Episode 41
How To Build A Sleep Tech Startup - w/ Supreeth

Can you build a startup while you are still on an F-1 visa? Supreeth Reddy, co-founder of Serene AI, talks through the sleep-tech problem he chose, how he found mentors, what it took to raise support, and the legal and practical mess of building as an international student.
Who this is for
- You are building something real and need the unpolished version before the launch post.
- You would rather hear Supreeth's version while the mess is still fresh than get another polished hindsight sermon.
Key takeaways
- Build A Sleep Tech Startup - w/ Supreeth
- Can you build a startup while you are still on an F-1 visa?
- If you want the real version of founder life before the glossy headlines, start here.
- What we do is we build this personalized AI for sleep. Did sandwiches let you use their machine for free? No one is...
- When you've went out to build this app or you didn't really care about any of this legal stuff, right? When we did...
- If you were doing all of this in India, it might have been a bit easier. We took care of the flutter. I took care of...
- All the LLM got railing it because people are like, you're keeping this in our room. What was the amount of the funding...
Transcript
The full conversation, right here. Auto-captions, lightly cleaned, still very much a real human conversation.
What we do is we build this personalized AI for sleep. Did sandwiches let you use their machine for free? No one is going to invest in a team with just two techies, right? You need sleep expert. When you've went out to build this app or you didn't really care about any of this legal stuff, right? When we did customer feedback, we were like, what changed did you see? So we mended the product that way. If you were doing all of this in India, it might have been a bit easier. We took care of the flutter. I took care of the machine learning model. The hard part was creating our own architecture.
All the LLM got railing it because people are like, you're keeping this in our room. What was the amount of the funding that you see? So I can't share that because... Welcome to the Ready Set Do podcast, where we learn from journeys of not experts who are just two steps ahead of us. I'm Naman Pandey and in this episode, featured not expert is Supreet Reddy. Supreet is the co-founder of Serene AI, which is a sleep tech startup. And more on what Serene AI does in just a bit.
For today's discussion, we go back and forth with two themes that are very tightly coupled. The first one being how Supreet built his startup and secured investment for it. And the second that he did this while being a master student on an F1 visa. For those uninitiated, the F1 or student visa in the US isn't exactly the ideal visa type to start a business. We do go over the exact problem that Serene AI seeks to solve, the tech behind it and Supreet's journey through finding mentorship, shipping the first version, building in stealth mode, and collaborating with lawyers and regulatory bodies and think tanks in Silicon Valley.
To finally go public, which hasn't happened yet, but is upcoming. Any person who is interested in starting a company will find immense value in this episode. In keeping with our theme of learning from somebody that's just two steps ahead of us, instead of an expert, my goal is to highlight Supreet's journey not just for the entrepreneurs, but also for any student visa holders who feel that their visa status might be a detriment to them starting a company.
Because as Supreet and many others have shown, it's really not. This is a ready set to podcast and to support it, please subscribe to me on YouTube and leave me up to a five star rating on Spotify or any podcast app of your choice. There are also links in the description below to support me more directly. And now my friends, without any further ado, here's Supreet. -Supreet, welcome. -Thanks for accepting me for this podcast as a guest.
It's my pleasure. I'm so happy to have you here to talk about your startup and other things, other related things. So let's kick off there. What is your startup? What does it do? And yeah, I'll let you give me a pitch here. It doesn't have to be an elevated pitch, though. So yeah, I mean, my startup is about isn't a wellness industry. First, we tried with a lot of sleep talking coaches. We'll build that because we are techies at the end of the day, right?
Then I got this amazing mentor called Curtis Carlson, which, who I actually told us how to empathize with our customers, who are exactly our customers. So the startup kind of molded into a sleep tech slash wellness step. So what we do is we build this personalized AI for sleep. All you need to do is journal with us. Like we have a veterans as our customer or district kids as our customer. So basically based on the journal, the AI gets to know about you more and recommends you personalize sleep patterns, which are good for you.
So we focus on sleep hygiene. We have hardware as a software. So yeah, the goal is to sleep to a very healthy disconnection from the world. Because we are targeting. So, yeah, as somebody that's personally very, I mean, pretty deeply interested in this space, I learned all about sleep on Matt Walker's podcast with Andrew Huberman and other places, you know, I think Matt Walker single handedly can be credited with in our, at least in, in these times.
Really just bringing a lot of attention to how sleep works and why it's important and why it's the single most important metric. One should be optimizing for it. Just as far as overall health goes, right? Like if you are sleeping eight hours every night and if that does your sleep quality is healthy and it's good, that pretty much guarantees that your overall health is good. So that's how important that is. So, yeah, super interested. I also agree.
Sorry for, yeah, I think it's like, it's like, it's like, it's a demon in the shadows. You know what I mean? We are like so used to sleep. It's like for, you know, when you're hungry and you eat, right? So something like that, you know, you shouldn't be able to sleep only when you're drowsy or you're tired. You should create this battle. So it's kind of built into this demon in the shadows. Well, no one is like, okay, I need to sleep at nine and wake up at six.
They don't only until their motivation lasts for some time. So no habits are created. So you fall into the same loop again, you know, what are you taking a lot of information load at night? You know how much information we consume at night, right? All of us in social media, infinite scrolling, all of that. So the goal was to, you know, create this path where you don't feel bored to go to sleep. Rather you enjoy going to sleep by talking to someone, by playing the sleepy games, by scrolling to sleep or like some people use perfumes for sleep.
