Full Transcript

Dhruv Sharma: Today is 14th of January. How are those resolutions holding up? It's also Sankranti and Pongal and many many other harvest festivals. So you know we wish you the very best. It's so great to see you.

Utsav Somani: Good to see you as well. What's happening at your end?

Dhruv Sharma: All good, it's of I I mean the world got know that the creator of bill Burtt Scott Adams who you also introduced to us at Angelist India. Remember in a in a handbook you. Yeah, it was one of his articles.

Utsav Somani: Anyone who's trying? I mean everything is changed in the world of AI, but that single piece of article that he's written. He was the creator of Dilbert, of course, so he's done a lot of writing and I mean brilliant writer and brilliant author as well. He's done some pretty good pieces. So now I was a big fan of him and I think a friend as well and they've done a podcast together as well. And novel in the early handbook of Angelist. They did this piece. They highlighted this piece called the day you become a better writer. It's literally I think maybe 5 paras and that honestly single article on the Internet can make you a better business writer. You don't need to take any business courses. It basically says cut away any fluff that's needed, like how you can say that I'm very interested. If you say I'm interested, it's pretty much the same thing. Adding very is not right.

Dhruv Sharma: Conversationally short sentences, clarity of thought, that sort of thing.

Utsav Somani: Yes, and with the world of yeah, no, I'm dashes as well. Yeah, what are the cool things that we saw on the Internet? So Anthropic released this new tool called Co work. It's basically a copilot for an average person like who doesn't want to code, but just wants to use AI in their daily life, right? You get access. It's locally run, so you basically can sort of give it access to any folder and say that summarize this, create a presentation out of this. And it's sort of like your copilot on your computer and many of these platforms will now try to embed themselves in your daily life and daily workflows to get distribution and live on your computer and embed themselves and make you a daily active user. Have you checked it out?

Dhruv Sharma: I have, by the way, in the launch announcement, they say very much later than the launch announcement. They say if you don't instruct, you know this carefully enough, it might even delete all of your folders, so just be very careful. But that's that's in the fine print. Andrew Reid, in fact, of Sequoia put out a tweet calling it Claude Code, not for code, but yeah, I mean, I've used both of the tools pretty good and Claude suddenly seems to be dominating the conversation.

Utsav Somani: Dude, it's every day something new, like even Google released a new VO model, which is their video generation model. So now they have 4K upscaling as well. So it's insane. Like every day there's something new. How do you keep up with all of these announcements and launches apart from like, of course, listening and updating us on TUN? TUN is where it's at.

Dhruv Sharma: Yes, this is this is where we cover what's worthy of your attention versus what's not. But look, I mean the only way to know what's I mean the only way to know is to try and and maybe also have a few people who you can call or check in with about what to try because there's so much. So I tend to do it that way, right? Like I mean, it's a handful of accounts and people who I will check with before trying out anything new.

Utsav Somani: Awesome, closer to home, Tamil Nadu is building India's first sovereign AI park. They've also launched a 100 crore deep tech startup policy. This is being done with Sarvam AI, which is one of the leading local AI players as well from India. Do you have any insights into this?

Dhruv Sharma: Yes, I mean, I read the launch announcement. By the way, Sarvam is going to release that model next month. There's that big AI conference taking place. It's an AI impact summit 2026.

Utsav Somani: Everyone's coming down for it. I think Akrit was mentioning, right? Jensen and a couple of other big names.

Dhruv Sharma: I'm sure he did. But you know, a very interesting thing about this is, you know, pretty much competitive federalism at its best. I really hope state governments start competing with each other to, you know, in a very good way to attract the best and the brightest. I really don't know what it is that can be done within the bounds of Tamil Nadu. I mean, maybe they'll put up a data center. Maybe they'll subsidize. It'll create a few jobs. But I mean, this sort of support is quite welcome. You know, apparently in China when DeepSeek made that announcement, it's obviously, I think it's located in the same places. I could be getting this wrong, but the same place is Alibaba. And so a lot of provincial and city governments that compete with each other in terms of trying to attract the best technology talent, they actually did a lot of introspection officially on why didn't that moment, you know, that announcement, that launch come out of a company from our city. So it'd be nice to have something similar.

Utsav Somani: Talking about Indian government, they are now spending time on things like removing the 10minute word from the advertisement of these quick delivery apps as well. 31st is when I think all of this came to light, where I think there was a couple of controversy and Dipinder Goel got involved where everyone was saying that gig workers are at risk always and these 10-minute deliveries put them at, I mean, risk of like crashing into this thing or cutting corners or skipping red lights, thereby affecting the people on the city. The roads as well. And we employ over 12 million gig workers and 200,000 gig workers took to the street or protested against this gig economy culture in India. And they were, of course, staging a ban. And then everyone got involved. Raghav stood up from the government side who took a stand that none of this should exist. We're compromising many other things just to get, I mean, quick deliveries. And then Dipinder Goel get the other side where he says that the stores are so closed that you don't actually need to deliver fast. Like you're not breaking speed limits. They're just abiding by this thing. And it's always like a switch on, switch off thing. So if you're not a full-time employee at a company, why should you get the benefits of full-time employment as well? But government stepped in and they've released a policy which we discussed earlier where companies will have to contribute to a common fund to do welfare for these gig workers as well.

