Full Transcript

Speaker 7: This is a test.

Utsav Somani: Hey listeners, welcome to the second stream of 2026. This is Monday, the 12th of Jan, and we have an exciting lineup for you today. How was your weekend?

Dhruv Sharma: Very good. It was a good weekend. So I went to something called the World Book Fair that happened in Delhi. It had, you know, counters and booths from all the major publishers, but then people said, even though there was Penguin and Harper and Hatchet and what have you, the most popular booth was Chole Bhature, because winters in Delhi.

Utsav Somani: India mein mirch masala vikta hai, actually, basically, I think that's the crux of the thing.

Dhruv Sharma: Samajh gaya hai sab logo.

Utsav Somani: All right. So let's welcome our first guest. We'll cover news at the end today. We've got an exciting guest to break down, it's an article that he's written on Twitter, and which I think is very, very topical because A16Z has also raised a massive fund. So let's welcome Anmol. Anmol, welcome.

Anmol Maini (Untitled Ventures): Hello. Hi, guys.

Utsav Somani: So tell us about the consensus capital article that you've written.

Anmol Maini (Untitled Ventures): So I think about a week ago, I saw Kyle Harrison, who is a VC at a firm called Contrary Capital, talk about how a very small subset of firms raise a majority of the capital that VC firms in America raise. And I was just curious what that ratio might look like for India. And it turns out India is a very concentrated ecosystem where there's just a handful of firms that raise a majority of the capital that goes into backing venture companies. And these are the same firms that typically then end up backing some of the largest winners in venture capital in India.

Utsav Somani: So A16Z raised 15 billion, which is almost, I think, 18 percent of all the VC dollars raised in 2025, which is pretty wild. And then Kyle Harrison, I think, mentions in his tweet where he says that 75 percent of all venture dollars can be raised by just 30 firms. So the word that he used was consensus. And you mentioned in your article consensus capital as well. What's the look on that?

Anmol Maini (Untitled Ventures): Well, I think if you kind of go through some of the conversation that's been happening on Twitter over the last year or so, it definitely does seem like every year there's fewer companies raising capital, but the capital that is going towards them is certainly increasing. In India also.

Utsav Somani: Concentration basically. Yeah.

Anmol Maini (Untitled Ventures): And in India, we've had, be it privately valued unicorns or all these unicorns that have gone public. They come from categories that are fairly well understood or in an essence, a consensus category. Now, these categories might or some companies within those categories might not be consensus when they were started. Like maybe Zomato in 2009 was not a consensus company. But by the time Sequoia was investing and Swiggy was raising growth capital, that had turned into a fairly consensus category. And so what I was kind of just pointing out is that ultimately a lot of the winners in venture in India end up happening in consensus categories and we don't have firms or companies being backed that are kind of contrarian or outliers to the very end of them going public.

Dhruv Sharma: And were you convinced of this? I was just going to ask him. Before that, I think I want to plant a conspiracy theory today, which is, I think the A16ZGP that they have been trying to embed for a long time, I think Anmol might be that GP. But the question I was going to ask is.

Utsav Somani: He's been super hard to reach. Like Anmol doesn't respond to us anyway. So yeah, he might be like, yeah, playing hard to get.

Dhruv Sharma: Playing hard to get, yes. But Anmol, so are you forwarding this as a, like, are you convinced of the theory or is this just an open question you want everyone in the ecosystem to engage with?

Utsav Somani: I think Anmol, you mentioned, I think the Manish Singh article as well, where he said the past decade of returns have come from basically these consensus categories. And I think the firms are raising or rightsizing their fund sizes in India, because where does the next decade of returns come from? I think that's a question that he leaves the article with at the end.

Anmol Maini (Untitled Ventures): I think that, like, it's certainly, I don't know if, like, for example, Peak has over two and a half billion to return in a single fund. So I don't know if that's the proper sizing or where that nets out. But I definitely think that there has not been a lot of competition for the top six, seven firms over kind of their inception in India. And there doesn't seem at the moment to be a lot of competition trying to take them on. Like, I think the closest example we have of this is Tiger kind of being less relevant in India in the last couple of years and GC making a much stronger push into the country. But that doesn't seem to result in Peak or Excel or Nexus or Z47 being any less relevant in the country. So I think we're still a ways away from seeing if that actually happens. And given how all these firms seem to be getting allocation in the best companies, it seems hard to think how they'll kind of lose relevance.

Utsav Somani: And one stat that you mentioned, I think two interesting stats that your article mentions is just seven percent of the venture capital firms in India led the series A rounds of 60 percent of Unicorns in India. And there are only 12 Unicorns, 12 percent of Unicorns from the whole Unicorn universe in India that did not raise capital from the 16 firms, which have raised 70 percent of the Indian venture capital dollars, basically. Yeah.

Anmol Maini (Untitled Ventures): Yeah. I mean, so I think it kind of just proves the point that if you want to build a very, very large company, you kind of have to raise from one of these firms as early as possible.

Dhruv Sharma: At some point or the other, even if they don't lead your seed or A, they come back to join the capital and lead a round.

Anmol Maini (Untitled Ventures): Yeah. And but it does seem that it is probably beneficial for them to back you as early as possible, possible from their point of view, because they get a lower cost of capital for the same amount of equity in your company.

Utsav Somani: But what I'm reading is basically like it's signaling, right? I mean, so you think that signaling in the world of VC exists?

Anmol Maini (Untitled Ventures): It definitely seems like that. I don't think there's any denying that signaling helps successive fundraising. And I don't think India's India's unique in this, I think in the U.S. as well, like when you hear, oh, Chetan at Benchmark's leading a deal like your next round, investors are pretty much already trying to line up to invest in that company.

Utsav Somani: It's sort of like this flywheel where LPs concentrate capital in these funds, these funds concentrate capital via consensus, non-consensus into companies, and then exits come and then like it's just a flywheel, basically.

Dhruv Sharma: I can't think of an Indian counter to this argument, by the way. But there is one from Keith Rappoi where he quotes Peter Thiel and apparently Thiel says that in order for you to, you know, for you to have the best investments, you have to be right and everyone else is to be wrong. And so Keith's argument is if if these were truly consensus bids, then you know, the price would just go up and therefore any. But I think it's a point.

