Summary

The Offline Network Episode 37: Fintech & Wealth (aired 2025-11-24). Guests: Dhruv Pathak from INDMoney. Topics: venture capital and funding, AI and LLMs, consumer brands and D2C, health tech. The Offline Network is India's live show on startups, tech, and venture — streaming M/W/F at 4 PM IST on YouTube.

Full Transcript

Utsav Somani: . traveling right now. So we've got Shrijan Mahajan who's stepping in. He was a drummer at Parikrama and also on the show before. So we have a literal rock star to kick things off with us. So let's do music first. Today we're going to talk about AI in gaming but on to our guests in a bit. Spotify wrapped. Why are people obsessed with this thing?

Srijan Mahajan: It's crazy man. I've never seen and I remember when it came out first like a few years ago it just went insanely viral right and then obviously like your product POV it's great because you everybody wants to show off good taste in music right and they're very well curated because they'll make a nice design story and they'll give you stats that you want to show off to people including now like where they're showing you like your age and your listening age and whatever they've sort of got these things and mine is horrific if you look at it and you can bring it up I think.

Utsav Somani: Let's share it like so like so I'm 18 by the way. I listen to lounge and lo-fi beats. There's also some AC in it. Yours is 72 man like what are you listening to? I've been listening to so much old school stuff right and Hans Zimmer as well. It's crazy because as a rock star drummer I could have assumed that there will be a little bit of indie rock as well in there.

Srijan Mahajan: There's a lot of old school rock and roll that I've been listening to but somehow it hasn't come out over here. So I don't know why but it's there but it's crazy man they've launched DMs also and they've done this and I don't know maybe like there's speculation that they might be I mean obviously every app becomes a dating app and you put in messages.

Utsav Somani: Hey you listen to Taylor Swift that's the way to like slide into somebody's DM later but it's one of those features like even YouTube has come out with their recap this year. It's one of those features like how Snapchat stories right everyone tends to watch it because it just became so viral but I think the originality of the concept really stays with people and people are generally I think very self-obsessed like I think they just show off good taste and good culture and in December I think everyone starts doing these 2025 recaps so I think it fits into that theme as well.

Srijan Mahajan: It's amazing they've done it really really well and I think they've nailed the format. They've nailed it directly to share the stories. They've got just the right amount even the kind of design that they've got it's amazing and people are voluntarily like everybody who's a listener is voluntarily sharing Spotify on their stories. I mean what better growth hack can you think of right it's amazing.

Utsav Somani: That's crazy but talking about a very different topic Indio flights are getting cancelled left right and center. People are even calling it arm twisting the government to roll back the rules and I think as of last hour the government has also rolled back the rules where pilots were supposed to get 12 extra hours of rest. Do you have context on this overall? Have you been following this?

Srijan Mahajan: I haven't actually except for the three phone calls that I got from friends last night saying that hey actually flights are cancelled can we crash at your house and I'm just like yeah come to my house but that's all the context that I have but it's crazy because I mean obviously there are these questions of like a duopoly and whatever else is right but at the end of the day I mean this is free markets right so there's some tweet about you know Indigo flights are cancelled so Air India is charging 50,000 bucks to go to Bangalore of course they are charging 50,000 bucks for the demand and supply so these are all like these are this is how it is right what do you do? Cloudflare goes down half the internet goes down Yeah, we were worried that today's stream won't happen.

Utsav Somani: Raj was working like one hour back, crazy what it does. Even Sumit of CoinDCX was posting that CoinDCX is down so like 30 minutes of like downtime I think can be a loss of business but I think I mean pilots are saying that they want more rest but I mean I saw some pictures of what's happening at the Bangalore airport and what's happening at I mean there was a person who couldn't attend his own wedding reception because he missed a flight

Srijan Mahajan: It's crazy man, it's also I think somebody had put up something really really nice where like they said that you know when there is some of a system to function well need some amount of slack in it and when you make a system or design a system like indigo which is so tightly wound right the one domino falling in just sort of breaks everything down and if you if you notice like indigo they're like running like minute to minute right like if you go to a like if you go to on an indigo flight they're so paranoid about every minute and then like on time performance this that the other and now suddenly one thing happens and whether it was orchestrated or not that is a separate matter all together and I don't think we'll ever know that it was or wasn't but man like this is how it is.

Utsav Somani: They put out a tweet today like all their flights from Delhi are cancelled Chennai 6 p.m all flights are cancelled and I mean I had some indigo flights booked later this month and they had posted that February 10th is when they'll resume proper full operation I was like wow but let's talk about some startups and finance Nexus of course closed their latest fund fund eight at 700 million and they're going to dedicate half this to India and half of this to US and you were mentioning one thing that we have to discuss with Akrit as well that why is Indian AI founders or Indian AI founders like this workflow that we covered in the DPS segment Tane all of them based in the US raising from US global products but is there not enough capital for this I think something like that is a question that Akrit might be able to help us.