Dhruv Sharma: They've certainly formalized the gig worker, the gig economy as well. But also on this, I think it's the ministry of labor just quote-unquote nudging Blinkid and their peers to remove any references to the 10-minute promise. Yeah. You know, making jokes about what's gonna happen next. Maybe Maggie gets a similar nudge as well to remove the two-minute promise.

Utsav Somani: All right. That's us from today, at least from the news segment. Let's welcome our first guest, Gautam. Good to have you on the show. Hey, Gautam. Hey, Dhruv and Utsav. Good to see you guys. Thank you so much for giving us the time today. Give us, I mean, tell me what's behind the name, Inuka Capital.

Gautam Shewakramani (Inuka Capital): Oh, that's a loaded question. So it's actually a very, there's a personal story and then there's of course an analogy that ties into sort of values and startups. Very quickly, personal story is my, when I was shutting down my first startup, this was 2017, my two-year-old daughter, almost turning on three, she asked me, you know, what I did. She couldn't understand like, you know, our work is very abstract. You're sitting at a computer all day. Unlike my wife, who's, you know, she works, she's a therapist, she works with children. So it's very easy to explain that. So at that point, I just shut down my startup. So I told her, I said, well, you know, right now I don't do anything, but I'm going to start a new company. And then I just casually asked her, this is at bedtime. I said, what should I name it? And she said, you should name your new company Inuka because that was the name of a stuffed toy that she had. And the stuffed toy we've actually purchased in Singapore at the zoo, Inuka was actually the name of the polar bear in Singapore zoo. This is a bear that was born and lived and eventually died, lived a good life in a climate controlled enclosure. So it's the only polar bear that grew up in the tropics. So there was a metaphor there from a startup ecosystem perspective is that founders tend to, they have to get comfortable in the discomfort of being in a polar bear in a tropical environment. So that's the startup analogy, but obviously on a personal level, I promised my daughter I would name my next company Inuka and I did end up starting up, did not end up starting up immediately after shutting down my last startup. But I think of Inuka Capital as a fund and a startup as well. So when we did start, I had to live up to the promise. So that's where the name came from.

Utsav Somani: And give us some numbers. How many investments have you done? What's the fund size? What's the portfolio split and all of those things.

Gautam Shewakramani (Inuka Capital): Yeah, so we are a $10 million fund. We're fully subscribed. We started raising just at the turn of 23. So probably at the worst time possible. And of course you understand this business as well. I remember one of our first capital calls was 24 hours before Silicon Valley Bank went down. So, and guess what our bank details said. It said Silicon Valley Bank. So yeah, so we started raising at the worst time. It took us a good, almost two years to raise. We are fairly concentrated as a pre-seed fund. So we write 300, 400K checks. We like to lead. So we're usually the first check. And usually the largest check in what typically is about a million-ish dollar round. And so we're fairly concentrated. We're targeting a portfolio size of only about 27 companies, 26, 27 companies, which at pre-seed is fairly concentrated. We're currently at 21 companies. We just made an offer for our 22nd investment a few hours ago, actually. We don't know that it'll consummate, though the founder still has to say yes. And in terms of splits, so we're an India-focused fund. So we invest primarily in businesses that are being built in India. I would say we have, I would kind of split the portfolio up into three buckets. I would say broadly. One bucket is what I would call sort of the applied AI or deep tech sort of bucket. In all cases, these are companies that are solving real business problems. I heard you right at the beginning of this call talking about how to be a better business writer, right? And one of the things that, one of the internal rules that I have is when we write our investment memos, even if it's for an AI company, don't use the word AI, right? Like, can you explain what this business does and whether they solve it with sticks and stones or whether they solve it with AI is actually secondary. It's a means to an end. So that's sort of a big chunk of our portfolio. Another big chunk of our portfolio is what I call enabling Indian businesses. So we believe that Indian businesses are now ready for purposebuilt software, especially outside of enterprise. So mid-sized businesses, small businesses, I think we think are ready for software and for full stack solutions. We do think the business model needs to be a little different than what you traditionally see in the US. A per-seat SaaS subscription model is not gonna work. So whatever our portfolio companies do, they make sure they embed themselves very deeply into the cost of the revenue cycles of these companies and somehow are able to take a percentage of that value that's flowing through. And a third sort of broad big bucket is what I call increasing consumer aspirations, right? So we don't do consumer. So we don't do consumer non-tech. So we won't do a DTC brand that makes protein or granola or shampoo. Although these, I think they can be great businesses and we've seen so many great outcomes. We are playing consumer aspirations through technology and through sort of the second order effects of the growing consumer spend. So we've got a bunch of companies in that bucket. So those are the three broad buckets that we sort of play in.