Utsav Somani: What Anmol said is Zomato back in the day wasn't consensus. You could get as much of Zomato as you want, like in terms of Infoedge investing in Zomato, but as it eventually became clear that food delivery has to become commonplace in India. So I think a factor of that.

Utsav Somani: Enough talk about the world of venture, but I think Anmol, if you were building a firm, which you are at Untitled Ventures, how would you break the club? Like if you have the aim or dream to like sort of break into this large fund journey?

Anmol Maini (Untitled Ventures): I think it's really hard to see how you would tackle these firms directly. I think you'd really need to come in with competing with them at all levels because just being the best series B firm in India isn't enough when you're competing with the top six, seven firms because they self-select the top series A's and then you won't get access to the top series B companies and so forth. So if you're kind of competing with them, you try to compete with them head on. I think it might be a little easier. And what we do a lot more is try to become feeder funds for some of these large venture firms because something that's definitely true in India is that the top six, seven, ten firms now don't want to invest any lesser than two, three million dollars in a company. So if there's a founder who only wants to raise a million or if the team's a little raw to not be able to raise a substantial seed round, that's where we try to play in. And I think you can still build a great firm just executing at the early inception stage while you are able to find categories that are consensus and companies that are non-consensus yet that go on to become consensus in the future.

Dhruv Sharma: I have a question to ask you guys, which is that, so Anmol, you said the top five or six firms, tier one firms in India have largely remained intact and uncontested. Do you think it has anything to do with the fact that the firms also haven't changed shape or form in the sense that, you know, if partners leave, it's just one partner here, one partner there. No firm is split through the middle. There's no breakaway faction. Nothing. That's what has happened.

Anmol Maini (Untitled Ventures): I think that's probably true to a degree. I think the only example there I can think of is Helion kind of being a big part of Indian venture in the early 2000s. And now that has kind of given birth to both Stellaris and Arkham. Now, we're probably still a couple of years away from both Arkham and Stellaris having a lot more unicorns and IPOs. But other than that, the large six, seven firms.

Dhruv Sharma: That's another name that comes to mind.

Anmol Maini (Untitled Ventures): Yeah. Yeah. But other than that, like the managing directors at most of these six, seven firms are still the managing directors in the current age compared to when they started in 2006.

Utsav Somani: Anmol, as a closing note, maybe you can give us a one minute description of what Untitled Ventures is for our listeners. Yeah.

Anmol Maini (Untitled Ventures): So we are a very early stage pre-seed fund. We started, both Vedika and I run the fund. We started a newsletter in the year 2020 and then we started Untitled in 2022. And our current stage, we primarily invest in consumer companies. So consumer brands, marketplaces, gaming companies and the like. And we started our first fund in 2022, which was a million dollar fund that we just ended up deploying last year. And now we're kind of setting up fund two this year.

Utsav Somani: All the very best. And last two investments, anything that's publicly available?

Anmol Maini (Untitled Ventures): Last couple of, I think, let me instead of last two, let's kind of talk about some of the recent hits. So I think Pronto is one of the companies we backed last year and they've been off to a flyer. And a company we invested in just announced their fundraise a couple of days ago. It's called Flint and they help you find great homes to live in in Bangalore.

Utsav Somani: All right. Let's have you back on the show very soon and we'll break down some of these investments that you made, Anmol. Thank you so much for coming on the show. Perfect. Thanks for having me, guys. All right, listeners, we're switching gears up a little bit. Our friend recommended our next guest, Boris of Virgin Ventures. He's invested in this company and we have Sree Poorna from Arctis Aerospace. Sree, welcome to the show. Hey. Yeah. So before I think we begin, there's a cool video that South Park Commons put up. Prateep was also on the show and that video was very, very cool and captures the journey. I mean, better than anyone can explain right now. So I think let's see the video for two minutes and then we'll have questions for you.

Shreepoorna Rao (Arctus): Yeah, sure. The first time I saw a space shuttle launch, I was in the fifth grade. Fire, smoke and this massive machine just lifting off the ground. After that, I was hooked. Space, aircraft, fighter jets. This was like magic. Years later, I built my first RC aircraft. The moment it lifted off, I knew. I don't just want to fly. I want to build the things that I do. Hi, my name is Sree Poorna and I'm the founder of Arctis Aerospace. We build unmanned airplanes that fly at 45,000 feet, can stay in the air for 24 hours. So, you know, take off today, land tomorrow and can carry 250 kgs of payload. We are the commercial unlock for real time, high fidelity Earth observation. While it is an engineering problem, it's also equally an art. The endurance and payload at this scale means we don't settle for off the shelf. We rethink everything. The airframes, the structures, the avionics. We rebuild the systems from the ground up. You need people who back you all the time and are as excited. If you look at our team, right, it's like a bunch of Red Bull credit engineers. We need someone who can match that energy and that's something, you know, I would say Aditya and Prateek and all of those guys have been able to do. We are young, driven and unafraid to take risks. If we can commercialize this at this price point, which we can do by building everything in-house, we can actually take a crack at it and make the idea of, you know, Earth observation being as ubiquitous as mobile data. Imagine having the capability to zoom in to Indira Nagar and seeing how's the traffic there. Real time on Google Maps. That is what we are enabling here.

Utsav Somani: Man, that's motivational stuff. So, I mean, you've explained, I mean, that pretty, pretty much explains what you do. But if given a two minute segment to give an elevator pitch, what does Arctospace do in your words?

Shreepoorna Rao (Arctus): So my elevator pitch usually goes like this, man. I just talk about the product a lot. And, you know, the use case just follows through. But yeah, our single line is we build these unmanned aircrafts flying at 45,000 feet, you know, 12, 13 kilometers above us. They take off today, they land tomorrow at the same time. So 24 hours of endurance. And, you know, they can carry 250 kgs of payload. The reason we want to carry the payload is we want to put these optical cameras, different types of cameras on these aircrafts, which can zoom in and click images at like really high resolution. And when you talk about resolution, it's like one square centimeter per pixel. So you can see the color of your jacket, phone, car, what's happening, what are you carrying? All of those stuff. And yeah, the use case from that is immense. You know, a lot of industries need this kind of data. They don't have the access to this data because these aircrafts are built by the military for military. There's no commercially available option for you. And the only commercially available options are either the small quadcopters, which DJI sells or you go to satellite companies. The issue with satellite is there's a high capex involved. You have to put up these number of satellites to do like imagery every 24 hours. And then also physics disables you from doing real, you know, extremely high resolution. So, I mean, when going through like, journey has been smooth, you know, we didn't fall on to this. This was not something that came in my mind and overnight that, oh, let's do this. But after speaking to a lot of people, going through a lot of pivots, and through my journey, you know, building aircraft, this was like something that excited, you know, a lot of people, including me.