So the hardware was built with an idea around that with like modularity or sending out perfume before you sleep or showing them some dreamy journals before they sleep. Yeah, interesting. So we are very excited to, you know, deep dive into all of that. But first is taking a step back. I'm more interested in what you started with specifically like what was your good, like, you know, the idea that you started off with before you pivoted to, as you said, after you met your mentor, it sounded like you went, you just pivoted just a little bit.
But I'm trying to understand before that pivot happened, what exactly you were seeking to build when you first started. Yeah. So I'll tell you this interesting story, right? So me and my co-founder were working in lab all day all night. We were doing this R.A. in my university, WPI. And then we both just to try to, we tried, we tried living together in summer, then we were like, after working so hard, why are we not feeling tired to go to sleep and like a zp, like a snap.
So then we were like talking to each other before going to sleep. That's when we realized, hey, man, not everyone has a partner or not everyone has a guy who or something which you can communicate through or exchange information before going to sleep or which has a compound effect the next day. So we are like anyone who is living alone, anyone who is living individually, working by themselves. They need something or someone to exchange that information, dissociate from all the information load from morning to night and go to a healthy sleep.
So we're like, let's make this AI, right? Let's make this AI with just talks to you. It's just going on asking you questions and gives you some ASMR music, let's say, and binds that in between changes your mood based on that. All of that we discovered slowly, slowly, and then that's when we built our MVP, right? MVP is like most of the people are like so pious, since most of the time, all of us are techies in these kinds of industry spread. We are so biased in creating these solutions for people. We're like, what can we do better?
Instead of asking the customer itself, instead of targeting what problem you're actually fixing. What the thing is the empathy can only practice with the problem because, you know, how pain and pleasure could create this empathy in human mind. You know, to exchange information such a way that both of them are benefited. So kind of with each of our customers, when we did customer feedback, pre-interview and post-interview after doing the product, we were like, Hey man, what changed did you see? So we mended the product that way.
So you will answer what exactly was it MVP was it a phone app or was it a hardware? What exactly was the MVP? So it was a basic software application with just journaling and like kind of sleep coaching. Like it just gives you recommendations when to have coffee before you sleep, something like that. Okay, so just like a web app and I'm trying to drill down only exactly how you present it. Yeah, like a mobile app. Yeah, it's a mobile app with no revenue, no passive income.
Because you know how I'm still talking about the MVP, by the way, not resis. So maybe then the good part is the university supports you, right? It's hard to get IRB site. It's hard to go to the place where you can pitch for human or beta tests, human subjects. So the institute offers you something called IRB, like institutional review board, which you need to pitch to them, expand the workflow of your entire app without revealing the secrets.
And then they'll be like, okay, we approve it because it's harmless. They go through all the regulations. And we got the IRB and then we started doing the experiments. Okay, so just to make sure I'm tracking so far, you had your MVP, which was really just a bundle of some different features that you had put in. So be it about when they should consume coffee have, have be able to have a conversation with this app. So on so forth, right? Am I missing something there? No, no, that's that's exactly how simple it was.
We tried to make it simple because sleep is a very personalized problem. So in US consumer market, right, when we enter with a lot of solutions, you can't actually capture the niche market because the key players, at least they're solving some part of it. Let's say we create any open air can do some part of it, right at the start. So you need to create this niche fields. So for that customer empathy was really important, which made us designs. And who built the app was it just you two? Or did you have like a team for that? Or how did you go out building the app? Yeah, so my co-founder is from CS. I'm a data science grad. So we both mended together. He took care of the flatter, all of that. I took care of the machine learning model.
Then we built this architecture, it was way before land chain or anything. So the hard part was creating our own architecture going through going through all the LLM guard railing it because people are like, Hey man, you're keeping this in our room. I don't feel secure because humans can't accept AI as a species right now. We are like too early for that, right? So so we need to create a product such a way that it isn't it.
It's weightless as well as useful to the customer in their room. They don't feel that they're adding something that they feel like him. It's just a thing to use. So what were some of the like most notable results from the, you know, IRB? Like what did people report? Yeah, so first thing the problem we were targeting about was the loop, right? The feedback loop, the loop in which people are going getting back to having more that synthetic dopamine hit of that information load again and again, right? So we are like first we need to validate the problem. Is it really existing?
Or are people going through these phases at very less or very rarely or the frequency of a car. So you're like, how many times did you go through this in a week that you are not able to sleep well because you have been going through your phone or trying to get more information because you're not yet tired. Something like some of the questions pre-interview were very direct, focusing on the problem statement, which validated it.
Then we are like, we selected some part of them, like out of 200, we selected 40 people first. And then we were like, okay, use our data because you like journaling too, as well as you want to sleep well, cutting the information. So then they used to have, they got back to us, explained us, Hey man, this is what I wanted different. This is how I wanted the AI to talk such a way that I don't feel that it's, you know, literally stalking me. You know how the AI banes like how people are.
