I use my Deep Technical expertise to solve a problem? To give you an example, let us say in the healthcare, just using the things that I know of. In the healthcare, so there is a problem of, let's say the diagnostics, cancer detection.

So today what does the cancer detection is once somebody kind of symptoms are. So then you kind of do, can we detect cancer at the early stage, even before it manifests as some of the symptoms? So what does it take, or can we simplify the diagnostics in terms of just taking a blood sample and detecting cancer, or predicting that this person could have cancer.

What is the probability that this person will get cancer in future? So some of these require you to have can you kind of fundamental shift in the way you approach these things. So some of our startups or some of our research in this area is towards those things in detection itself.

But let's say you want to cure. How can you cure today it is more chemotherapy or to do some of these things which have its own positives and not so positives. How can you solve that problem?

Can you send a robot to the place where the cancer is happening? Can you just cure only that without any side effects? That's the kind of Holy Grail. It's curing. So in the whole spectrum, you take anything. If you have a fundamental expertise, technical expertise, you could look at multiple different fields and see how this technical expertise can be put to use. And once that understanding is developed, and then you need everybody else.

As I said, you need not just technical experts, but you need people with the business understanding or a customer understanding.

You need all the different fields. For example, the regulatory is a big thing in the MedTech space. You need people with that kind of background as well. So the careers are, as I said, divided. To your question on the careers, it can be technical experts or a researcher, or it could be the business people or it could be the people with legal, safety, finance, and accounting.

So you will have careers, even though it's like any other field. It has like required expertise of different people. Its different backgrounds. So I think the careers are quite varied and quite open.

L Tulasi Gandikota (Guest): What is AI? Basically it is able to learn from the data and be able to do some decision making like an intelligent way of doing that. So let's say you want to, maybe I'll just take an example of this thing. So there is again, it's much easier to explain in the healthcare. There is a lot of data, there's a lot of imaging that is being done for all kinds of diagnostics.

Right now we are dependent on the real expert in terms of the doctors to be able to identify the patterns. And probably correlate with the disease and then solution. So some of these drudgery or expertise can be simplified and much nicely presented to the clinician, in this case a doctor.

So with the help of AI, and then they, it saves a lot of their time because all these image analysis is done and is done much faster, much easier through this AI ML because it learns on its own. Let's say if this pattern has been kind of more likely, this is the problem, then they will present that, and the role of clinician would be then to kind of take this and correlate using his own expertise to be able to diagnose and maybe saves time for the clinician.

In the case of healthcare, that's one example, but you can see in every field, for example, you're doing project management and AI could make a lot of these work easier for you by just putting the information in. It could be even simple application where you are trying to read multiple pages to prepare some simple document.

So again, as I said, can be varied from a simple task to a very complex task and the human or the expert's role can be more of using their expertise and reducing their time.