Utsav Somani: And out of that 250 kg payload, how much of that is actually batteries?

Shreepoorna Rao (Arctus): By the way, we are not electric, but we are internal combustion engine. So, we use turboprops, we use XP-100 fuel with like a lot of additives on, of course, which we design, we build in-house. We build the entire aircraft other than the engine. So, we use a turboprop engine, which is commercially available on the shelf for like, you know, $20,000, $30,000. Put it on the aircraft, then build the rest of the aircraft ourselves. We understand every single component that goes on the aircraft, every single, from the nuts, to the screws, to the wires, to like the avionics. The material which makes the aircraft itself is also handcrafted by us.

Dhruv Sharma: Let's talk about that for a minute, Sri. So, I mean, if this is a turboprop aircraft, and you manage to keep it up in the air for like 24 hours at a stretch, how have you guys built the airframe? What materials have you used there? How have you made it lighter?

Shreepoorna Rao (Arctus): So, that is where the analog is actually.

Utsav Somani: By the way, Dhruv is our in-house aerospace geek.

Dhruv Sharma: Or, like an enthusiast. And I love hanging out with fellow enthusiasts and builders like yourself. I think, over a period of time.

Shreepoorna Rao (Arctus): It will be done, of course. The way the space is exciting, of course. Everyone should love aerospace, in my opinion. But yeah, see, traditional aircrafts, which fly in these unmanned aircrafts, by the way, they exist in the world. US, Israel, Turkey have the capability to manufacture this at scale. And they have been doing it.

Dhruv Sharma: They are called Predators and Sentinels. And they do really dangerous things.

Shreepoorna Rao (Arctus): Yes, yes, yes. We are commercializing that. But for not killing people, of course. So, these aircraft, the weight of the aircraft is like, you know, 10-15 tons. And they can carry 250 kgs. Now, these were great in 1990s to 2000s era. When titanium was the material to use. Like, you know, the manufacturing process involved making those traditional ribs. Everything was just basic geometry, you know. Circles, rectangles, those made the aircrafts. But now, things have changed. Technology has reached a stage where people are using carbon fiber to make rockets. Now, carbon fiber traditionally is seen with a negative perception in these sectors. Because carbon fiber, when you subject it to radical changes in heat and temperature, they start, you know, charcoalification or like in a burner and like a lot of issues with that. But right now, those are solved problems. They have been done. They have been done by researchers. They have been solved by a lot of people and have been used. There are cars which are made out of that. There are fighter jets that are made out of that. But in this sector, nobody has done it. Nobody has gone into that. Because if you look at the customers, it used to be, it is still defense. And they already have order books filled for the last 30 years. For the next 30 years. Sorry, for the next 30 years, the order books have been filled. So, there is no incentive for them to, you know, iterate and do something new. Their iteration traditionally has been, we put a better engine. So, it will run for two hours longer. We increase the size of the wings. So, it flies longer. Like, there hasn't been fundamental rethinking of the way we build an aircraft than just looking at an aircraft itself. We are changing the whole thing. We are building a whole aircraft with carbon fiber, with the hints of 3D printing, metal 3D printing, and making complex shapes those, you know, in subjected areas. That is our IP.

Utsav Somani: Elon Musk way of looking at rebuilding from the first principles, basically.

Shreepoorna Rao (Arctus): Exactly. Like, check everything out. We need an aircraft. These are the mission parameters. What are the things that, what is the thing that can help us build this at like the least weight possible? Like, check everything out that we don't need on the aircraft. Do we need it? Does that help us fly longer? Does that help us fly at that altitude? If yes, it's on the airplane. If no, get out.

Dhruv Sharma: Like, commercially, Elon wants to reduce the cost of launch. Do you want to reduce the cost of imaging? Is that a fair assumption?

Shreepoorna Rao (Arctus): That is. The fundamental number which we look at internally is operation cost per hour of this aircraft. That is where the whole unlock of the business end comes in. Like, that just makes the use case for us, like, obvious. Because cost per square kilometer is what Earth Observation cares about. The customer doesn't care how you click the image.

Dhruv Sharma: How many square kilometers can you observe from 45,000 feet with the kind of optical instruments you carry?

Shreepoorna Rao (Arctus): Yeah. So, we get a swath of around 20 kilometers. And that is great. If you, even if like satellites wanted to, like, good resolution. When I say good resolution, two square meter per pixel. That is good in satellites. They also have a swath of 15 to 20 kilometers. Because it requires you to align your optical lens and all of those in particular ways to get that image. So, for us, you know, the operation cost per hour is coming somewhere around 70 to 100 dollars per hour. This is including our renting, airspace, all of those costs. Our fuel cost will be somewhere around 20 to 30 dollars. So, when you look at that and flight of 24 hours, the cost per square kilometer comes to like 50 cents. At this resolution. You know, one square centimeter per pixel. This.

Utsav Somani: What were the other decisions that you made when you chose this versus a satellite? We had the founder of Pixel on the show as well.

Shreepoorna Rao (Arctus): Yeah. So, see, satellite is a medium to capture images. You look at different ways. Like these companies build cameras, great cameras. And their job is to now capture images. What is the best way they can capture images right now commercially is satellites. That's the only way for them. Hyperspectral cameras and like, you know, of course, the other co-founder is a great friend of mine. I know him. And ever since, like, I started this journey, they have been helpful, you know, the Bangalore crowd here. They like they use satellites as a way to put cameras up there. They can come to us. They can put their cameras on us and we can do much better for them. Their cost of operations will reduce. And that is the unlock we provide by doing it. The aircraft is the unlock for 45,000 feet. If you want to put commercially anything in space, you think of SpaceX. If you want to put commercially anything in 40-45,000 feet in that region, we are the unlock. That's how we look at it. Now, the Earth observation thing, a lot of things which we can do inhouse, we will do to provide data. In optical and SAR, there are like a lot of commercially available cameras that can do that. So we do that. But in the niche sectors, we can, of course, partner up with folks.