So if you're going in a consumer market, you're taking the risk of, you know, not keeping your AI secure out there because people are like, Hey man, you're taking my information. It's in my bedroom. So, so a lot of empathy, a lot of, you know, talking is what made our app weightless, which I think that was the first key deliverable for my MVP before any money or before any revenue before anything. Interesting. So you were. So if I were to understand this correctly, the two biggest features of this app were the ability for users to journal.
And the second would be not, and this is not in order of importance, but so just generally the two are the first that I just said, and the second would be to break the feedback loop that of dopamine that comes when you scroll Instagram and then replace that with, you know, just having a conversation with an AI. So was this verbal or were they typing that stuff in? It's verbal. Everything speaks to the journal verbal is well, or do they have to type that?
No, the journal is verbal too. It's a bi-directional conversational journaling. Interesting. Okay. Because that, yeah, I was wondering because you know how the, I'm sure, you know, the effect that blue light has on our psyche, especially around the time we're trying to sleep, it's just not helpful. So if, if a user were to have, you know, a phone right close to their eyes, they're wildly tight. That's probably counterintuitive, but that makes sense why you did it in a conversational way. So they don't need to be looking at a screen. I'm sure that's something you've considered.
Yeah, that's like the principle of disconnecting to people to a healthy sleep. So we are like, Hey, man, it starts with the disconnection itself. And you know what, like, since see this, how the market molded, right? It's amazing. Once you build this little MVP and you don't need the look to look at the phone. So who are the people who are not tech savvy, as in 2024, not tech savvy at the base level to veterans for enjoy the old people who are not able to use work or the autistic kids. So that's what turned out true. That was interesting.
Very cool. So then when you met your mentor, what changes came about? What did he recommend? Because it sounded like he was very instrumental in your journey so far. So what was that like? So he's an amazing guy. First of all, thanks to Dr Curtis Carlson for being your supervisor. So he taught us this framework. It's called an ABC framework. He's been teaching the world around going around the world, teaching that to people. He's also started in India, by the way, using an ABC. He's been connecting workshop.
So an ABC is simple, right? And it's about the need more than anything. It's like an abstract level of finding a problem, like the need for having that people who cannot, the people who need that in their life, something like that. And then comes the approach, how your solution is mending to your need. And then comes the be part. B is very important to actually do a business because it's like benefits over costs. What benefits are you offering to the people based on how much cost they are paying for it?
So basically sleep is a very intuitive problem for most of them to have around the world. So in America, one in three people face the problem of sleep, face the problem of insomnia. They're driving themselves. They're driving the economy towards insomnia. So more sleep products coming out. But the point is the benefits over costs part is where all the key players in my industry, sleep, like industry, like calm, headspace, we all know about them.
So the customers generate. So people are jumping around in the top key players, the customers, because they're not able to find the personalized sleep store. You know what I mean? So headspace is offering just meditation, calm is offering meditation, especially offering sleep stories. So people download this, use it for three months, jump around. So they're jumping around in those key players. So we're like, Hey, man, we can offer a personalized solution and we can get an umbrella out of the market where there are people who needs everything at the same time.
People who wants to take it off or people who wants to grow with the app rather than use the app. So, and the last one, the C part is competition. That's why a lot of the analysis come from here. So interesting. It's like a machine, you know, n n works with a BC and then you reiterate again and again. It's amazing, man. I think it's a magic you guys should write. Everyone should. Yeah, no, for sure. Yeah. This is not something I was aware of, but it really sounds very compelling to, you know, break down what you're trying to build.
In these easily solvable components. And then you can kind of optimize for each of the four pieces to get to the point where you're not trying to get. So that's very interesting. I will point out just just my personal experience with. I don't know if it was calm or what it was, but yeah, maybe we can just do like, you know, like a role playing exercise essentially. So I am somebody that loves tracking their C by do it pretty much every night that I possibly can. The sweet spot. I have found actually before I share all of that.
I want to say that maybe I'm curious to hear your perspective on whether I'm even a prospective customer because most nights, so I will say like 80% of my nights, I have no trouble falling asleep. That's good, right? I'm assuming, but there are definitely 20% of the time where it's a bit of a struggle like I'll just lay there, you know, contemplating life. And just, you know, you know, like just being tormented by not being able to sleep. But however, the way I do it is I just wear my Apple Watch.
And then it sings it, you know, auto calculates sleep data and all of that. And then I use this app. I'm sure you may have heard of it. It's called sleep plus plus. But it does it is just grabs all the data from the Apple Watch. So like my sleep stages and all that. And with that, it clubs what my testing heart rate was what my heart rate variability was. And it must is in all of that information and spits out a score out of 100. That's my sleep score.
So that's all I do right now. I'm just trying to understand. And actually, you never mentioned the name of your app. What is your app called? It's called Serene AI, on Play Store. Perfect. I am with the first thing I'm going to do after this is check it out. But where would you say Serene AI fits in with this type of user base? If at all, this is even applicable. What you have talked about is a quantitative measure of sleep. Like you only check it when you wake up that day.
But you have been through the problem of not sleeping that particular previous night. So you're like, how can I improve? If you're thinking about the same things or like going into that route, the goal is to assess where that route is. Like our goal is to assess your pre nap ritual and post nap ritual rather than how well of a sleep you actually have. So that's where this part we have no connected devices. We can connect Apple help through APIs and all that.
