Tanay Kothari: Building the Next Generation of Personal AI // #63

They're solving the wrong problem. Every single voice transcription service today, they optimize for this metric called word error rate. Of the words that you said, how many of those words do we get perfectly right? And they're like, we're 98%, we're 99%. It doesn't matter. Being 99% good at the words you get right means that in every sentence, about one word is going to be wrong. And that actually is problematic because that means you can't trust a single sentence that comes out of them.

Welcome to Forward Obsessed, where we talk to breakthrough business leaders and rising entrepreneurs about their journeys, origin stories, and aha moments.

Hey everybody, welcome back to another episode of Forward Obsessed. I'm your co-host Pete Senna, and I'm excited to be on this conversation today because Tanay Kothari, he is the founder of an incredible startup called Whisperflow. It's a product that I use every single day and quite literally couldn't live without. So when I had the opportunity to jump on the show with them, I was really, really excited. Unfortunately, my co-host Dave couldn't be here today. We are going digital. We're going remote first. And let me tell you, we started this thing remote. We went to in-person, we're back to remote. And I have to say, it's hard, but I'm super excited. This is a product that has absolutely changed my life in every sense of the way. Tanay Kothari, welcome to the show.

Yeah, fantastic. So I'm Tanay, the CEO and co-founder of Whisperflow. Which in the simplest form is the best voice dictation tool built for professionals. So if you're somebody who has to spend a lot of your day communicating, whether it's writing emails or messaging people on Slack or message, sending a lot of messages to your mom, writing code in Cursor, making docs in Notion, this is a tool for you because I think it's 2025 now and it's about time we stop using our keyboards.

So tell me a little bit about like where you grew up and like how did you get the bug? Cause you've been building apps since you were a kid and I'm just really curious to understand more about your story.

So I grew up in Delhi for most of my life and back when I was, I think, young, I was very into math. And so in school, my teachers let me skip math class and I would just go sit in the computer lab, play games. This one time there were a few seniors who were there and I think they were writing code, building a game or something. And I went to them and I was asking them what they were up to because it seemed very interesting to me. And they said, oh, you're too young, you wouldn't get it. And no one says that to me. So I went home. This is, we're talking about 4th grade. So I was about 9 years old. And that was when I pulled my first all-nighter. Because I wanted to teach myself how to code. So I was sitting there on YouTube and this website Dream and Code, which had a lot of tutorials. And that's when I started my journey. And a year after that was when the first Iron Man movie came out. And to me, that caught my attention because I looked at that, I looked at Jarvis, and I looked at what I was doing on my computer, and I was like, This is just going to be archaic. We need something like Jarvis to, uh, and that is what the future is going to be like. And so this is what I was thinking when I, when I was 10, because that's when the movie just came out and I wanted to build Jarvis. So the first operating system that we built was supposed to be more language-first as well. You were able to just say things to it and things would just happen. Which is what the initial idea for that was. And that has been pretty much a consistent theme in my life ever since.

So tell me a little bit about your parents. Like, you know, like, were you a first-time entrepreneur? Were your parents also entrepreneurs? I want to understand a little bit more about your upbringing and your parents.

So both my parents are pretty entrepreneurial. They are both in finance. So my mom built and runs her own private wealth management fund. My dad at the time was a CFO of a company and then now builds and runs India's largest venture debt fund.

So where would you say that, like, did it come more from your mom or from your dad? Like, what would you say, like, who was more inspiring for you?

Both of them, actually. My mom is, with her, it was like she was building this new company from scratch while taking care of me and my younger brother. And so I had a just a lot of empathy for her. And my dad, from then till today, just works 18 hours a day. I had to stay up at nights to write code because otherwise it would get counted in screen time that I would have only an hour of every day. And I used to watch cartoon and that. So I would wait till my parents went to bed at about 10 PM. Then I would call my buddy up and then we'd both get on the computer. We would start building things. And then at about 5:30 in the morning, when my mom would be coming to wake me up, I would jump back into bed, pretend to be asleep. And I think for maybe all of middle school and high school, I just slept out in the nights.

So what was the first coding language that you started coding in? I'm also a coder myself, and I always have to ask.

Visual Basic.

So I'm really curious to understand, like, you know, you grew up in India. You're now, like, in Silicon Valley running a Silicon Valley tech company. You know, for those that are tuning in all over the world, like, Was it always that the world had this, like, felt this open to you? Like it was the world is your oyster? Or, you know, as someone growing up in India and like, did you think that you wanted to one day move to the United States? Like, you know, like, did you see the separation between, you know, India culture versus American culture and how that all comes together? And the reason I want to ask that question primarily is because I just want everyone that's listening to this to know that like, no matter where they're at in the world now, like we're all connected. And that's really what I mean by that.