Srijan Mahajan: Yeah I mean he'll have more insight on this but this is a very interesting thing because I read somewhere recently that like I think only about 30 percent of Indian VCs are putting in money into Indian AI firms and the rest of it is going abroad and in India people are putting in money into very different bets right so you're looking at a lot more consumer bets and a lot more infra kind of bets and whatever it is so it's a very different game and I don't understand it well enough because I don't think I know enough about it but we would want to maybe explore this with Akrit because he'd be the best person to tell us about this because he's returned from the US he's built in India and now he's the fund for AI in India so but also interesting key all of the

Utsav Somani: funds like Lightspeed is I mean of course one foot in the gate in the US even Peak 15's hired somebody in television was at the YC demo day so everyone's going to YC and YC suddenly I think they were declining and suddenly they're back up and every AI founder wants to go to YC again because of market access and capital.

Srijan Mahajan: Yeah it's crazy I mean I saw recently I saw PGN, Jessica and TBPN you know which is like I was just like wow this is amazing.

Utsav Somani: Yeah dude even TBPN announced a partnership with NYSE yesterday New York Stock Exchange and they said that 2026 ad slots are all sold out. Those guys are crushed dude but Nexus a bunch of them I mean will be going live Postman, Zepto so they've got a pretty healthy portfolio Rapido is also there so I think four five five IPO pipelines in the next two three years. OpenAI bringing their Stargate project to India negotiating a 500 megawatt lease with TCS Hypervolt for India Compute so India AI mission and all of these things are taking off finally.

Srijan Mahajan: I have one one question and I mean I don't I don't know enough about this but maybe you do like do you think there's space for like like or compute or like data center kind of players for like smaller players or only these large scale players can actually step in right or is there a like I mean because obviously the capex is so high that you can't really do much but is there a way of let's say fractionalizing this and maybe setting up something where smaller private players can actually jump in.

Utsav Somani: I honestly don't know what business model to follow here but I think it's a big capex game here I think we've not really had like I mean if it was eternal like maybe 20 years down the line where they've built up 800 billion in cash reserves then make it out and then offer it to other people like AWS does right like then it would make sense but none of our IT outsourcing firms have that innovation mindset or R&D spend or the capability or the I mean the leadership required to like build out a compute scale like this but I mean take some movement like OpenAI taking India very seriously they've given away that chat plan for three one year everyone's coming for usage but this is the first I think instance where we've seen they're actually playing the infra game as well which I think I mean I just feel like this is

Srijan Mahajan: literally the rail the internet for the next like 20 years right and I'm conservatively speaking like 20 years so if that is the case how does a like and I think Rahul from BBC had done like a deep dive onto data centers if I'm not mistaken but how does like a common man get access to

Utsav Somani: uh the upside into this yeah I mean this one is a clear tata project so tcs I think I mean there are companies which are raising at 8 12 12 billion so there was a tweet which said private markets are the new public markets like I mean you go when you have to dump stock to retail but apart from that all the value capture and everything is happening in private so anthropic IPO is happening next year they find this banker so it'll be interesting to see what value gets created for even retail investors that are going IPO at like I don't know 300 last round so I mean yeah now we're in the trillion dollar era basically so quickly recapping some things that we saw on the internet peak 15 has crushed it dude how times change when they split off from sequoia everyone in the ecosystem was saying your shit but now in the last year 2025 most year-end review of peak 15 will look very insane grow is the big one then there's me show wine labs wake fit going soon mama the office exigo like I mean they've

Srijan Mahajan: really done well it's crazy man yeah like and what a turnaround right and I don't even think it was a turnaround I think they were just waiting because they were just like yeah they place all these bets and everybody is crying and everybody is talking about you know oh what are these guys going to do this at the other end by juice the entire thing that by juice happened and whatever and all of that happened and they just sort of stood their ground patiently waiting and being like it's okay we'll play the wrong game you wait and watch and now suddenly they're sitting on 30,000 crores worth of holding that's crazy yeah all right let's

Utsav Somani: our guest is almost here so just uh quick recap uh do you think it'll happen and another interesting thing that I saw is dude Netflix is possibly in the line and finalized a deal to take over warner brothers uh discovery and uh they'll own hbo max as well so they might reboot game of thrones in 10 different ways now the funny thing is 5 billion dollar breakup fee man convincing the department of justice to let this deal go through I think will be insane uh so they'll be confident billion dollar breakup fee

Srijan Mahajan: rocky here they think i'm sure and i'm just hoping that's all

Utsav Somani: yeah india remember the tvf guys did something i think very funny but apple bingo card is looking very strange where everyone's going tim cook might go next year

Srijan Mahajan: why do you think that's happening i don't know and i mean i just think that it's a great thing um to be honest because as a like a lifetime fanboy of apple i have seen it declined right and in the past few weeks i have had some horrific encounters which are too long which i won't get get into right now but i also feel like this is exactly what apple needed uh i don't know if it will be come out if the result of this will be good or not especially because tim cook is also and all of the leadership is going to leave uh but i think apple has been resting on their laurels for way too long and they need a big shake down whether that shake down comes up in a good way and the result of it is good or bad time will only tell but i think it's a great