But it's like if we are not calculating how well you sleep, it's about we are calculating how well you can go to sleep as well as how well you can wake up from sleep. So we are like, if you see the personalization factor, see once you sleep your shed on, you have no control like with your conscious part of your brain most of the time. So when you're not before actually when you're able to recognize that, that's what we are targeting that when are you thinking about sleep, not while sleeping right before sleeping or after.
Right. So that's a big sense. So you're just you're not really concerned with what happens the morning after you're more concerned with the pre sleep side of things. But yeah, a brain intuitively makes a lot of sense for sure. Yeah, yeah. So it's like, yeah, go to a healthy sleep with us rather than will tell you how much sleep you're actually having. Because come on, like we have all been through that where the app shows that we have like really good quality eight hours of sleep, but still you're tired in the morning.
Right. I don't know. I feel that because sometimes I'm just lazy in the morning. So for sure. And usually, so that's the reason I like sleep plus is because even when I have slept eight hours, sometimes it'll give me a score of like 23 out of 100. And I'll just be like, Oh man, I wish I hadn't seen that. And the reason usually is that either I, you know, woke up a bunch of times or my heart rate was too high. Or I don't know what I was doing. As you said, it's impossible to control what happens after you've gone to sleep.
But yeah, it's just such an interesting problem to solve. So yeah, I guess jumping back to kind of your journey. At what point did you start, you know, going live or inducting real world users? What was that experience like? So once my co-founder graduated, we moved to California. That's where we were working with SRI. Right. Stanford Research Institute. We were trying to pitch to them. We were trying to build something.
But you still on college at this time, were you still a master student? Yeah, I was a master student, but my co-founder was graduated in May. Okay. So in May, we shifted to Cali for the summer for me. So then we we pitched to SRI Stanford Research Institute. We pitched to mainly the researchers there because we were reading their research papers. We were like, which of these papers reflect with our product right now? What can they promise us?
Because no one is going to invest in a team with just two techies, right? You need sleep experts. You're not good. You're not in a good board room. That's why it accepts you. Yeah. So in Cali, we luckily, we had this NGO like a veteran building right next to our gated community. And we are like him. There was an old guy standing there and we just went there and we are offering this. Would you like to take this? We started like this, right? Like we have problems with like just a cold pitch. Just an in person cold pitch. Yeah.
That's how you know. That's how you know exactly what the market thinks about it. It might be my bias in doing business. But I feel that you actually know what the market is thinking about you without knowing who you actually are. I think that's where the magic is because you will actually know how ready are people to accept you when you put it out there. So then the veteran suggested to his friends and then we targeted NGOs. We pitched them. We regulated the API a little for their use.
And then yeah, then it took up from that. And then where you did the Stanford meets at all feel or prove fruitful or not. Yeah, because they have polysomnography machine, right? So this machine is crazy because it actually very accurately, quantitatively measures your sleep. So the goal was to measure a compound effect of if they're using our app for 21 days, if they're habituated to using our app for 21 days. Will they have? Is there a change in their sleep pattern or is there a change in their happiness or feedback with respect to when they wake up in the morning?
So we were able to measure this quantitative measures because those matter end of the day because sleep. Most of the experts who are like judging the sleep industry are like, are like research scientists, sleep experts, sleep psychologists, cognitive psychologists who actually want an evidence based on the corpus site. You can't be like, this is giving a reward up. So this works well, or this is not giving a reward. This hypothesis doesn't work well. So we created nears like that.
I'm very interested in how you went about this research phase because it sounds like it's a lot of effort. It's a lot of investment that sandwiches like you use their machine for free or how does that. This is where, so Dr. Curtis Carlson was XCO of SRN. So he made HDTV and Apple City. That's when he got appointed as CEO of SRN. And that's how we got partnerships, right? Because he could get us a meeting in the first place that I saw before us. Wow, took off from.
Okay. And then what were the results? Like did you see any noticeable or any improvement? Basically, when we mended the MVP to veterans and all, we could see that the rate in which the accuracy or rate in which the bidirectional conversion generally, the good part of it, the good feedback is increasing is by 6% every month. So like this is a great compound effect because people are going to pay very less, right? How much are they paying? They're paying just $10 per month or $5 per month because you know how it goes with the first 10,000 customers right most of the time.
So they are like, okay, we are giving the customers the chance to calculate how well they slept in June versus how well they slept in July and August and how it's improving daily or monthly basis. So how were you calculating that? What were some of the metrics you were looking at to get to this 6% increase? So some of the metrics were their app usage and so the app usage and the time they talk to Seren and the more they journal with it, the Seren is able to understand them more.
There's a memory cache which is able to remember about their temporary activities. Let's say, let's I'll explain you this way, okay? Let's say last week you've been with your friends who are swimming pool and you're tired, you're a little exhausted after that, you come to your home, say to Seren that, "Hey man, I'm exhausted today, I don't need Seren today, thanks for..." and you go to have a good night and you go to sleep.
The next week, on the same Thursday, you're like, "I feel a little active, I'm not exhausted." If you're like last week, you went to swimming pool with your friends. Do you mind calling your friends if they're free? Do you mind messaging them? So you're creating this real world connections between people, between their activities, between their psyche, like psyche at the present day. So based on their different modules come out, like let's say you're like, "Now you're like, I don't want to swim today man, come on, I don't want to go out of my home." It's like, do you mind listening to ASMR music now? Do you mind going to listening to sleep stories? Or do you want to scroll for some time and come back to me?