Dhruv Sharma: That actually gives us a lot to pick on, Gautam. But before that, maybe you'll humor me with a story, right? Or rather, I will try and humor you with a story. So when you spoke about how Inuka's name came into being, have you heard Jamie Dimon's story about when he went out of work for like a short amount of time in his life? Apparently he had three daughters and he said, this bank hasn't quite worked out. So the first one said, Daddy, are we gonna still have a home? And he said, yes, sweetie, we're gonna still have a home. And the second one, he says, was obsessed with going to college for some reason. She says, can I still go to college? And he said, yes, you can still go to college. And the third one said, Daddy, well, can I have that phone? Because you're not gonna need it anymore. So kids, we're humbling you even through those career transitions. Exactly, exactly. But you said something very interesting about say business writing, right? So the investment memos, and I know you take pride in writing great investment memos. So nothing related to investing.

Utsav Somani: It's evident from Gautam's Twitter that he must be a good writer. He's done a bunch of tweet stones. Yes.

Dhruv Sharma: So Gautam, what's the difference from, like what's your perspective on the difference between people who use a whole lot of jargon as opposed to just very specific vocabulary just so that you can be more efficient and use fewer words?

Gautam Shewakramani (Inuka Capital): So I think, so my take on jargon is that it's inefficient. Like I think that's the big sort of takeaway for me. And I keep going back to a lot of my, you know, of course I have kids, right? My wife, I live in Mumbai, which I'm proud of. Like, I'm proud of the fact that I'm not in Bangalore which is the heart of the sort of tech ecosystem here because I think it gives us a lot of objectivity. And, you know, my weekends don't, I have amazing friends in tech in Bangalore but when I hang out with my friends in Mumbai on weekends, like we're so deep talking about Claude and ChatGPT, 99% of people aren't using these, right? Or they're barely using them, right? Like just scratching the surface. So I keep kind of going back to, you know, if I had to explain a portfolio company of ours, right? Which falls under the deep tech and applied AI bucket, right? Or even any of our companies, right? Like, can I explain it to somebody who I run into at the gym in Mumbai?

Utsav Somani: Let's give Gautam a task. We had one of your portfolio founders on the show, on arrival, Ankit. How will you explain that company to your daughter? Amazing, right?

Gautam Shewakramani (Inuka Capital): So actually it's interesting and maybe I'm throwing Ankit out of the bus here but one of the things that we used to joke about early on in, you know, this is just after we invested and before he had sort of like really kind of gone out to raise his next round of financing, my co-founder used to joke and say, you should communicate what Honor I will does because you do a better job communicating than Ankit does. And so we spent a lot of time like talking about that and explaining that, right? So, you know, for somebody who understands, so even, so he uses the analogy AWS for travel, right? Anybody can spin up, anybody can spin up an AWS server and deploy an application. Now any company can spin up on arrival and start selling travel. The problem with that is, you know, this doesn't fit my rule of it should make sense to someone I run into at the gym because believe me, nobody knows what AWS is, right? We on this call know what AWS is and everybody listening to this, you know, podcast probably knows what AWS is. So I like to kind of simplify it and even further and say it allows anybody to sell travel very easily that's one way I describe it. The other way I describe it is, you know, and you talked about using specific words that make it easy. I call it a headless OTA. So imagine the core technology of make my trip or clear trip without the interface, right? And you can plug that into any interface that you want, right? Whether it's, you know, whether it's a bank or whether it's, you know, a real estate firm that wants to start selling travel. I just made that up, obviously that I can't imagine that would happen, yeah. So that's how I would describe it.

Dhruv Sharma: And Gautam, I mean, Inuka you said you started raising the first one in 23, right? Yeah. But you've been investing a lot longer than that. Yes, I have.

Gautam Shewakramani (Inuka Capital): So what's the progression like? Yeah, so I can tell you a little bit about, so I've always said that I've basically had two parallel career paths that both started in 2011. The first part was basically founder, operator, company builder, right? And that's the more vanilla part, right? I started a company, it didn't work out, shut it down. Then I was at Quora, Skillshare, OnDeck, and now the fund, right? So like that's sort of the company builder part. The second part I think is a lot more interesting and it was probably in hindsight, not the smartest thing to do, but it worked out. But when I came back from the US, I had some savings and instead of investing in mutual funds, which are sahi, right? I basically ended up getting plugged into the fairly young tech investing, angel investing scene, which at that time was, there was a lot of stuff happening in Mumbai, right? 2010, 2011, I'm sure Mumbai angels, rings a bell for those who've been around long enough. And what I found is from 2011 onwards, I was writing super small checks. I was one 10th the size of the next largest check in these angel rounds that were coming together. But I was the guy subsidizing conviction, right? Every other angel was a more traditional business, I should say businessman, but business person, right? There were women as well, you know, that were maybe running a factories, making tires or making ball bearings, right? For manufacturing, and obviously had a lot of excess cashflow. So, but for me, it was like, I was just trying to help people out. I was imagining the future, while I think a lot of traditional business owners or private equity mindset is like, what have you done in the past? And through that journey got an opportunity to invest in some pretty awesome companies. Exotel was one of my first investments, which did pretty well. Tonbo imaging has done pretty well. In fact, I just saw an article about them. I think it was today or yesterday in one of these publications. So yeah, and obviously I made a lot of mistakes as one does. So in fact, I would tell some of my early LPs that my angel investing was the most expensive education that I got, right? In the venture world. So yeah, but that was for a while.