Utsav Somani: thing that this is happening all right without further ado let's welcome anurag from felicity games anurag welcome to the show hey what's up hey shrejan how are you thank you for giving us the time for the listeners who are hearing about felicity for the first time can you spend two

Anurag Choudhary (Felicity): minutes describing what the business uh does sure uh so we are a game publisher uh what it effectively means in this uh world of uh you know low attention span where almost every digital first product is kind of you know yearning for your attention and the success rates are abysmally low when you launch a new game or an app uh what we are doing is trying to find games which have uh commercial uh you know potential and how we do that is by working with a bunch of studios uh on prototyping games and figuring out what ones which have uh those legs and wings to perhaps take off right and if if they do then we take over the game acquire it and scale it right so that's that's uh that's what we do we are 18 months old uh we have worked with around 10 studios uh tested around 25 games till now uh found two uh between which you know we do a dau of around 45 000 uh almost clocking almost 300 000 in monthly revenue uh and most of our revenue comes from the u.s but the studios that we have partnered is from india so a very different approach to building a casual gaming business uh more distribution first rather than more product

Utsav Somani: and how does the consumer experience your business like is it that they download an app felicity and then you can like a game center choose like different things like a playstation plus and you can download that game uh like an in-app thing or do you uh so you're not a publisher yourself so how does it work for a consumer uh how does it work so uh in that essence every game

Anurag Choudhary (Felicity): so it's on felicity's play store or app store handle and of course any other alternative channel where we would be publishing the game uh but yeah every game is very very sharply positioned so our largest game is a solitaire uh and we actually try not to bring felicity into the frame it could be an experience which you know is very felicity like but uh we want to shop we want to position our games sharply for the need state that the user has to play a ritualistic game uh we are kind of uh uh you know in the back end uh making sure that they have a experience which is personalized and tailored to the challenge that they feel is apt for them

Srijan Mahajan: but this is the thing right like distribution is the only mode remaining now because creation has become so democratized that like i mean your idea is obviously there but like how you distribute it and how you position it to fall into the context of your user's day is the only thing that is going to be of massive value going forward yep so i mean before this i attempted to

Anurag Choudhary (Felicity): you know build a short video app uh so this is like my second startup in a row and there we were trying to orchestrate virality and what i realized is that there is no point building a business on hope uh you know so the only way to kind of uh have uh you know clear line of sight to your pmf is if you can really master profitability at on paid ua right so at the core of it casual gaming and a lot of other app first businesses today are ua arbitrage businesses and that's you know the ability to uh test iterate and figure out the economics on paid cax if you can really really unlock that cash flow is give you a higher chance of success then only and only thinking product first of course that doesn't mean that the product is not important it's very important but where you end up spending your time at least that you can de-risk by uh you know trying out a bunch of things uh uh you know up front rather than uh figuring out towards the end of the journey that hey i

Srijan Mahajan: have a lot of money this is not right i think product product stays stable stakes right like product is stable stakes like a great product is stable stakes but the only differentiation now is positioning and and how do you take to market correct give us some specifics on

Utsav Somani: that like what games have worked and why have they not worked before or are there other players doing

Anurag Choudhary (Felicity): it like some examples or case studies i think might be helpful sure uh so if i talk about uh so let me also just break down gaming at large right and mobile gaming specifically pc console uh you know we'll not talk about it as much but mobile gaming can be largely split into you know casual gaming so casual gaming though there's a word casual attached before it so it's not considered gaming uh it's considered time pass right so and usually uh players here are mostly skewed skewed towards women and the need state is that of watching reels like you know you're just killing time and many times when you're playing a casual game you're not just playing the casual game you're also watching another screen uh right so you're doing a bunch of things together uh right and these are typically solo journey plays you're not competing with someone it's just uh you know anti-brain drought perhaps you know you're making some progress on the game you know you're leveling up so you're happy i mean it presents a right amount of challenge so puzzles uh something like is right up there right so this is casual gaming then there is mid core uh which is clash of clans where you actually you know uh uh i mean it's a multiplayer uh game wherein you know you're you're you make a plan and then you're you know attacking someone's village and then there are triple a games which basically like a first person shooters etc uh mid core and triple a are like hollywood productions they require a lot of deep systems working for them at the core and uh the amount of capital required to build these games is immense at least 50 million dollars today if you were to build even one game and even then uh it may not necessarily work out casual on the other side has gone through uh you know several motions wherein initially i mean the motions were set by zinga which is the og here right which actually brought in a lot of analytics into uh core game play uh they basically i would say more of an analytics company than a gaming company uh but over time uh zinga also enjoyed a lot of organic installs so organic installs are dead today and uh that's going to just the amount of digital first inventory so there are a few companies globally so globally i would uh you know uh voodoo v-o-o-d-o-o voodoo uh homa h-o-m-a homa uh these are all like uh unicorns wherein what they have done is they have innovated on the supply uh where they've worked with a very very large number of studios prototype games but they're going after hyper and hybrid casual games which are not very deep so effectively their cpis will be low 50 cents 60 cents and they will be able to you know churn let's say a one one and a half dollar ltv on it and they'll be able to do it repeatedly over and over again across a portfolio of games so this is a portfolio creation rather than building this one big game like one candy crush or one royal match it's a portfolio creation game and i would today call classify game populations into two two categories one which are trying to still attempt product first second distribution first there's no right or wrong uh rather i mean and i still believe that product first will have a much non-linear impact right but it's just more difficult uh so marketing first globally voodoo homa triple dot uh today it's almost as good as two billion dollars in revenue uh these are these are some of the large publishers now coming to the market actually that it's large enough like casual gaming alone is around 25 to 30 billion dollars and the thing here is it's not a winner take it all market so you know you can participate uh so the top game let's say in the genre that we operate in would be at 80 to 90 million dollars but even if you're the 10th best game in that genre uh that's a 5 to 10 million dollar outcome so what we are trying to do is not necessarily be the number one today rather we're trying to take that position of the 8th or the 10th best like our soliders right on top 50 card games in the us the idea is to take it to top 10 right so that's the next challenge uh once we get there will be 10 million dollars in the holiday but yeah enough of those aggregated compounds as uh uh compounds as a as a portfolio uh and largely the outcome that you're trying to drive is also a portfolio outcome rather than this one game or one ip which