So it reminds them like, "Hey, before sleep, I don't mind if you, I'm always there for you before sleep coming back to me." So, yeah, we have felt that, yeah, there are some people do. VDSD or go through COVID, we are all therapy sessions, right, CBT therapy sessions as a veteran. So they feel like this is a good problem to solve. So that's very interesting. I didn't realize this was also one of the features you offered, but that really opens the door to a lot of new, you know, just new new ground to run on.
So it sounds like it's almost like an AI assistant bundled into kind of a sleep and well-less type feature because when you said it, it'll help you talk to your friend who you were missing maybe. Like, does the app do that for you or does it just encourage you to make the call itself? So the app grows with you, Serene's AI grows with you, and we have this interesting UX part, right? So first time Serene is like, so UX plays a lot of role because you need to actually feel like you're interacting with a human before you go to sleep, not like a machine, because who likes to talk to a machine before? It's a very private space of your sleep, right?
So it's like the character grows up with you, the character grows old with you based on how much it learns with you. So once the character grows old enough, a new character comes out and interacts with you. It's like, and also the... It grows old as in it just dies. Sorry, what do you... It stays somewhere. Let's say it's a plant, it becomes a tree. Let's say it's a cactus, it becomes a bigger character or like something, all those metrics for an end-to-end life development.
Yeah, so it's like, so you know the cactus remembers your part of activities in August, and rather some person who's growing old with you remembers about your journey in May or August. So then he leaves. So there is a whole sleep world which is great and created for them. They're like, what did I do last year in June? Call the cactus man, Serene will call the cactus for you. So something like that. So it's like a very based, yeah.
So I definitely appreciate the need for having this data stored so that it can call back to certain experiences you had maybe in the last year or the year before that or something. I'm not entirely sure I'm following the need for these various type of plant types. I'm sure there's a reason. Can you elaborate on that just a little bit? So let's say it comes from the fact that what kind of things or characters do you want in your bedroom or near your private space, right?
People have a very close people have plants, people have fishes. So this is targeted to the things which you see it as a normal in the society. Like people are keeping, keeping aquariums in their room bedroom or people are keeping reptiles in their bedroom if they want something like that. It comes from the fact of acceptance to keep it in your allow it in your private space and then you got it and do all of them remember everything or do I need to remember that the fish remembers stuff from last year.
So you don't need to remember that the fish is fish toaster data in May or your birthday month. It's like Serene always tries to get more characters involved. It's like Serene first talks to you and then it's like if it is able to assist your mood for that particular day, it will call your old friend, the character who you like much. Interesting. So I see so you're almost the kind of doing some sort of reinforcement learning here where it's one night I was able to sleep really well with my snake, then it'll keep bringing back the snake just to have signal to my brain that okay snakes here it is time for me to.
Is it something like that? It's a mood page dividends. Like if your mood reflects around that cluster that it will call the cluster. It might call the second rank because you don't want the guide to feel monotonous right like when you're dealing with an AI you're like come on man. This calling snake all the time. So you're like so interesting. Yeah, this is so novel. I've never heard of this concept at least definitely never heard it being implemented in a you know user facing app.
So I guess my next question is yeah, this sounds from an engineering point of view fairly complex. What was it like to build out all of these features just like the two of you. It's like 100 hours per week every day for three months. It was crazy man. I was doing a and my co founder was doing a with another lab. Just to get some initial funding to pay for the credits right API switch use all of the things to bind in and we still were able to try to fit in time of like 100 hours 90 to 120 hours per week like crazy.
And then then when we met our advisor a supervisor doctor Kurt we pitched him. It was just a four minute pitch. He called us to California. He paid everything all expenses paid and just to pitch right he likes WPI that much. Sorry, sorry, sorry for that. Yeah, I didn't even see what was happening. Sorry. Yeah, never mind you saying he like WPI for. Yeah, so then, then once he liked us. He made us, you know, like he made a part of his community. There was this value creation forum going on.
It's an open forum for anyone to join. So he made us join that community community. He made us iterate the product better. Then in a month there was TechCrunch TechCrunch disrupt 2023. That's when we were like man, girl's magic came out because we talked to so many investors. You know how harsh the investor world is right. You know how harsh they are like reject you on. No, actually I have no, I mean, I've heard stories, but it would be great if you can share it.
Yeah, yeah, it's like crazy harsh because end of the day they are spending their hard on money for you. Right. You know how the money culture is around the world when it's come my people, the relations needs to be more stronger. So most of the investors at pre seat stage, which we were at. They invest in people more than the ideation part of it. They invest in people. They invest in problem. Just do just two of these things.
Everything else is a complex part of how their firm is going to interact with them later. How the due diligence is going on. All of that is like, it's just like a business part of it, the agreements part of it, the due diligence part of it. The legalities of it. But when it comes to investing as a person, you should be ready to pitch anyone in the world, right? Whoever reflects with your problem or sleep. Luckily, most of them are sleep deprived in the world.