Utsav Somani: Quora days as well. Apart from the investing, you spent time at a very large Silicon Valley company, right? Which you brought to India. And Quora, I think was already big in India, but I think you made it much bigger. So what were the numbers like? How is Quora doing now?

Gautam Shewakramani (Inuka Capital): Yeah, so it was interesting. Quora was fairly large in India at the time that I joined. What was interesting is they were looking for someone in India for a few years, actually. And the story of even how I got that role is actually kind of interesting. I ended up- Power user on that platform? No, actually I was so confused about my own career that I started using Quora to read about what other founders were doing with their careers after moving on from their companies. And then the more I used the product, the more I got impressed with how thoughtfully it was built. And I literally went to the jobs page and I saw that they had an open job for like an India role that had been open for like a couple of years. So I basically applied cold, right? But the thing is the way the job was written, it literally read like it was written for me. So I applied cold. Now, obviously the happy ending of the story is that, oh, I got the job, but that's actually not how it happened. Nobody replies to cold inbound, right? But I ended up finding a future colleague of mine at Quora who was at MIT where I went to university as well and sort of got a warm intro in. When I joined them, and that's sort of how the whole thing happened, and the first few calls that I did, it was on, there was no Zoom at the time. I think it was, I used Google Meet or Skype or yeah, one of those, right? It was basically like, look, if you guys want to grow in India, here's what I would do and here's what I think you guys are doing wrong. Like that was basically my, I'm quite blunt when it comes to these things. So, and the culture at Quora is very direct. So I think that was like very appreciated and that's sort of how it all happened. When I joined Quora, I can't share too many numbers, but there are some numbers that I can share. That are public. When I joined Quora, we were doing about 30 million monthly unique users in English. And English was the only language available at the time in India. When I left, we were doing, which was about two and a half years later, we were doing about 70 or 80 million monthly uniques. We had launched nine languages in India, Hindi, Tamil, all the big languages. I don't want to give you a laundry list of languages.

Utsav Somani: And your mandate was to grow the platform and were there any specific task items that you were given?

Gautam Shewakramani (Inuka Capital): Interestingly, I was literally the only person outside the United States. So my joke at Quora was that if you need to, if they needed to buy a printer that was physically located outside the US, like I was the guy to do that, right? Like it was.

Utsav Somani: It must also be very alienating, right? Being the only person outside India for a big company, which is this thing. Alienating in the sense that, I mean, culture and- Yeah, yeah, yeah.

Gautam Shewakramani (Inuka Capital): Well, here's what was interesting. I made it, I actually wrote a couple of answers about that on Quora about sort of remote work. I was doing remote work before it was a thing, right?

Utsav Somani: Fully metal, basically, like fully metal. Yeah.

Gautam Shewakramani (Inuka Capital): So, I mean, I was in the US every nine weeks. I did that for three years. I got the highest tier of status with Singapore Airlines and didn't pay for many vacations after that because of all the airline points that I earned. So that, yeah, it was challenging. We did end up hiring and building a team out, not just in India, but across the world. So I was not the only one. I was the only one only for probably a year. And then we sort of started hiring more people. I had a counterpart in Japan very soon after. And in terms of my mandate, my mandate was primarily on product and growing product usage within constraints of quality and all of the other things that a consumer internet company would look at. But again, it's very entrepreneurial culture. So when I saw an opportunity to start the revenue engine in India, we had a very frank conversation with a very limited budget. And we started, we hired out, we started our sales and I spent some time, I spent a quarter doing sales. That's when I realized that doing enterprise sales wasn't for me, right? I want to move back to the product side. But we hired a sales team out and then started to add sales in India. So my mandate was sort of growth in the broadest sense of the term for international. And obviously it started off with a big focus on India.

Utsav Somani: And in the world of AI, now people are asking questions and getting answers from these chat tools. Do you think- Yeah, is it relevant?