Utsav Somani: really scale so it's sort of like a venture capitalist for casual games basically for us

Anurag Choudhary (Felicity): yeah people do make that uh you know parallel but the only difference is that uh uh you know the uh you i mean unlike you operate the games that's one that's one key difference and the same operating model can also power a bunch of other games right so uh that's the that's the only difference because if i just talk about player experience a lot of like i i talk about this a lot so why does someone love a game it's about how much challenge did you present to them and it's very subjective it's very difficult to measure this like some feel challenged enough versus region that is something which comes down to being able to personalize at a cohort of one you know personalize your in-game journeys and that engine if you're able to build you can run it across multiple games right same with the user acquisition funnel right the way you run creatives the channels that you operate them on the predictability that you can drive in terms of ltv so can you figure out in three days whether you'll be profitable or not all of these are compounding pieces of infra which you build as a publisher which can power a set of games so yes i mean it sets you up to become perhaps more of a private equity perhaps if not we see that you know over time you can start building your own you build your infra but you can then be a little more inorganic about getting games but i mean that's not a focus at the moment right now it's mostly

Srijan Mahajan: building from zero to one crazy um and is this like like your ip acquisition thesis this is mostly in the casual gaming space or like how do you think about this and what do you what does what makes it worth you to get a game into felicity yeah so initially to be honest

Anurag Choudhary (Felicity): we did spray riddle and i personally don't come from gaming as a background before my previous startup i was with like swiggy and snapdeal for almost seven years so i'm in the two years of operating this what i've realized is there are two types of games one which are impulse led which will get you a very cheap download but the novelty dies fairly quickly right uh so it's like um you know something it's like a trend at best right yeah and then there are evergreen games uh which have been like you know which you don't need to explain so solid air it's an evergreen game maybe ours is a 10 000 solid air out there word games uh sudoku and what these games typically have is great long-term residual retention uh so you know you will have d60 d90 d180 d270 i don't have a d365 to report as yet on my game but yes i do have a d270 user still playing the solid air uh and what i noticed at least between these two genres uh the uh uh the evergreen genres are very well set up for long-term retention which gives you the ability to even to push back your paybacks so you know your cap will only increase that you scale so if you were spending a hundred thousand dollars a month maybe you will get a user at two dollars if you spend million dollars the gap will go up yeah but if you have the ability to have that residual retention at the long term you can keep pushing back your payback and still uh turn out profitable right so so executive summary on the games that we picked today are ritualistic puzzles at least that's what we're focusing on at the moment which have long-term retention beyond d120 um and those are like your typical suspects you know like card games word games uh uh you know uh number games uh

Srijan Mahajan: sorting puzzles etc so quick quick follow-up on this i play like sudoku a lot right and i have like five different sudokus on my phone and every time i'm playing five different sudokus on your it's just always because i get bored of one kind of puzzles because then they become very predictable and i finish like the extra hard one and then i'm just like okay now let me get another sudoku app and it's the same thing right but then i get that and then in their ads they have these weird games which are very very interesting which are these sorting like test tube sorting color sorting those those kind of games and those are what i like i i am really tempted to download them and i play them very properly for like a day or two days and then i'm done with them correct

Anurag Choudhary (Felicity): correct correct correct so that's what you know months basically yeah yeah so they're i mean so they will so those are more hyper casual nature wherein they're playing on an impulse uh actually gaming everything is playing on your impulse but uh you know it's like uh having a pizza versus having dal chawal so like uh you know pizza games you like you know just devour them quickly and that is the end of it um at least we have come to the conclusion that dal chawal games is where the longer outcomes are and like you can sustain their revenue and scale for six to eight years let's say the earlier one where it will be a quick blip

Srijan Mahajan: so even with this sorry with this analogy you're looking at um pizza games to be hyper casual dal chawal games to be your kind of uh gaming and then like a very healthy salad to be rock