So we were able to pitch to host of them and got rejections at the start a lot. And then this guy, Femi Moito from empty three capital, he gave us a call back and it was. So how does that part of the whole thing work? Maybe there's stuff that you're not allowed to talk about. So I respect that. I mean, I can say, yeah, sorry. So yeah, how does the negotiation work? Like how much money, how much equity is it for? Is it just like Shark Tank? Because that's what I mean.
Yeah, it is Shark Tank. But you know, what's the intense part? It's like a one on one session with Shark Tank with the Sharks. So you're like, they'll ask you literally anything. It's not on media. So they ask you anything. Right. You're your anecdote. It's about to know actually who you are. So we had like five six interviews with empty three capital with Femi with our lawyer just for them to meet up to partner. And then, yeah, that you did.
Then comes the due diligence part due diligence is one of the hardest phase for the for early startups to, you know, to cross over because of the part that it involves the 17 different factors of legalities of who your name is, what kind of patents you have. How does the market fit? How's the product market fit? All of that. How's your team? How are you as low getting resources? How's the finance? How's the expenses?
So there are the 17 things which so it took a month. It took a month for them to go through actually research about our company, re validate our partner quadrants, have a competitive analysis, and then they get what back to us one day and they're like. So you mentioned there was a lawyer there who pays for the lawyer. I hear lawyers are reducing. Yeah. Oh, so the thing is, we got a connection. So you don't have sport, right? Yeah, we know we had a.
So Jim O'Neill is one of the founders of HubSpot. We luckily got to dinner with him and he became our friend. He suggested his lawyer when he started HubSpot at the start. So Steve Kagnata. So hey, Steve. So Steve was our lawyer. Is our lawyer still? And yeah, he worked for us. He incorporated the company. He took care of the tax structures, the agreements. I mean, yeah, I guess I have a bunch of questions around the legal side of it, but it feels kind of boring because I know you need to be incorporated in certain states.
Every state has its own separate rule. So I'm sure you would have had to look into all of that. So I guess overall, I'm just trying to understand how did you navigate all of this? You know, like when you've went out to build this app, you won't really care about or you didn't really care about any of this legal stuff, right? It must have been such a hassle. So I'm just curious about how you navigated this whole part. It's like once you got this once we got this lawyer, right? As a student, you have this advantage.
Even the professional environment as a student accepts you wants to teach you things. So Steve came on, so he didn't ask us any money at the start. He's like, Hey, man, I'll help you do this. Then create this quarter cycle for me or create this kind of paying as a bit by bit thing for me because you're student. He knows that you don't have money. So yeah, that's how the relationship between us and our immigration lawyer fell into the picture and this guy fell into the picture and that's what makes it amazing as a student.
We have this power over using resources more efficiently because the depth which you will be going as a professional is more since because you're all by yourself. Here, your university will provide you help. There are professors who provide you help. Yeah, all the networks. So your university's network are yours as a student. Once you go out, your networks are your so research. That's very well said, actually. And yeah, sorry, we got distracted. But how much like what was the amount of the funding that you see received that you say.
So I can't share that because I'm still in stealth. But yeah, I can say we got an evaluation. We shared our equity for 17% because see, I'll tell you, this is amazing by us with with early stage founders. So this early stage founders, since they have built everything from scratch, they have attracted like obsessed to the solution or problem anything that so much that they forget, they think that their equity is like a diamond, a gem.
But no, it's not true because you need to be ready to give out equity at the start because 0 100% of zero is nothing compared to 5% of a million or 10 billion. So people forget that because they're so attached to the initial part of it. So when I say 17% for some amount, people are like, man, you gave out 17% of your startup to a random stranger, you'd run and bootstrap it, get it to a better stage, like show a better customer base and then try to raise that. But it depends on your industry. We needed research equipment. We needed hardware. We wanted to attack the industry in different ways.
So it's like you either burn at the start or then build up in the future. And another thing you said kind of caught my attention, you said you're in stealth mode exactly. What does that mean for a founder? So it's like, so see, as an immigrant, see, we haven't got the hiring and firing rights here in our this thing. And we also can't earn passive income as an immigrant, right? Because we are waiting. So our supervisor is the one who is our financial, who is our mentor, who is our recruiting manager, something like that at the start until we get appointed as for a full-time position.
Got it. So when we are giving out our app out there, the government, we need to apply for something called Dune's number. So it's not easy to get the Dune's number to put the app out there. Once the Dune's number is there, then we can start marketing on social media on whatever about sharing. So we're just waiting for that. Once we get that, then as an immigrant, it's like legally allowed to get through. It's the government saying that, "Hey, man, we are okay with your financial pipeline through this app. You can start earning some money and doing some marketing. You can be out there."
Despite being a not a citizen of the United States. Yeah, yeah, yeah. All you need the first hard part is get a supervisor. If you're thinking of startup, get a supervisor who is an American citizen. I mean, there are many people who are ready to advise. That will be surprising because once they connect with your idea, right? The people are ready to advise. And like all day all night, man, I'm in with you. That's a great part about USA, like a lot of opportunities for entrepreneurs at the start.
Makes sense. And in your case, or you would just generally recommend for students at least because most of most Indians here are at least when they start of their students. So it sounds like you're recommending that we just, you know, they go to their departments wherever they're at. Look at people, talk to their whoever is advising them from an academic point of view and then pitch to them in terms of also being an advisor in terms of, you know, this whole app or whatever business stuff from that POV.