Gautam Shewakramani (Inuka Capital): Yeah, good question. So I definitely don't think Quora as a product is very much in the cultural zeitgeist anymore. You don't hear about it. I think Reddit on some level has sort of taken away that crown maybe, or if the crown ever even was with Quora, depending on the use case. I do think Quora the company, the Quora the company definitely still exists. I still own the options in the company. And there is a new product that they have that they call Poe, P-O-E, which is sort of like a front end for any foundational model provider. So imagine if you're someone who's built like a small language model, right? And you don't have the distribution capability of a open AI with chat GPT or a cloud, how do you actually make sure that you get your model and your capability in front of millions and millions of users? That's what Poe does, right? Poe is a front end for any model. It's the same, if I'm not mistaken, 20 bucks monthly subscription, but then you have access to everything. And I'm guessing a lot of the data set and the data and the expertise that we have or that we had in the platform and the Q&A platform has informed the product's ability to be actually very high quality for consumers.

Utsav Somani: Funny enough, how I remember Poe is because of that open AI board drama. I just remembered the Quora founder is on, was on the board of open AI, right? Yeah, yeah, that's right. That's when Poe actually, I mean, it just hit me. Dhruv, any final questions for Gautam?

Dhruv Sharma: Yeah, I mean, I wanted to ask this a little bit earlier, Gautam, but you mentioned the three themes across which you invest, including enabling Indian businesses. Yeah. I wanna ask the timing question, right? Because companies have been at it, myself included back in the day and firms have been at it since mid 2010s. Do you think it's only starting to, space is only starting to open up right now or for a specific type of business?

Gautam Shewakramani (Inuka Capital): So I think the space to sell to Indian businesses has always been there. I think the question is really whether you can have venture scale outcomes in this space. And I think the way to have venture scale outcomes from an exit and therefore, to some extent revenue value perspective, I think comes from the ability of these businesses that we back from being able to capture some meaningful value, right? That grows asymmetrically if the business is successful, which is why one thing that we are very, very clear about is the companies that we back. So, I mean, just using an example, like let's say Propacity, right? It's a real estate tech company that we back. They are actually helping midsize real estate developers sell more homes, right? Primary sales sell more homes. That's a hair on fire problem for real estate developers. It's extremely expensive to be a real estate developer sitting on unsold inventory, right? They have done this in such a high quality way with such high velocity and such high quality pricing without heavily discounting that they are now at a stage where the developers that they're working with are paying them a percentage of the sale value of real estate, right? So you're talking about, you know, let's say a home in a tier two city, which, you know, my sense is probably- That would have been no one's calculations. Exactly, you're talking about 60 lakhs to one crore, quote unquote, average order value, right? Now if you're enabling that, what is a, you're talking about a four to 6% take rate as brokerage, right? And developers are willing to give that to Propacity, right? So look at the revenue scale up potential there because they're solving a hair on fire problem and they're solving it well. So that's the kind of stuff that we look for, right?

Dhruv Sharma: Maybe one final, final question, Gautam, before you go, like just give us some tidbits from your productivity stack, your personal productivity stack, managing Innoca Capital.

Gautam Shewakramani (Inuka Capital): Oh, wow, we could do a whole other session on that. But so I think the big thing, we use a lot of AI, like a lot of AI ourselves as a firm. A lot of our deal flow ingestion is completely AI driven. So if a referral comes in, it just hits a catch-all email address that we send it to, an internal catch-all email address, that triggers off a cascading series of automations and agents that sort of review the material, found a background, everything, and give it to us in one place. We haven't outsourced judgment to AI yet. If we did, I guess we'd be working ourselves out of a job. And the whole point of this business is actually finding the outliers and the exceptions. And on a personal level, my stack's pretty simple. I use Raycast a lot. Superhuman is my sort of go-to email client. And I actually use text, well, now it's called Beeper, instead of WhatsApp desktop to manage all of my- He sold for 50 million, right? Yeah, Kishan sold for 50 million to Automatic. And then Automatic had also bought another company called Beeper. And now Beeper and text have sort of like married each other. And the love child product is pretty good. And now I've actually just started playing with the MCPs, like the local MCPs, using cloud code. So it can actually automate even my outbound WhatsApps, but on the client side. So there's no API involved.

Utsav Somani: So. Amazing. Lots for us to try on in text, is something that I do want to try. Truly one inbox for everything?

Gautam Shewakramani (Inuka Capital): Yeah, yeah. So now it's called Beeper. Oh yeah, it's one inbox. Honestly, in India, it doesn't really matter because WhatsApp is the only inbox, right? I don't think any other inbox really hits that volume, besides email, of course. So I have my email box and I have my Beeper and that's it. That's kind of my, and everything runs from there.

Utsav Somani: Awesome, Gautam. I think you've given us a segue into our next segment. Akhil, he's doing a lot in the world of social as well. Universal APIs for social. Akhil, welcome to the show. And Gautam, thank you so much for giving us the time. Cheers.

Akhil Bhiwal (Phyllo): Thanks for hosting me.

Utsav Somani: Yeah, we were discussing the world of social. So simplify what Fylo does for our listeners.