Dhruv Sharma: 45,000 is also a tricky altitude because commercial aircraft, at least like private aircraft, still fly in that altitude.

Shreepoorna Rao (Arctus): So, you know, cross-Atlantic or cross-continental aircraft fly at 40-42,000 feet. 45,000 and above is where a lot of flexibility for us comes in. We want to avoid that crowded airspace anyway, makes it regulation-wise easier. If we have nothing for like 100 square miles of radius, it makes it easier. By the way, the FAA already has regulations for unmanned high-altitude flights. Not at this weight class though. We are the first ones doing it at this weight class. I think it's a problem which we figure out with them because we are the ones who are doing it commercially.

Dhruv Sharma: I don't know if we push this whole AvGeek thing too far, but in my mind, the mental imagery right now is like, she's building like a YouTube dragon lady, you know, in all the way.

Shreepoorna Rao (Arctus): Exactly, that is how it looks.

Utsav Somani: There was a question on our YouTube comments like now, so let's bring that up. So, does Arctis, I mean, most of these words don't make sense to me, so you'll have to help me out here, guys. Does Arctis compete in the hail and mail category? Do the avionics comply with STAG norms and do the ICR missions complete with low-Earth orbit satellites? Do you have the IPR for the turboprop engine?

Shreepoorna Rao (Arctus): Yeah. So, on the IPR side of things, we are partnering up with a few firms. We NDAs with them, so we can't name them. But these are like firms who supply turboprops for commercial use cases for other UAVs. So, can't comment a lot there under NDA, but on the other side of things where, you know, hail and mail category, yes, our aircraft is literally the hail category aircraft. And this is the defense classification of these aircrafts. Although we don't want to, although defense seems obviously like, you know, the biggest market, Earth observation is much bigger. We see a lot of unlock here, and that's what we're going after. We don't want to go after defense at this point of time.

Utsav Somani: And in the defense sector, a very different question, but defense startups, 100% FDI is allowed? Or I think 70% is allowed?

Shreepoorna Rao (Arctus): 70%, yeah, there's FDI, there's some issues, there's some regulation that keeps changing, you know. That's not my expertise, I am an aviation guy.

Utsav Somani: What are the other use cases that you're personally excited about? What other industries can benefit from this?

Shreepoorna Rao (Arctus): So, when I say Earth observation, we look at oil and gas, where we have been getting like the biggest amount of interest and also optically mapping. A couple of, you know, agriculture based companies do need data to make, and insurance, like these three, like has been a lot of inbound interest primarily, which we have been converting. But yeah, it takes some time for us to build the aircraft. That's what we're in the process of right now. That is these three industries globally itself will be a massive use case for us. And plus there is shipping, logistics, and a lot of things that come about, like every single day we figure out new use cases, because this didn't commercially exist. That gives us the biggest unlock, we are figuring out. Right now, the best hedge for me to tell what we want to do is we use the satellite equivalent, but there are a lot more things that can be done. This is an unlock, which we look at it that way, like a market unlock. If you look at the pie of the kind of Earth observation data, we are creating a new pie. And that is what makes us more excited.

Dhruv Sharma: And, you know, whatever sensing equipment you have on board, Sri, like the information that it relays, the data that it relays, can it be used as is, or do you have to like then do some post processing, etc. In other words, like, can you make use of the feed you're getting from the aircraft while it's still airborne?

Shreepoorna Rao (Arctus): It depends on use cases. Now for mapping, yes, we need to do a lot of post processing, depending on the kind of data our customer needs. See, at the end of the day, the customer dictates what we want to do. And real time video is not an unlock, the resolution is. And then the frequency of which you can do is, like, if I can give you...

Dhruv Sharma: Traffic management will need real time.

Shreepoorna Rao (Arctus): Real time, yes. So that is where like optical cameras, optical cameras have reached a level of sophistication where we can just use the real time feed, just run one layer of clearing so that we remove the noise and all of those. But we are looking at it or solving it on an engineering level as well. If the data collected from the camera itself doesn't have noise. So we put a camera in a great way. There are cameras which do that. We just need to attach it to our aircraft, you know, like vibration, jitter and all of those. Those are engineering problems that have been solved. We can solve it. So yeah, like see traffic management, those like, I look at it this way. Imagine an oil refinery has like a six month time lapse of 24 hours of every single second of what's happening in the refinery. Like when the crude is loading, you can literally see the guy connect the pipe, cut it there. Is it turning the wall? How many times? There are SOPs to be followed. And like to give like the amount of scale of impact we can have just with one refinery. Every single process has insurance in it. Because, you know, in refinery things can go wrong. If things go wrong, they go wrong terribly, you know, bad. Each of these processes has SOPs to be followed. If they're followed, then the number of incidents are low. Now, right now, there's some people down there. There are people right there. They have ground infra to do this inspection. Then they have to collect data. They have to do multiple investments in order to do data collection. We become that one point in the sky where we can do everything for them. We target it for them. And this data can help you firstly observe whether your third party contractors are doing right. You can go to insurance company and, you know, say that, look, I have data of the last six months. We had zero incidents or we had 10 incidents. That's below your threshold. So we can reduce your insurance cost for you. And you can see whether your optimization on like, you know, what if you invest 10 million in inventory here, which can increase your output. We can show that for you. A lot of use cases, just one sector coming out. And this, you know, if you talk about insurance for agriculture, we can literally go and fly for 24 hours and say, like, if it is raining, this is where the leaching happens. This is how your chemical components flow through. This is how water takes out the chemical components. Like individual level at square centimeter. So let's give them insights on that. You know, if you cut off here, if you add more pesticide here, this is where the weeding usually weeds start growing in this corner of your massive farm. And that's how, you know, your output is reduced. We can target and tell you that. We can see when it's happening in real time. That is way more exciting. That just gives you the amount of control and data you get. It's like almost controlling every single parameter. You can, you know, control, almost touching, you know, the level of entropy things has. And like different parameters that control it. And then identify, have data to solve. Data is that.

Utsav Somani: All right. As a final closing question, give us a one minute download about how was the experience of being a fellow at South Parkments. We had Prateek on the show. So he gave us how the operating mind works at building this program and the structure behind it. But how was your personal experience?