Yes. Most of the time people should be, you ask, you get an answer, man. You ask 100 people, you get, you'll get at least one answer. You should be able to do that. You should be able to take notes, which most of us practice at the end of March. That's why we're applying for full times in this current job market. You know how it is, right? So, yeah, they're ready to accept knows and then move forward. Yeah. For sure. And yeah, it's and also just to quickly close the loop on the Dunes number. What I'm hearing is if I, well, not me, I guess, or maybe me, right? You said it has just has to be an immigrant.
So if I were to start a company or if I were to want to start a company, I would find somebody that's willing. That's a citizen that's willing to be my advisor. And then they would have to do people work file. What exactly for me to get that Dunes number? Like, what's that process? So that process is like, you need some testers, you need to show your organization's hierarchy who is working on the app, which is going public.
All of that you need to send to the play store and to the government organizations, which are involved in sleep industry and then get a permission from them, like a certain reaction. Yeah, that's the hard part because once you create this financial pipeline, right? You can do anything out of it. Like you can, I'll be like, Serene will be nothing. Serene will be free. I'll just do it. I'll support it to one revenue. So different kinds of revenue streams can be added when there is a nap, which is public out there, publicly out there.
So just to cut that out, just to cut, just to create a safe regulatory environment. The government is like, if you have a Dunes number, you're ready to pay taxes. You're in. I guess with the part I'm not entirely understanding is why would you need to go to those like sleep department or whichever regulatory body he takes care of that just because your app claims to help with sleep is that why? Yes. Okay. Because our app is solving the problem of pre-nap ritual of sleep or anything like whatever you're solving, right? You need to get a certification from someone who's actually been there in the industry or as an expert.
That guy comes from the government. That's a big yes. If that guides the guys from a private organization, that should be an esteemed organization to create this good finance part. Interesting. And then when you were navigating all of this, this just sounds like a lot to me. It just sounds like just endless amount of hassle. So did you have a person that was helping with that or were you just literally googling all of this and figuring it out as you went?
No, no, no. So we had a lawyer, right? So we're paying our lawyer. He's crazy, but yeah. But yeah, don't mind to you, but yeah. No, I mean, that's what I said. Yeah, when in my first question, right? As soon as I heard lawyer my years back up, I was like, that's, you know, that sounds expensive, sir. So very cool. And then you said that for your app to go public before that you need to do certain steps. So is Siri not public right now?
No, it's not public. We will literally have a marketing guy who is downloading the app on our customers or customer markets profile. So it's not public. It's like an API base. So if I were to go on absolute and look for it, I wouldn't find it. No, you wouldn't find it. I would send you the API to use and give me. Got it. But you cannot put it on the. That's so interesting. Okay. So that makes a lot of sense because at first, I thought I would just be able to use it villainily, but it sounds like that's not true, which makes sense that you said that.
You need to check those boxes off first, make sure all of your due diligence is done and then go from there. Yes, yes. So, yeah, that's all. So, yes, I guess what lies ahead, though, for you, are you going to be looking for jobs or what's the person so next? This is what, yeah, this is what I've been doing. Like, I'm like, I need to, I'm a data science. I'm a pure data science research kind of wife cut. So I'm like, let's do a job, which is 70 to 80% efficient with whatever tools I know just based on skills.
I'm not saying one job is like easier than the other. I'm just staying based on the number of skills you provide to the company. Let's do a job, which can take only 70% of my skills and not squeeze me to end because I need some energy left for to work on Serena at the end of the night. So I'm like finding job around, yeah, I'm trying to find jobs with like 70% of a data science role, not that role research intensive that I can manage both.
My goal is to do a full time since I have a loan like any one of us. So I have a loan, I need to clear that, do full time. So I have this key deliverable in my mind. So once that key deliverable of the number of customers is set, I shift to full time in Serena. I'll be appointed as a CEO of Serena. What's that number? I can't say that it's company secret, because that's where you're like getting managing partner, all of that.
And you're still referring to the number of users that that's still what we're talking about, right? Yes, number of active users per month. Got it. That's so interesting. And like, yeah, I'm just trying to put myself in your shoes and I thought it just feels like, you know, it just feels such a like you're so close to not have to work and to, you know, focus your all of your energy and all of your time into this thing that you've built from the ground up from scratch. And yet you cannot do that just here. Yeah. Does that make sense? What is that like?
I mean, most of the time, it's like, you know, it's like you have so much, you give so much value to a time when you're working on a particular idea, right? Just that time, a little part of the time is taken away from you. You're filling it with financial reward in that aspect. So you're like, okay, trying to manage your mental health at the same time, like your mindset so that you'll be ready to take and, you know, your own companies work at the night or whenever you have a time space left.
So resource, basically two things, right? I'm practicing resource allocation along the way for my own self because maybe there will be more things for me to deal with in future with respect to setting. So it's preparing me for that. And the second thing is, I don't need to worry about how much salary I'm going to get from my own investment firm. Are they going to give me salary? So, yeah, I don't even think about that. That's the reason you're not guaranteed a salary, right?
Yeah, exactly, because if you take salary, right, you're just an employer to the firm because firm is a huge, like big organization who just employees. They have other investments obviously. So you're saying a boardroom might take a potential threat if they want to. That's what happens with most of them before entering the siege stage because everyone wants to get hold of that equity, wherever they are investing money, all of that.