Akhil Bhiwal (Phyllo): So we are a social data aggregator. In simple words, what we do is we integrate with platforms like YouTube, Instagram, TikTok, normalize that data and provide it into a common set of APIs. Those APIs are like identity, engagement, reputations, audience demographic. This largely gets used for influencer marketing. That's a big use case. When you want to understand an influencer across their digital footprints, which might be on YouTube, Instagram, TikTok, Facebook, or any other channel, we provide that data with a common set of APIs so you can profile the influencer. Many of our customers happen to be influencer marketing platform, marketing agencies globally, marketing agencies, PR agencies, background verification agencies globally. So these are companies like WPP, GroupM, Dentsu. They become our customers. And we got inspiration. In India, there's a product name, Account Aggregator Framework. So there was a company in US named Plaid. So we took inspiration from Plaid. We started as Plaid, but for social data. We saw how Plaid solved the problem for FinTech and it enabled FinTech to grow. So what we realized is something like Plaid is missing for social data. Integrating with these social platforms is a pain and it's really tough to work with the social platform. And the key over here is, these platforms provide some APIs. These are one of the most advanced tech platforms, but they don't expose their APIs or their infrastructure for developers. And this largely started when Facebook faced Cambridge Analytica. So they were too worried in case of working with social platforms. There were companies like Zynga and others which built, but then an event like Cambridge Analytica happened. So after that, all these platforms made their developer docs very obscure, their integration process very obscure. And it's by design, not like by accident. This obscurity happened because they don't want their system to be, their data to be misused in any way. So someone had to work with them. It was very complex. And these platforms are like governments. So someone has to work with them to understand, hey, what your incentives are, what can we do so that you are ensured compliance happens? You don't have any, another Cambridge Analytica kind of event. And then provide that API so that developers can build on top of them. So in nutshell, this was a very long one, but we provide, we are a social data aggregator. We provide APIs for developers to build on top of social platforms.

Dhruv Sharma: Okay, just to be understood correctly, did you open up a conversation with Meta and with all the other social media giants?

Akhil Bhiwal (Phyllo): Yes, so our customers are companies like FinTech companies. So they use this data for fraud, et cetera. But when they have to integrate with these platforms like Meta, we open up that conversation. We build our integrations with Meta, with Spotify.

Dhruv Sharma: You're a trusted partner for all the large social media companies and platforms.

Utsav Somani: And were some of these harder to get than others? You said that I think 100 plus platforms can be accessed via your APIs.

Akhil Bhiwal (Phyllo): Yes, so these platforms are difficult. Like, so initially we thought like, hey, we will do a revenue share. And then once we started the business, we quickly realized like, whatever revenue we give to them doesn't matter. They like Meta's last year revenue was around $165 billion with $65 billion profit. We trying to share 1 million, 2 million, 10 million revenue, 100 million revenue has no consequence. No one would even listen to us. So revenue doesn't matter to them at all. I mean, revenue share partnership doesn't matter to them. So each of these platform has a different incentive and someone has to work with them. Like Spotify, they want to promote more indie artists. So it's a publicly listed company. They want to increase their margins and they want to promote more indie artists. Can something be done around it? Meta, after Apple's policy changes, they want to promote some APIs so that they could capture how the ads are performing. So to understand each platform's incentive, work custom with them, that's when they start opening up their APIs. One of the key thing we do is, we've worked with all their customers who are spending $100 million plus on these platforms, like YouTube, like on Google's platform or Meta's platform, et cetera. These are companies like Unilever, Pfizer, P&G, et cetera. But you can't spend like $100 million right off the bat, like on these platforms. You need a lot of custom work to be done. So that's where we work. We come into play, we build those integrations, we build those infra, et cetera.

Dhruv Sharma: And- Akhil, when brands are about to partner with a creator, help us understand to what extent, to what lengths do they go trying to do a 360 on the creator and their legitimacy?

Akhil Bhiwal (Phyllo): So before Fillo, this was mostly an approximation. So someone would just discover a creator on Instagram or TikTok, and then try to find it. I mean, try to make an approximation of it. Now, more and more it's getting data-driven. They want to understand how much is their audience relevant. So let's say if an example like Nike, if they're launching shoes for women in the area of 25 to 45 who are into running, and they find a creator whose audience matches the kind of Nike is looking for. But the influencer is talking about food or is talking about alcohol somewhere that doesn't match with Nike's brand tonality. So prior to that, it was very expensive and not scalable.

Dhruv Sharma: So with tech- And not empirical either, like you said. I mean, maybe they were making approximations and then the campaign never went as intended.

Akhil Bhiwal (Phyllo): Exactly. And we provide this data. So with data, they're able to make more informed decisions. So now, one is Nike understands like, hey, my brand tonality matches with this creator's brand tonality. So then they can perform a much better, they can run a much better campaign.