Anurag Choudhary (Felicity): star games kind of gaming correct yeah so those those would be uh actually rock star game so even their uh pc console gaming we i didn't touch upon them but uh the analogy that i'd like to give is console games are like watching movies right this is a point doing high production to make and uh you know you will give your undivided attention to like just playing them so that's what rockstar is it's like making movie uh right with a billion dollar plus budget then still waiting yeah i'm still waiting here and then then there's uh and some games of course you know are both on pc and console then then there are like mid-core games which are like more of otts their chapters you know you you'll wait for the next and the next but mobile gaming largely is like watching youtube or reels you know it's not very high investment uh but yeah you will keep some games uh and and your repeat will not be daily uh so perhaps in a 365 day window you will fire those games maybe 90 to 100 times a year but if you manage to stay in that phone for a year you are a very profitable customer for the for the publisher and when you

Utsav Somani: acquire i mean you said you work with 10 indian studios like when you work with them the ip acquisition strategy i think must be very interesting so share some insight into that because if you're say buying a game what do you tell the i mean the producer of the game key why why should they sell to you instead of like just doing it themselves on the app store apart from the distribution angle and when you do this decision matrix framework in your head key i'm acquiring this game there must be some science into this whole thing where you said this is the white space that we're observing in the thing or it's a completely new category like i mean angry

Anurag Choudhary (Felicity): words actually we're not uh we're not trying to go for new categories rather we are trying to go after the boring categories which already exist but yes there is a way to kind of you know make that funnel and but first i'll take the question around why does the studio work with us and are we in

Utsav Somani: that game like do you expect it to pay back in six months one year like i mean it's basically

Anurag Choudhary (Felicity): like a content library that you're building yes yes yes so uh so let me take all these three questions uh right so first is uh why would a studio uh you know even choose right so today uh since organic installs are dead uh actually the cost of testing the game on real users uh and running split tests and experiments and improving the metrics that's higher than actually building the game right so let's say you build a game uh uh and you want to now test what what it's d1 retention is you need a thousand users to even test the first day's retention uh you can't extend the uh same retention that you have in india to us so you would have to acquire and us is the market which you're going after which is the largest in mobile and casual that's at least thousand two thousand dollars spent additional right uh and now imagine testing d1 d7 d30 maybe you don't even have that kind of content many a times what we have seen that the testing budgets are actually higher than the production budgets uh right and thus if you're a bootstrap studio uh unless and until you have some cash cow it's very difficult to uh you know like put that kind of investment into just testing your game on real users there's no such thing as play testing in mobile because um you know you want to really really get that feedback on paid users which is where you'll be spending your money uh your play testing will give you cute results wherein a lot of people will say yeah it was a good game you have to basically nobody states it's not stated need at all right so that's that's one now uh which so uh in india there are 400 studios in vietnam there are some 5000 studios so some studios self-publish and they found that one cash cow and yes for those studios there is no reason for them to work with felicity but there are like hundreds and thousands of studios which don't have that access wherein they are looking to build better games they want feedback loops on how to build better games and thus a publisher like us when they come with okay this is a genre that perhaps you can attempt we are going to test this and also align your incentives and payouts basis how good the game is so what we do and this is now the second part of the question which is what is the process uh effectively let's say we commissioned a solitaire with you uh what we will do is we'll just give you three to four thousand dollars and uh then once you've built the game we'll test the game in the us for early retention and session time what we're trying to test is did the users like the core game blue right and this is where most of the games die but we do align an incentive that if we have certain thresholds that let you cross we will give you even more money this time you'll get twenty thousand dollars instead of five thousand dollars right to build this game further so it's in a way stage gated uh payout right the next payout they build this deeper game uh this time we are testing the game not just for uh retention and acceptability we're testing it for monetization potential right so we are trying to discover our doubt average revenue per daily active user and because if you think of this there are only three things in the equation cac and i mean cac ltv and ltv is a breakout of after one retention so we are trying to discover those and if those are good where we can see profitability is there that's when we make an offer to uh you know buy out the game and that will be an even bigger cash cash payout right so that way the studio's incentive is also in line and they are also learning on the fly because there's a lot of failure that happens in the top 100 sales so they also try to improve their games to you know adjust to get to that better d1 and d7 and uh third return on investment so today the games that we operate are not very very expensive that i mean you know uh the best the biggest amount that we would have spent on acquiring a game would be around 80 to 90 thousand dollars uh but that game is already doing like 150 thousand dollars a month in revenue so uh so it's uh it's uh it's already got a good investment but if i look at it from a cap perspective then typically a six month payback uh is what we uh tend to be disciplined around i mean you know there are games which operate at two three year paybacks as well but given our scale given the amount of capital in the bank we don't think that's the game that we can play today we can play it maybe you know once we are much bigger we can we start to make a lot more cash so today we have kept our uh we we we operate with discipline that in 180 days we have to break even uh and uh typically 150 percent in a year so sorry 150 percent in a year or 180 percent in two years is considered top quartile in terms of returns so because if you can do that at like four to five million dollars in monthly us spends that's like uh that's basically a hundred million dollar business right so so those are the benchmarks that we operate around awesome anurag thank you so much for coming on