If you ever wonder if you were doing all of this in India, it might have been a bit easier. So the thing is, this is my second startup. I had my first startup called Brains Language in South Delhi, past year. I started with this photographer, Ashish, and we were trying to build recommendation systems for cab jam. It was service day startup. So in COVID time, we were like, at home, I just resigned my work because I wanted to do something amazing before I turned 25.
So I was like, I looked at risk, right? It was hard, but yeah, I took that risk and started with people in India. Indian consumerism is very tricky, man. Indian consumerism with respect to anything, right? It's very tricky because anyone can take over your market or you need to dance with the market, which happens with a lot of expertise. So when you're taking the risk, even for the first time or second time or as a serial entrepreneur, you need a lot of value already built in the market.
You need a lot of brand already built in the market. You know how it is with podcast, right? A lot of brand value creation in the market and then you take it off from there. So something like that. This is what, by us, I have in Indian market because without traction, no one is going to invest in you in India. But in the US, without traction, just with the proof of the concept, there are a lot of investments happening.
That's very interesting what you said about needing traction to get investment in India. Like, that's the thing, right? That's what kind of trips me and in my very limited understanding of just, you know, starting up and all of that. Obviously, I haven't started up yet, but that is something I plan to do. And a constant, just a tussle in the back of my mind that I struggle with is on the one hand, you hear all of these people Peter Peter here and all of these people say that if you're if you're solving a problem, that's a genuine actual problem.
You will get paying users. It's just how it works. And then on the other hand, you hear, and I do not the first person to say this. I've definitely heard before that, like for Indian consumers specifically, it's just that hurdle of relinquishing their money. It's somehow just way higher. So it's almost as if two contradicting pieces of advice, right? So, yeah, maybe you don't have a solution to this either, but have you ever wondered about this or is it this way?
Yeah, I think the Peter Thiel's way of perspective is about the process in which you're going to reach the paying customers more than the end result of getting the paying customers or like, interesting. You will find at some point of the time, it's like, it's becomes intuitive. And once you empathize with who is going to pay in the room, you're like creating or minting your product in that way. And then the process of thinking about that doesn't need you to be an expert in that field or something like that.
I felt that magic because we were going to an ABC framework, right, coming back to the framework. So that framework enabled this process of thinking about it again and again about that. Especially with the benefits, right? It sounds like benefits would be what exactly attacks this exact problem that we're talking about. The benefits over cost place, a huge role. And the benefits should be greater, greater than cost in India.
That's true. Yes. I mean, I know ideally that should always be the case, but you're 100% right that especially in India. Thank you so much for taking the time today to walk us through. You know, you're such an amazing journey. It just feels really when you were talking about it felt like, you know, just watching Silicon Valley all over again. It's just, you know, and it's just something that I've always, you know, it's just, I'm pretty sure for most entrepreneurial people out there, it's just really the, you know, the dream.
So in a lot of ways, you are living all of our dreams and, but obviously you only have your own hard work to, you know, thank for that because you didn't magically end up in this position. So really excited for what you built and what lies ahead for serenee. I will be looking for that day when it's finally available on the app store so that I could just use it and recommend all my friends to use it. So thank you so so much for taking the time here today.
Thanks for calling me, man. It's amazing to talk to you the way you take over the story, the way you take over the podcast. It's so natural and talking to you. I think you reach great heights, kudos to that respect to you. See, thank you. Appreciate it. That brings us to the end of episode 41 of the Ready Said Do podcast. Thank you all for sharing these conversations with those that continue to benefit from them. If you would like to support me, please subscribe on YouTube right now.
Leave me a profile rating on Spotify or any podcast app of your choice and literally tell every single person or friend that you know about what a great podcast is. That's a little bit cringe, but I've just started saying it just in case that works. Some people have told me that they've been doing that, so there's that. Catch you all in the next one. New episodes, everyone.
Transcript-backed moments
A few lines worth stealing before you hand over the full hour.
What we do is we build this personalized AI for sleep. Did sandwiches let you use their machine for free? No one is going to invest in a team with just two techies, right? You need sleep expert.
When you've went out to build this app or you didn't really care about any of this legal stuff, right? When we did customer feedback, we were like, what changed did you see? So we mended the product that way.
If you were doing all of this in India, it might have been a bit easier. We took care of the flutter. I took care of the machine learning model. The hard part was creating our own architecture.
All the LLM got railing it because people are like, you're keeping this in our room. What was the amount of the funding that you see? So I can't share that because... Welcome to the Ready Set Do podcast, where we learn from journeys of not experts who are just two steps ahead of us.
I'm Naman Pandey and in this episode, featured not expert is Supreet Reddy. Supreet is the co-founder of Serene AI, which is a sleep tech startup. And more on what Serene AI does in just a bit.
Show notes
Can you build a startup while you are still on an F-1 visa? Supreeth Reddy, co-founder of Serene AI, talks through the sleep-tech problem he chose, how he found mentors, what it took to raise support, and the legal and practical mess of building as an international student. If you want the real version of founder life before the glossy headlines, start here.
More in Build It
Same mess. Different guest. Pick the next conversation that feels closest to your real life.