Shreepoorna Rao (Arctus): Yeah. I mean, it's been great, I would say. Like in South Parkments you meet a lot of people. By the way, I'm 23. And usually if you look at the age group of people in South Parkments, they are like 30 and they have done shit in it. Like great shit. The shit. And like when you hang around with people like that, and you talk to them and you see how they have seen stuff and that experience you have from it. That helps you, gives you, that puts context in your mind of what you can do. I now think globally. I now think what's the best I can do. How can we conquer the world with this? And like, you know, that kind of mindset, that kind of, you know, thinking where people think you're crazy. Like, you know, when I say, you know, imagine we in 10% of aviation, 10 years down the line has to be Arctis Aviation, where our aircrafts are flying and collecting data. I see it happening. I can see that working out. And that's what South Parkments is like, yeah, we see it with you. You know, people like that, you know, Aditya and like Pradeep in Bangalore, of course, have been helping out. You get to meet a lot of people. Their community is great. Yeah. I mean, everyone in this journey has been wonderful.

Utsav Somani: Awesome, Shree. Thank you so much. I've learned so much on this segment from you. So all the very, very best. You seem crazy enough to change the world. So go do it.

Shreepoorna Rao (Arctus): Awesome. Thank you for having me. Cheers.

Utsav Somani: All the best. All right, listeners. We've got a couple of minutes in the middle before our next and final guest joins. So Dhruv and I will quickly run down the news. Dhruv, UPGRAD Unacademy deal is off. The ESOP exercise was rolled back. Did you read about this?

Dhruv Sharma: I did, yes. This was, it was very much in the news last month. People almost thought the deal is going to happen and then it didn't. I think Ronnie made a statement that it was probably valuation differences because we should not have fell apart.

Utsav Somani: I think the deal was pricing Unacademy at 300 million, which is 90% below the 3.5 billion valuation. And I think I read on Money Control that Roman and Gaurav were also spinning out AirLearn, which was their new AI learning app as well. And that's scaling up very well. So they wanted to spin that out. They were raising 17 million from it, from I think Peak 15 and Bloom Ventures, many of their existing investors backing that. So that I think also is on hold. And Gaurav wants to remain CEO. All of this is hearsay, but yeah. Was it hearsay on Money Control? Hearsay on Money Control.

Dhruv Sharma: Hearsay on Money Control.

Utsav Somani: Yeah. India has become the cheapest place to train an AI model. That's a tweet that we saw. I wonder where that is. I mean, some of the facts, I mean, have to be checked out. But 50,000 NVIDIA H100s now live. Startups in India pay 65% less than AWS and Azure rates for it. So what I think the government is trying to do is build distributed access. And I think the bottleneck is shifting from can we afford compute to what can we build with it. What can we put it to. Yeah. So I think the advantage doesn't last forever. So I think it'll be exciting what Aakrit is seeing with his fund as well at Activate AI as well in the world of AI.

Dhruv Sharma: And we've covered this before. I think it only makes sense when everything is in its infancy for, again, the state to step up, master resources, subsidize what they can, ask the best people to come up, share what they can. And there was a point in time when they were creating that GPU cluster. It was almost like they asked all the private players, how many do you have? And that's how they sort of showed up capacity so that people could come and start training their models. Very innovative structures too, by the way, the investments from, someone had done a breakdown of that too. Maybe we'll cover it in our next episode. Yes.

Utsav Somani: And CES 2026. Many cool launches. Anything that caught your eye?

Dhruv Sharma: Yes. I mean, lots of things. And I'm still sort of seeing, that's the one event to go attend, by the way. So I think it's one of the better events that will take place in Vegas.

Utsav Somani: But the Wild World Congress, which one would you pick in Barcelona?

Dhruv Sharma: Which is the other one? I'm sorry, I missed that.

Utsav Somani: The Wild World Congress. That's another big gathering as well in Barcelona.

Dhruv Sharma: I think I might just pick this over that. Personally, at least at this point.

Utsav Somani: One of our guests, Gaurav Khatri, was there as well. He had a booth there. They've launched a ring, apart from all the products that they do, they've launched a ring called Duna Ring. And they are now doing a form factor which is similar to Woop as well, which is like the Luna band, basically.

Dhruv Sharma: That's amazing. In fact, even at CES, longevity tech and wearables as a category were doing incredibly well. I think there was, let me see if I have a note of it somewhere, but this company, this brand called Fender, which launched headphones that have batteries that last 100 hours in advanced noise cancellation mode. And they're also like, you can take the earphones, remove the little piece, the thing that you have on the earpiece, and swap the batteries. But I mean, that wasn't the most exciting thing that was announced at CES. The Jensen Huang, I think, obviously, attracted a lot of attention for all of the announcements that they made. One in particular really stood out, something they launched, an open model for autonomous vehicles that's already live with Mercedes, like the CLA, called Alpamayo, which has been trained on a combination of real world data, which are drivers driving cars, and that's how that data has been picked up. But also synthetic data, because you can't wait for drivers to drive these cars and for all sorts of scenarios to play out on the road. And so they've used one of their own models called Cosmos to create synthetic data that has then trained Alpamayo. So everything is starting to, it's everything all together, all at once, in the world of AI.

Utsav Somani: Elon Musk posed about this. He wants him on as well. He wants him to succeed because I think Tesla cannot be the only one collecting real world data at that level of scale for this to work. And Tesla is one of the rare companies that open sourced all their patents as well. They wanted the world to win. Like, I mean, Elon's, I think just...

Dhruv Sharma: Although BYD is the biggest EV maker at this point in time. And they're doing...

Utsav Somani: They're heavy state subsidized. So I think there might be something to it. I'm not really driven up because of that fact. Talking about protocol wars, there's Google which has stepped into the world of commerce and AI. Universal Commerce Protocol. It's basically an agentic way of doing shopping as well. And we saw something which OpenAI and Stripe also did. You have any insights in this?

Dhruv Sharma: I did see the Google announcement and it appears to me as though they want agents to do everything that you would do. As in browse the internet, shortlist things, compare A, B, C, and then finally take the different steps that you have to in order to close that transaction. And the other thing Google's also doing now is it's bringing together not just the purchase experience from a customer standpoint, but also the delivery of customer service, customer support. And that's pretty much end to end. So that's going to be exciting to watch.