Utsav Somani: and introducing felicity games uh thanks for having me over cheers thanks anurag all right uh folks let's uh switch gears up a little bit uh hot topic ai and we've got the king uh in the room now akrit welcome to the show oh my god you were the first guest on our show uh we've improved uh today i have a guest co-host frejan uh he curates a very good newsletter and is the drummer for this band called parikrama and he has a keen interest in uh tech startups so i mean we had to get you on congrats on this announcement as well 75 million you've launched a fund called activate introduce

Aakrit Vaish (Activate): the fund uh to our listeners activate is india's first homegrown ai fund um the um sorry about that no worries you can start again i'm gonna quit whatsapp all right activate is india's first ai fund over the last decade or so i've you know had this opportunity to play different hats as some of you may know i built this company called haptic which was one of india's first ai companies we built chatbots in 2014 using things like nlp and machine learning and all those techniques and then i've been investing as an angel for a long time and then i had this privilege of going and working on the india ai mission for some time so just felt like various threads of life were constantly pushing me to do something like this which was you know to play a bit more of a horizontal ecosystem role uh more than just a builder honestly uh which is uh you know how can you put together an ecosystem on a platform uh to uh really um for the lack of better words activate ai in india and that's really where the name comes from so i say this uh very often and you know actually i've not this is the first time i'm publicly talking about the fund to anybody so you know it was it was uh you know in a live conversation so uh uh you know um i often say this that you know for me or for all of us that activate right the the idea is to put ai in india on the forefront with the fund as the business model for it um so it really never was the starting point was never really let's go build another vc fund and you mentioned that activate

Utsav Somani: has to be india's own native ai institution so tell us a little bit more about this institution play apart from being uh just a venture fund yeah a great point man so you know if you think about

Aakrit Vaish (Activate): um how ecosystems but in ai have gotten built in every country uh they have been centered around some flagship institution in the valley for the most part it has been open ai and now of course there's a lot of open ai was first to it anthropic came out of x open ai folks um you know then you know just everything accelerated from there um you know in china to some degree it's been deep but even predating that there's been a lot of different ai companies uh that sort of build stuff around it um you know uh whereas the traditional mind might say that look how can a fund play a similar role that's what i'm going after which is can activate be the sort of institution of the platform around which you know we can again activate ai in india um and really rally um our young people our founders our technical builders uh to build amazing companies and i have enough of a reason to feel like it's a platform like this that's required

Srijan Mahajan: uh to just sort of put everything around it so a lot of your uh folks backing the fund right and a lot of the lps are from all over the world and we have seen very renowned names names in this and there are a lot of them are from the valley and from here how does this all come together in supporting in becoming this india's ai institution right yeah man so you picked

Utsav Somani: on a very important aspect and some amazing names in that like i mean perplexity founder armin and then like i mean of course all your friends from bombay as well in the indian startup ecosystem insane i mean what a solid roster of lps yeah actually that's a feature so that goes into

Aakrit Vaish (Activate): the platform story right so i was very clear that uh you know this is a um a fee the lp base has to be a feature of the product for the lack of better words right uh and all of them have to be significantly value additive to what we are trying to do so that's how we just sort of went about it right which is that look and everybody actually who's backing us is uh thinking of this very much as a sure you know we may make money returns will come and all that but it's all aligned to that same mission which is that look um can we can we do something about ai in india and can we put it at the at the forefront so uh my only ask from all of our uh lps has been consistently your time more than your money which is that um you know the money is more a way to just start the relationship but uh you know can you engage with the young builders in india can you engage with the founders can we do stuff together with them if we find something interesting can you validate it for us or can you sort of help us think through it uh and so on and activate i mean where do you

Utsav Somani: intend on playing in the ai stack i mean huge capex required for building out the models and compute and all of those things right and there are infra tools like so many different things that you can do in the ai stack do you have a particular focus and check size uh that you're looking to do

Aakrit Vaish (Activate): yeah uh look man for me i'm very very clear this is an ai in india plane um and with that i mean the with that i mean that um we are absolutely not doing what a lot of other vcs are which i think is not a bad strategy but it's their strategy which is sort of betting on indian founders wherever and particularly betting on indian founders sitting in the u.s right i think it's a strategy that i think can work very well we are saying that look we are wholly and solely focused on making things work here so our thesis also stems from that and the areas that we are focused on so first is anything and everything consumer ai for india so we believe that all of the different verticals are going to have a transformation which is going to be ai led healthcare education commerce entertainment finance uh are sort of five things that we think about a lot but there's going to be other consumer verticals which i think we'll have a homegrown completely new native ai experience native ai i'm sure you guys will appreciate this doesn't always necessarily mean a chatbot a native ai just means that use the data and models and intelligence very well to provide a differentiated experience so that's one big thing the second is ai led services um this is a for the world play but i truly believe and this is coming from my days in the government as well i truly believe we as a

Utsav Somani: country should own this space this is to build on infosys and tcs of the next era that's right