Utsav Somani: And there are these influencer marketplaces in India and globally who sort of, I mean, suppose they say, I mean, you can go and say that I'm a brand in the sportswear category. I send D2C yoga pants, basically. And I want to sell these $1,000 yoga pants to people in just Michigan. And you can find these creators who are local and micro, nano influencers, and you can fine tune all of this. So how are these platforms getting that data, which is so granular? Is it through tools like Philo?

Akhil Bhiwal (Phyllo): It's mostly through Philo. So 70% of the time, one of these platform would be our customer. It's like globally in marketing agencies, PR agencies, or background verification agency, they are our customers, most of them globally.

Utsav Somani: And you mentioned, I mean, APIs. So you have a focus on dev tools, right? Basically, or developers, basically. So how can, say, an agency which is not tech friendly, but still managing a lot of influencers work with you?

Akhil Bhiwal (Phyllo): This has been a constant ask, and we have not been able to service that till now. We did thought of building a platform, but that's what our customers whom we enable it to do it. So we generally direct them to one of our customer, like, hey, you can use this customer's platform. Or there's something lately we've been thinking is, can we build AI agents that can do some of this work? So those are things we've started building. And what we have started realizing is, these AI agents are more helpful for brands. So lately we are working with retailers like Ulta Beauty, Sephora in US, or consumer brands like Kellogg's, RacketBandKaiser, et cetera. So it's, apart from data, we've started also launching AI agents, which can do a lot of this work. Whatever the platform was needed, either we, in case of platform, we redirect them to one of our customer, whosoever can't build tech in-house, or then we provide them AI agents to do that task.

Utsav Somani: Give us some numbers, like how many influencers are you tracking? Or, I mean, it's very hard to put these numbers, or any clients and API calls that are happening via Filo?

Akhil Bhiwal (Phyllo): So we have around 200 customers globally today. Primarily US, 70% would be US. Around 100 million API calls per month, 25 million creator searches. Those creator search that happen, primarily many of our customers are looking for, so these are marketing platforms, or marketing agencies, which serve like a brand. So these, each marketing agency would be serving around 25 to 50 brands. So that's there. So around 10,000 brands that keep actively looking for creators. Around 25 million creator searches happen monthly, on a monthly basis. That's a high-level number. There's another category we recently started doing, was background verification. So a social media background check of someone.

Dhruv Sharma: Is that like immigration departments asking you for social media data?

Akhil Bhiwal (Phyllo): Yeah, exactly, figuring out, how can we sell to US? So, just a fun fact, you're gonna make some enemies in that process.

Utsav Somani: Income tax authorities as well, right? I think they're also tracking your social media data, like if you're going on these foreign trips and reporting, under-reporting your income, I think they might catch you.

Dhruv Sharma: By the way, so I got distracted, I think you said $1,000 yoga pants, my mind is like, $1,000 yoga pants?

Utsav Somani: Hypothetical numbers, hypothetical numbers.

Dhruv Sharma: No, no, $1,000 can secure you a lifetime of yoga, you should be able to secure your lifetime supply of yoga pants.

Akhil Bhiwal (Phyllo): No, but someone goes to Lululemon, they will take $1,000, in some cases, they're expensive. So, immigration has been, my sense is this income tax thing is more a, just a way to create fear, because I don't think so, that tech exists for most people. To give a sense, US Department of Immigration, I'm not forgetting, it's called USCIS, US. ICE. ICES, yeah, Immigration and Custom Enforcement. So, basically, US Immigration, they work with Palantir, they give them $55 million to do a social media background check annually, and that happens manually. So, and there's an officer at, I mean, there's an officer who gives those files to Palantir, Palantir has a team of people who is sitting and manually reviewing it.

Utsav Somani: Man, this is the first use case of Palantir that I've heard like properly, like, what does the company do? At least I know one thing that I can- Create value for shareholders to do nothing.

Dhruv Sharma: So, finally someone made it real, yeah.

Utsav Somani: The company's been having crazy valuation, so I'm sure that they're doing many other things, but yeah, this definitely is something that I can add to that list.

Akhil Bhiwal (Phyllo): Yeah, and so it's 50, and it's $55 million per year for manual check. And I'll tell you why this has been difficult till now, why I think Indian Income Tax Department and all those are just saying it, and it's not practically happening, because you can't assess so many social media content. So, one of the key value we provide to customers is, many of our customers are, let's say a brand like Sephora, they pay us $200 per background check report of a creator. You have to go through years of content, that would be thousands of content pieces, videos, images, audio, text, and you have to look into each video, so doing it in a affordable way was not possible before, some of these LLM models, and I can say it like before Gemini 2.5. So this was like maybe few, six months back. So, because we have been trying to do it in a cost effective way for quite a long time, and it was not working out for us also. Lately after these advances in LLM, we also were able to do it. So the tech just got there to make it scale.