Utsav Somani: And protocol wars have existed. I mean throughout the internet history. I think even in crypto, which are known as protocols, but I mean just otherwise also. So I think it'll be interesting to see the Stripe camp and the Google camp battle it out. Great for customers. And it's 100% open source. One final item before we welcome our next guest. FIFA World Cup is happening this year in US and TikTok has become their preferred platform ahead of this World Cup. They live stream part of the matches as well and they've been allowed to that League MLS, I think is doing well, which Beckham is driving a lot of viewership to. But I think overall, I think soccer is here to stay. Like I think it's the biggest watched FIFA is not native to the US.

Dhruv Sharma: TikTok is not native to the US. I think in a globalized world, it's incredible to see these things come together.

Utsav Somani: All right, let's welcome our final guest for the show. Krishnakumar Sir, welcome to the show. Good. Happy to be here. Thank you so much for giving us the time. Typically, what we do with our guests is we ask them to introduce themselves. We don't need that with you. So should we start with Meena Menchers or should we start with Mindtree? Like what would you prefer?

Krishnakumar Natarajan (Mela Ventures, Ex-LT Mindtree): I think if you think the listeners will get value to hear the Mindtree story, certainly you can go with that.

Utsav Somani: I mean, the most basic question, Mindtree, 2.5 billion, L&T acquisition, so much behind the headlines. Like, what was that journey for you? You started off your career at Wipro.

Krishnakumar Natarajan (Mela Ventures, Ex-LT Mindtree): Yes, I started my career in Wipro right out of business school and spent about 18 and a half years. And to be honest, one was fortunate enough to do that because you really had the advantage of when Wipro was small, closely watching YouTube, leaders like Mr. Premji, how do they look at business? How do they build the culture of an organization which is even beyond them? So a lot of that reflected in the process of building Mindtree itself. I worked with Wipro for about 18 and a half years before co-founding Mindtree in 1999, August. So in some way, I think the timing was right. It's really a stage where the Indian IT industry had got global acceptance. And a lot of us had got exposed to building global businesses. The whole idea of risk capital was also very early. There were really no VCs in India then.

Speaker 6: So, Mindtree was one of the first companies to reach 10 billion users around when we started.

Krishnakumar Natarajan (Mela Ventures, Ex-LT Mindtree): And subsequently raised another 13 or 14 million from another US-based. We weren't fully aware how to leverage the internet. Just like today, enterprises are struggling to use AI. 20 years back, when enterprises didn't know how to leverage the internet. But the fact was, while enterprises survived, a lot of our customers are dot-coms. And like you rightly said, the dot-com bus sort of took away almost 50-60% of our customers. And not only that, a lot of them had receivables, which again created another thing. For a company which just raised 10 million of capital, to write off 3-4 million wasn't an easy task. But we survived. We survived that Asia-Pacific crisis which again brought in some softness into the business.

Speaker 6: But we survived all those what I would call marketing disbelief.

Krishnakumar Natarajan (Mela Ventures, Ex-LT Mindtree): Coming from there in 30 years to establish a globally competitive where globally you are considered as being the leader in the industry is extremely credible, which is one part of it. The second thing which I'd really compliment the IT industry for is I think they set new benchmarks of how to run public listed companies with a high focus on governance. Be it Wipro or Infosys or TCS. I think clearly set the benchmark for governance which gave investors a lot of confidence. Yes, I think we can put our money and it will be safe in these. So it really built a public market investing confidence which in turn was a flywheel for the industry to grow because subsequently all the companies that I mentioned about went about listing in the US be it New York Stock Exchange or NASDAQ. And you've gotten a lot of global capital. At a peak in my entry, I recollect our foreign institution investor holding used to be about 48%, which means almost half the market cap was really funded by foreign institution investors. And the last one was really the focus in terms of talent and talent building. Because while the criticism now is that, hey, India lags behind US and China on new areas like AI. Yeah, I do believe that the talent pool which we have, which is really a reflection of the work over the last 30 years the IT services industry has put in is going to be a substantive advantage in the next phase as technology evolves. What I would certainly think the industry could have done better is I think they've always seen it as a business which every quarter churns out cash flows and money. So there was really no interest to take risky bets, which is why we always sort of lag behind being a product country. I think if even 10-20% of the returns made had been invested in a little more riskier type of ventures, we could have created another piece of business which I think will happen in the next decade, which is honestly one of the pieces of Mela Ventures also.

Utsav Somani: And I think that's needed, right? The last decade, I think, belonged to the IT industry which rode the wave of digital transformation, but I think with the world of AI that's coming up, which is a huge platform shift, many of the world leaders have said that it's one of the biggest platform and technological shifts of our generation. What do you think the Indian IT industry needs to do different in the next decade? And how can we adapt and benefit from this shift that's happening?

Krishnakumar Natarajan (Mela Ventures, Ex-LT Mindtree): No, you're absolutely right. I think the Indian IT industry was the leader in the digital transformation, again, not to boast that Mindtree was one of the first to see that opportunity in 2011 and set up a separate business for digital. And the markets really rewarded us. I think some of our, what I would call the premium in terms of market was really because of the outstanding work we did in digital transformation. Again, one of, what I call something which has survived is really the software which we built for Aadhaar. The whole element of trying to make that identity system a reality was really done by Mindtree in terms of things. What I do think is the IT services companies, while they don't take bold bets on the future, I think what comes as adjacency I think they managed to sort of really deploy that. Beyond the digital transformation, there's a phase when a lot of tools are becoming a reality, particularly in enterprise technology. The Indian IT services world obviously adopted it the best because that was delivering both cost efficiency as well as speed of delivery to customers. And they evolved an excellent model by which they deployed those tools and shared a part of the benefits with customers which in a way is reflective in terms of the continued growth which they did. But the next decade is going to be very different because with AI, with Gen AI, potentially an AGI coming in, I think the nature of business is going to be a lot more in terms of experimentative where you need to have far more deeper domain knowledge to try and deploy these technologies really have a measurable business value in a much shorter time. Which means experimentative type of work, projects which might be probably 12 weeks to 16 weeks time. Whereas, the Indian IT industry has got used to five-year contracts where your revenues are, I would almost say, committed for the next few years and so on. So the business model needs to change the way in which you upskill people. I think that needs to dramatically change.