Aakrit Vaish (Activate): infosys tcs genpak persistent just think about like all domains it services in fact i think is the hardest one to do and i think it services actually i'm not very bullish on personally as it services 2.0 but i think there are all other verticals think about all the kpos right which have been doing work in document processing accounting legal ops travel ops all of that workflow all of that tribal knowledge sits here like you know sorry just sidebar but related whenever people ask right what is happening in ai in india and then the conversation is where's the llm or where's the compute my standard reaction is listen as a country we've always had our own lean we've always had our own game when it comes to technology waves right that's how all these innovative tech got built whether it is upi or quick commerce so similarly i think our lane needs to be these i truly believe that one of the chat gpt moment for india will either be an ai tutor or ai doctor like i genuinely believe that's going to be the massive unlock um so that's our lane right our lane is go bpo 2.0 gpt5 has shown us gemini 3 has shown us that yes models are getting better but you're not replacing entirely 100 humans anytime soon that creates such a large opportunity for us to become the human in the loop uh part uh sort of center of the world and then finally the third theme is uh what i like to think of as sovereign ai um so you know every government in the world including india uh is now playing a more active role uh in helping ai companies right they're not simply playing an enabling role they are playing a producer role um so we think there's going to be opportunities within that set where government either becomes an anchor customer or an anchor backer or something um and those i think are also

Utsav Somani: going to be very interesting opportunities and your check size and uh the check size that you

Aakrit Vaish (Activate): look to do from the fund yeah the check size 500k to 3 million um we want to be first money in doesn't matter uh precede seed inception incubation all of that i think sure uh it's just we want to be first money in and whatever we think is the right amount of money required to get the company

Srijan Mahajan: off the ground sorry shreejan i interrupted you right i had one question other than this is something that you were talking about i think before before also when i was discussing this your sort of conviction is a very contrarian bet that you are taking into india right because we have seen a lot of funds and as you said there's a right or wrong answer right there are indian funds that are diverting a lot of money into ai in the u.s or abroad um and within india a lot more people are actually investing in other stuff like uh not ai so how does this how do you sort think about this and more so like why this this thing like apart from obviously like that you just want to do it but you can just do things uh but why do you want to uh like why do you see this needs to be done and why is it inevitable that it needs to be done because there is this uh

Aakrit Vaish (Activate): what you explained to us the first part is basically the entire gap that's been left behind oh because everyone's looking uh to the left so actually candidly like i mean i never thought i'll become a full-time vc i think you know so maybe you would appreciate that from just some conversations you had in the past right i always thought of myself as a product builder but i think this is the most in-demand product that i can see from my lens which is that anybody who's thinking about building for india what's the right set of ecosystem what's the right set of support system what's the right capital what do you need as a platform to help them succeed um and you know i mean if you just look very first principles it's inevitable this will happen right i just i just have to be careful that i got the timing right outside of that it's inevitable that all of these things will happen we will have our own ai consumer products we will have these companies india will be a large enough market to create large enough outcomes um you know it's just that the timing for the other funds and the size and scale of those other funds doesn't probably fit the you know thesis set today and me starting out today gives me like you know enough of

Utsav Somani: a time to sort of build up to it and i think size the fund right where you can get into early opportunity and still have a sizable outcome the fund as well and here i think i mean you'll be collaborative with other funds as well but you've made i think a solid entry but being the first year fund but do you ever feel keep follow on capital for ai companies the future vcs might say you'll shift to the sf market or go to us and then build from there and we might support us uh

Aakrit Vaish (Activate): you think conversations like this might come up you know depends what we end up investing in so if we do the things that i was telling you about the gnome right because those are the spaces that you'll have to be very close to the market here uh maybe in the ai services aspect i think eventually founders will have to you know relocate and be based wherever their customers are you know and funnily i was talking i've been i've been chatting with everybody in bpo 1.0 for the last six months um and you know uh a lot of their initial beachhead markets when they were building was actually not the u.s you know they had australia as initial customers they had you know europe as initial customers they had japan as initial customers because that would give them more case studies before they enter the u.s so we don't know right we'll sort of see where this all ends up but i'll add a second point of serve which is that i have never seen more demand for follow on capital in india what i mean by that is that um you know generally if an ai company from india and this i'm not even talking about activate if you look at even what the other funds have been doing if an ai company from india starts working there is a lot of demand sitting at the stages wanting to you know go after it um so that actually at least from a next couple of years perspective is just you know a lot whether they are whether they ask to move to the u.s sit in india all that is secondary but there's just a lot of demand for people saying you show me

Utsav Somani: something that works in india and you know i'm available and so what are the some interesting companies in the airspace that you've seen like what i mean you don't have to mention the companies

Aakrit Vaish (Activate): but interesting applications that you've seen yeah i mean let's go down my theme my themes like one by one so on the consumer side um i don't know whether they'll work or not but i've seen too many ai astrologers you know i've never been a superstitious person or believe in astrology so much but one of these actually i liked and the reason and the reason is because you're not you know in astrology the problem is i'm not sure if the human astrologer is speaking out of your genuine interest or instead you're hooked um you know with ai it felt like look you know feels like yeah i mean he's probably looking at a bunch of scientific scriptures and returning back some answer to me about how my day is going to be like um so i think those are some interesting ones we've seen there's a lot of stuff happening obviously on you know sort of borderline language