Utsav Somani: So, creators are now sort of influencing culture and many other things, right? Buying decisions and spending, I mean, so many things that they influence, and you have, I mean, you mentioned in your last statement that there's so much data that has to be analyzed and many of the models previously couldn't allow that. I know you can't pick topical things, but is there a way that Philo can predict the trends? If say I'm a brand, what will go viral? Like, can I use Philo for that in the future and say that, hey, this is what we should do?

Akhil Bhiwal (Phyllo): So, some of the work we are doing with consumer brands is around predicting trends. They are trying to understand, hey, what kind of packaging should I do for my product? What is working? Like, how many people are searching for sustainability? What is it? So, by the way, when you said creators are creating trends, we also call creators as tastemakers. They are creating the taste what many people get influenced with. So, many of these consumerLike, Dhruv, we are tastemakers, okay?

Utsav Somani: From now on, anybody who says we're podcasters or live streamers or anything, tastemakers.

Dhruv Sharma: We're tastemakers now, yes. Tastemakers.

Akhil Bhiwal (Phyllo): So, many of these consumer brands, so some of the work we are doing with Racket Benkaiser is that they want to understand how to package their product better. And they want to understand what moms- So, we are working with their baby formula division, which makes baby formula. So, the moms are the buyers. So, they want to understand what moms are looking for, what kind of packaging, positioning they're looking for. They want to understand the trends around it. So, one is understanding those trends to either change product or packaging, one of these things, or messaging. Another is understanding what competition is doing. So, what kind of new categories are emerging, which did not exist before. So, the trend is being one of the big thing that's- That's an important category for us.

Utsav Somani: Dhruv, any final closing question for Akhil?

Dhruv Sharma: Yes, of course. So, what's coming next for Fido? But before that question might take a while, Akhil, so, I was also thinking about privacy concerns because you're a global business. Of course, you're talking to three different jurisdictions, GDPR, this and that, and then the right to be forgotten. How does those factors sort of interplay with each other? Or for instance, if someone locks their profile, or if the profile is private by default, can someone like you gain access to anything that used to exist at a previous point in time? How does all of that work?

Akhil Bhiwal (Phyllo): So, let's take an example of a social media background check. If someone wants to do a brand, like to do a vetting of an influencer, and if a profile is private, we cannot do it. Practically. So, privacy is like at the core of it. And we have to comply with GDPR, CCPA, all these regulations. So, in a private profile, it doesn't work. It mostly works when the profile is public, or there is some digital footprint available.

Dhruv Sharma: Obviously, you can't vet retired influencers or creators because that's the final step in that process to lock their profile. But yes, what's coming next for, talk to us about your plans for the future.

Akhil Bhiwal (Phyllo): A big thing is we're trying to build around social media background check a lot more. So, for immigrations, we are figuring out how we can sell to governments.

Dhruv Sharma: Dude, you're gonna need a security detail very soon. That's it, that is the roadmap.

Akhil Bhiwal (Phyllo): We did serve Indian government during Operation Sindhu for a small use case. I can't talk about it much, but the social data is very important.

Utsav Somani: Otherwise, the Indian government will have to remove you.

Akhil Bhiwal (Phyllo): We did service them.

Dhruv Sharma: So, I think fighting disinformation becomes a very, very important thing during the heat of conflict. And so, if home-grown tech companies can come to play a role in that, then nothing like it.

Akhil Bhiwal (Phyllo): Yeah, that plus this lot of open source intelligence that comes, which gets recorded by satellites, bots, and gets posted on Twitter automatically. There's not much way to collate that information and make it capture insights out of it. So, we are figuring out how we can sell to governments and large enterprises. So, bringing AI agents for these consumer, for large CPG brands, so that they can understand trends I love your play to Palantir, Akhil, I have to say.

Utsav Somani: All right, Dhruv, people are making fun of our turtlenecks on the live chat, but there is one question that I think we can end with. Any predictions for the creator economy in the next decade from Akhil? Sorry, what's up? Can you repeat that? Yeah, any predictions for the creator economy in the next decade? Next decade, I think is too long, but maybe let's do next three years.

Akhil Bhiwal (Phyllo): I think people talk a lot about AI influencers. I don't think so AI influencers will pick up a lot. It won't be there. It would be humans telling the story and leveraging AI, definitely. So, AI influencers may not be there. I think 50% of our social media feed would be content generated with AI. That would definitely be a true fact. And that might lead a new kind of social platforms which don't exist. Like today's social platforms were designed for humans to post content, but when 25 to 50% of our feed is content, I mean, generated with AI, then the platform might also be a very different one. A new social platforms will emerge.

Utsav Somani: Awesome, thanks for sharing that. Perfect, I think that's a show ender for us now. Thank you, Akhil, for giving us your time. Cheers.

Akhil Bhiwal (Phyllo): Thanks, Adshanth. Thanks, Dhruv. Great to see you, Akhil.

Utsav Somani: Yeah, see you. All right, so let's end the Wednesday stream here and I will see you on Friday at 4 p.m. Cheers, bye-bye.