Utsav Somani: Interesting. And now, I think that's a good segway to start talking about Mela Ventures. That's your bet for the future of Indian technology. So, can you tell our listeners about Mela Ventures?

Krishnakumar Natarajan (Mela Ventures, Ex-LT Mindtree): See, Mela Ventures, we started about five years back. When I exited Myntree, the opportunity was saying, hey, should we build another brownfield IT services company? And, to be honest, the capital markets had evolved so much that we had a few large global investors, PE funds, telling us, saying, hey, we will fund you and build another Myntree billion-dollar company. Honestly, after looking at it for a few weeks, I realized, what am I trying to do? I'm just trying to sort of be in the same game and trying to demonstrate that I can build another Myntree. I would rather be excited trying to create other successful entrepreneurs, which is how the whole idea of Mela Ventures came in. Because what we saw, because I've been involved in the startup world since 2010, I used to be the chairperson of the industry association, NASCOP. So, in 2010, I initiated a large program for startup called the Tech 10,000 Startups, which is probably the first, I would say, structured effort to build a large startup ecosystem within India. And that's sort of really gone well. And in that context, I started building a lot of network within the startup community. I personally did a lot of angel investment. It sort of gave me the thing that one should really deploy whatever you've learned to help other entrepreneurs become more successful. And then we looked at some data. See, if you really look at in the US or Israel, and you take the percentage of companies which raise seed funding and go on to a series A, B type of investment, probably about between 4% and 5% reach that stage. Whereas in India, that was less than 1%. And clearly, the logic or what we saw as a reasoning was that it is not because the ideas are not good. It's not because the entrepreneurs are not capable. It's purely because they don't get the right amount of risk capital at the right time. And they don't get adequate mentorship and support to really grow their companies. Which is how we evolved Mela Ventures more as an operator-led venture capital company where we say we bring in the three E's. One is an empathy with the entrepreneur. The experience which we have in terms of building up companies. And clearly, the key areas where entrepreneurs have failed in the past because we do think number of technology business in India can be made global. And people fail in terms of accessing the U.S. market because they don't adopt the right go-to-market strategy. They don't build the right partnerships. Which are all areas where we have in a way done many mistakes and corrected them and built businesses. So how do we leverage our experience of what works what does not work to help these companies become more global in nature and really become the next generation tech companies which have built global businesses in India.

Dhruv Sharma: You know what's very interesting sir is that you mentioned the 10,000 startups at NASSCOM. Yes. And there's something about that number about 10,000 so then SIDBEE is 10,000 crore fund of funds. And now there's a 10X larger pool of capital.

Speaker 6: Yes. Exactly.

Dhruv Sharma: RDI. So we'd love like from again from you know your experience from your perspective we'd love to understand how this journey has played out because it seems like finally there's a you know there's an acknowledgement a thumping acknowledgement that public funds can actually be used to fuel private innovation in India.

Krishnakumar Natarajan (Mela Ventures, Ex-LT Mindtree): Absolutely.

Dhruv Sharma: Yeah it must be very gratifying.

Krishnakumar Natarajan (Mela Ventures, Ex-LT Mindtree): No it is very gratifying and you rightly pointed out when we launched the Tech 10,000 startups in 2010 at NASSCOM a lot of people said hey how can there be 10,000 startups in India? It's something which is very imaginary. Yeah. But the fact is today there are 195,000 startups in India. So it's a reality which happened. So as long as you can provide the right enabling ecosystem this can be made to happen. And to be honest I think the government is playing a very pivotal role in doing this. Like you mentioned about the SIDBI fund of funds the first fund of 10,000 crores has been completely deployed with the right business model where SIDBI says you know what we don't understand the startup world. We are not the experts. We would rather pick up fund managers who have that experience and really enable them to go and pick up the right startups and do. And I was just looking at some data. Already they have created value in terms of investments they have done because clearly the funds in which they have invested have been able to pick out winners which now has led to the government confidence saying hey the model of not interfering but really being an enabler the government has now launched the RDI fund which they have even gone one step ahead beyond the fund of funds they're saying if private companies want to do a research which they feel is going to have good outcomes in the future they are willing to give them probably a 15 20 year loan at very small interest rate which means the private R&D energy which also will start flowing. So to that extent one should also appreciate in the last 10 years really the government has done the right job of being an enabler and in a way trying to accelerate these ecosystems without trying to regulate them without trying to play an operating role in them because they don't have the expertise to do that.

Utsav Somani: And during the early part of our chat you mentioned something about Wipro's culture that stayed with me basically you mentioned and highlighted how early days culture was so important and culture became a big word when startups were becoming household names as well. So when you invest from Mela Ventures like are there specific things that you try to pick on soft learnings or sort of impart to your portfolio founders as well like what are the key sort of insights that you give from your early days at Wipro?

Krishnakumar Natarajan (Mela Ventures, Ex-LT Mindtree): There are both soft and hard elements and when I said we are more an operator led venture capitalists we do take an active role in the company not in terms of running the company but in terms of setting up basic foundational elements like a very clear basis of an accounting system a sort of tool to manage that and to extract out of it a basic MIS which the management reviews and we also review with them on a periodical basis every month or every quarter we do that. Similarly we help them set people process because every company talks about ESOP but how do you really set up a ESOP system which will be motivating for people to feel ownership of the company and really do go that extra mile for the customer. What are the people process to be done? How do you when you recruit your next level not necessarily go to your dorm mate who sort of was with you in the hostel but rather look for the right person with the right skills. These are the sort of areas where we contribute and we do consciously tell our founders saying that culture is not something which you can just bring in as an add-on once you've grown to a certain scale. It needs to be built up right from the time you have set up an organization and when you start getting in the next level of people and it can take different forms to in a way how it gets implemented within an organization. If a founder feels comfortable he sort of with the initial team has a set of values tries to identify what is the purpose for which the organization is and get people aligned to it and then really starts looking at initiatives by which these values can get embedded in each and every person within the company. And some of these soft differentiators clearly make a difference between people who love to work for the organization and give that extra mile rather than just treating it as another job.