Utsav Somani: learning speech learning you know and even ai companions right i mean like just people i mean loneliness is going to be increasingly i mean it's going to become a big trend like i think in our world of like social media next gen is having problems and stuff and they might feel comfortable talking to a chatbot maybe or even a voice assistant to just i mean hash it out

Aakrit Vaish (Activate): versus say going to therapy yeah so there's a bunch of new ai uh companions um and um i don't feel too strongly about that opportunity personally so even as an individual investor before setting up the fund i have invested in nothing on the companion side um because i have a different take i feel like uh they err on the borderline of help versus waste time and i tend to feel it's going to become more of the latter um and when that happens it just creates a whole set of new issues and then also business models are questionable um so i hope what one of those works out in the way that you were saying it ends up being a true sort of companion for mental health and a friend but at least my anecdotal evidence of using all of these and seeing the data doesn't feel like that's going to be the case um in ai-led services you know there's a couple of companies so uh one is there's a lot of data labeling and annotation work happening from india so think of a scale more core sort of 2.0 built from india so i think there's a lot of opportunity there um you know a couple of these bpo 2.0 companies one in accounting one in legal ops uh one many in it services actually so many guys trying to go after it services 2.0 i think all of them are interesting and then the interesting part about those businesses is they actually by design should not need too much money start generating revenue and profit right away uh and then the capital should be mostly for just building out the ai agent or you know growth eventually when you need it so i think that's an interesting characteristic about those uh and then you know one other interesting company public um is uh um you know uh agrani labs uh the reason i only like to call them out is they are trying to build an ai chip oh wow from india something like a gpu tpu yeah basically taking on nvidia crazy yeah so they are based in bangalore and xamd team and they've raised a decent amount of cash from tier 1 vcs again before our time um and uh more power to them uh you know to see sort of how far they can take it so and that's the sovereign ai play right so now that's a play that for example we want the government to back underwrite support create policies for

Utsav Somani: because if they're able to do that then it's a homegrown chip nice and uh perfect so what else

Aakrit Vaish (Activate): is next for you uh this is it man i think this is now life uh you know when you sign up and you're

Utsav Somani: doing this with uh pratyush right from together fund that's right so pratyush uh and i have known

Aakrit Vaish (Activate): each other for six seven years um you know he actually interned for me six years back wow um and then we went together and then life comes a full circle for us so super excited to have him uh he's uh somebody said on twitter which was i think the right way to describe it which is he's uh excited to see deep nerd gps uh and i think that fits his description well um i think compliment each other well and uh he's in bangalore i'm never going to be bombay as all of you know so uh i think that also allows us to sort of you know be in both places and yeah just now next for us is uh you know um we haven't actually uh formally spoken about the amount raised so we have to get all of that in place do a final close and and so on right now we just talked about what we're going to be doing so over the next couple of months is wrapping all of that up um you know starting to invest so we're open in that sense to have conversations and and work with founders

Utsav Somani: and then you know um i saw him tweeted that uh he's helping you spend the money uh he used the

Aakrit Vaish (Activate): word spend and so you've done something with light speed yeah yeah i mean uh let me say i can't comment on comment on that but a month a month decided to so uh yeah so we're obviously open to doing business with others and also again i think where we sit in what we're trying to do it's great to see people bringing us in because they see the value that you know as an ai fund we can provide with a set of things we have so uh and then you know for me like a lot of my mind space or everybody's mind space these days is uh we have the global ai summit in delhi in february um feb 1920 is the first time india is hosting it you know it's going to be done by pm modi everybody's coming down all the big names so that's a big milestone just for all the names coming there oh you know i can't uh confirm but at least uh at least i've at least i've heard uh that i think uh jensen is uh fairly certain to be here uh and uh i think then you know sam and are you and others fairly likely they will also they will also come so it will be quite the gathering so yeah so for all of us uh especially back to my spiel and mission about you know the ai platform for india that becomes a fairly significant milestone to just work backwards from there's also be a ton of announcements around that's time from government startups we'll hopefully do a few things amazing are you still i mean involved

Utsav Somani: as an advisor to the india air mission or not yet focus is full-time activate no can't uh

Aakrit Vaish (Activate): in government you can't do uh you can't do these things obviously right but um so you know we wrapped that official engagement up a few months back but yeah still uh you know help out wherever they

Utsav Somani: need me or call me perfect akrit all the best hope to see you very very soon and congrats again on all the milestones and all the best for what lies ahead thanks thanks bye bye bye all right listeners thank you so much for tuning in have a wonderful weekend lots for us to think about and finally india has its own first ai fund as well and thanks to akrit for coming on the show and chatting about it with us and thank you shrejan bhai uh for doing this with me uh i'll be back again soon anytime see you sir all right listeners on monday bye