Utsav Somani: And I mean about leadership styles as well I saw that you became CEO of Mindtree slightly later in the journey as well and Dhruv is now the CEO of AngelList India we both were colleagues there as well. So how do you identify when is the time to from your journey at Mindtree a leadership change is needed or a style change is needed in leadership?

Krishnakumar Natarajan (Mela Ventures, Ex-LT Mindtree): See I honestly think again this is my personal view that leadership beyond a certain period in fact that's why I love the American presidential system where after two terms you can't be a president term. Because the fact is that you're used to a certain sense of how you see problems how you solve them many times a new view brings in a different context a different way to solve the problem. So essentially the role of a leader is also to create the next level where you're constantly seeing you're making them ready to take on bigger roles and bigger responsibilities. So I don't believe in the system where there's a CEO for 20 years and 30 years to be honest many times because they're so used to the role. I think they probably miss some areas which can be thought of differently. So I think it's a good idea for organizations to look at maybe an 8, 9, 10 year time period beyond which people should move on and provide an opportunity for somebody else with a different mindset to come and change a Dhruv, you have something?

Dhruv Sharma: Oh no, I was just soaking in what we just heard.

Utsav Somani: I mean, fascinating. I have so many mind tree questions that I can go on for hours and hours because I think that journey is fascinating in itself. Absolutely. A little bit of drama as well. I think you hit a low point in your personal journey. I think I can from whatever I can read online with the L&T takeover. How are you handling that conflict internally in your mind, personally, what was going through your mind at that time? I think it'll be relevant for the founders or people involved in the startup ecosystem because I think everyone's dealing with so many different identities and challenges and conflicts in their personal lives, professional lives. So I think it all comes together and would love to hear your experience during that time.

Krishnakumar Natarajan (Mela Ventures, Ex-LT Mindtree): I think it's again a great question. And to be honest, as founders, many times we get into this what I call identity clash in terms of you as a founder thinking that the entity which you create is only yours. So I think that's the wrong premise to start with. Because to be very honest, in a capitalistic world, at the end of the day, I think anybody who has the ability to sort of gain majority ownership will tend to own the asset, which is something which you have to start, I would say, making yourself comfortable with. So again, to be very fair, when this hostile takeover was initiated, I guess there was a lot of what I call negativity in the mind. And I think as any founding team would do, we did our best to sort of try and see if we can still continue to sort of do what is right for what we thought was all our shareholders. But we could have gone to a situation where we had some sort of what I call almost like a tie break between L&T and us because we had a set of foreign institution investors who were completely supportive, saying that, no, I think if you guys want to own it, we'll support you. So we would have ended up with something like a tug of war. But then what I realized was, what am I going to achieve with it? Next five years, the company will go to dogs because two major shareholders are constantly fighting, which is when I still remember the 28th of May was the day when the offer was closing. I personally told two of my large shareholders, one who held 11 percent and the other one who held 12.5 percent to say, hey, please go ahead and put your shares in for the thing, not because of anything else. One just said that, hey, there's no point trying to run your next five years on a negative contact. At least let the organization survive. But I'll tell you the positive thing of that. I didn't sell a single share. The three co-founders along with me, we owned about 13.5 percent. We didn't sell a single share to L&T. And again, touch wood, God was kind. After that, the markets went up 3x, 4x. So all of us got a much better outcome out of that, which in turn enabled us to do a little much more on philanthropy. Like my foundation has created a geriatric center of excellence within St. John's Hospital. My colleague Subrata and Partha gave 400 crores to ISE to build a multi-specialty hospital. So we have been able to, because that was clearly the intent. Whatever money we make extra, we'll sort of do back to philanthropic areas, which we love to do. And to be honest, if that had not happened, I would have still continued as Executive Chairman of Mindtree. I wouldn't have had the opportunity to do something like Mela Ventures, which honestly is exciting and making me high energy every week because I meet so many young entrepreneurs with great ideas. So at the end of the day, if you look at it, it's all turned out very good for all of us.

Utsav Somani: So Mela Ventures, I mean, as a final closing question, I know we're over time. So thank you so much for giving us the time. As you think about legacy and what part of legacy at Mindtree, and I'm guessing Mela Ventures is a derivative of the work that you've done at Mindtree and the success that you've enjoyed because of that, is part of that. But what part of the legacy that you've left behind at Mindtree makes you the most proud?

Krishnakumar Natarajan (Mela Ventures, Ex-LT Mindtree): No, I think the legacy which makes me proud is, to be honest, I'll share two instances. One, after all this L&T thing happened in one of the Bombay events, I met one of the very senior person of L&T who was a part of it. He said, you know what, we have done many acquisitions. We have never seen a company which is as clean as what you guys are. So you guys built something very valuable, which is just an individual sort of view. But even today, when I meet a lot of people who have just heard of Mindtree, the first positive comment is saying, hey, you know what, I think Mindtree was a great place. That's what I keep hearing. I think the focus in terms of saying that you can build a high performance organization, at the same time, be a caring organization. It looks like an oxymoron. You can't build a high performance without just driving people mad. But you can also build a high performing, high caring, which is really the genesis and the core base of the Mindtree culture, high performing, high caring. When people express saying, hey, you know what, we saw it in action. We still think that was something which you guys built. It makes, personally, me very happy.

Utsav Somani: Amazing. Sir, thank you so much for giving us the time today. We'd love to have you back as you progress in your Mela Ventures journey as well. Thank you so much.

Krishnakumar Natarajan (Mela Ventures, Ex-LT Mindtree): Absolutely. Pleasure being here and all the very best for all your great efforts in terms of trying to create that excitement.

Utsav Somani: And as portfolio founders from Mela Ventures as well, we'd love to share the stories on T1 as well. Absolutely. Yeah. Thank you. Thank you. Thank you very much. Awesome listeners. That was it from us. Monday, slightly longer episode, but we've covered a whole bunch of different topics. You've heard from news. We've heard Anmol Meni talk about the world of consensus capital. We've heard Krishnamoorth Sir give us a tweet.

Dhruv Sharma: I was just thinking we had a 23 year old CEO and then we had somebody who ran a public company all on the same show.

Utsav Somani: I mean, I will need to process this transcript well. I believe there were some glitches in the stream as well, which we will fix. So just look out for the full interview by today evening. Thank you so much. We'll see you on Wednesday. Bye bye.