You know, it never really felt to me that I was in another country per se, because of how global everything felt. Like I taught myself how to program looking at YouTubers from US and Russia and India and all over the place. And from a young age, because I had access to the internet, it just never felt different. When I moved here when I was 18, felt the same. So I think those global barriers that used to exist very deeply are just gone to a very fundamental level now. With regards to the Olympia, so I represented India for the IOI, which is the International Olympiad in Informatics, which for people who've done LeetCode interviews, it's that on steroids. And you have to go through multiple phases. To be selected as top 4 in your country. And then you go get to participate with the top 4 from every other country in the world. And I got into that for both informatics, uh, so that was competitive programming and linguistics. And for me, it was kind of a continuation for my love for problem solving, which has just carried me through the rest of my life, where you're given some extremely hard problems. Very little time. And for me personally, like that adrenaline drives me like nothing else.

So, so you're doing all those LeetCode interviews and that sort of thing. What language are you programming primarily?

So the informatics was in C++. And I joke, but I think it's probably true that I've written more C++ than I've spoken any actual language in my life.

So. I want to talk about entrepreneurship for a second. Is that something that you always wanted to be an entrepreneur or did it really kick in? You're like, you know what, I definitely knew from the age I was 12 that I wanted to be an entrepreneur.

I had wanted to be one since I was 12 years old. I remember I was 12 and I was launching apps on the stores and I decided to name my company Simple Inc. And my dad was like, you can't just add incorporated in front of something randomly. And I was like, nope, I want it. And I just had this dream of building a company that felt like magic, just like Apple. And I was very inspired by Steve Jobs at that time. And so Apple Inc. is going to be okay. It's going to be Simple Inc. And we're just going to build a lot of tools that feel like magic. And I remember the tagline for that company was simple, smart, and elegant, which I think still is how I think about Whisperflow and just everything that we build is like, that is the fundamental foundation on which all of this is built upon.

Now, back then, were you like monetizing all these apps or were like, you know, were you making money from them or just kind of making cool stuff and putting it out on the internet?

I was not monetizing them. I actually was like, I hate paid apps and I'm just going to give everything for free to the world. The moment when the realization became very conscious in my head was when I was building this product, which was a voice assistant. And at that time, I think it was one of the first voice assistants in the world. And I was at a party and I saw this one guy use it. No one knew I had built this thing, right? I'd kept my involvement very under wraps because I would get scolded by my parents otherwise. And he was using it and he was showing it to other people like, hey guys, you got to check this out. This is so cool. Like you can just say things and things happen because people had never seen anything like it before. And to just see that spark. In the eye of that one kid who was using it and the excitement of everybody around him. I think a switch just flipped at that moment and it's never been the same ever since.

So I want to talk about ConvertCC and just for the audience who doesn't know what that is, um, you know, you got millions of people using this thing. Like what was ConvertCC? Talk about it a little bit.

Of course. So this is a product that we built right around when LimeWire shut down. So to everybody who's, who's younger and doesn't know LimeWire, it was how you got music for free because iTunes charged you 99 cents per song. And so when LimeWire shut down, there was no good way to download music. So we built Convert as a platform where you can go and say things like, play me the latest song by Imagine Dragons. And it would find that and it would download it for you. And at that point it grew viral, which is like no marketing, nothing. I was a kid. I didn't even have any money. And it grew to about 2.5 million users. Around the world. And right after that, Google shut us down with a cease and desist because they didn't like what we were doing. Because there's one thing we had, which was if it didn't find the song online, it would go to YouTube and then convert the video into a song and download that. And Google didn't like that too much.

That's crazy. So, um, who was forking and paying for those server bills? Because those server bills had to be crazy. Was mom and dad kind of COVID that?

They had no idea. Uh, so I went to some hackathons and I got a lot of Google Cloud credits, which were like in $100 increments. And so I just kept creating new accounts and using them.

So let me get this straight. So you were literally building something on Google's platform that then Google comes after you on. So you're using their cloud for free, using their servers for free, and then that's all happening.

Yeah.

So what happened when you show your dad this? Like, I'm just curious, like, you get this like big scary cease and desist letter, like, we're coming for your life. Like, what happens?

Yeah, I showed it to my dad and he was like, please don't do whatever you're doing and shut this down. And I still think there was a way to just keep it up and revive it, but I was terrified. I was like 13 and I was like, I'm gonna shut this thing down.

So at that point, are you like super discouraged? Like, you know, or are you going to keep like, just keep building? Like, you know, like what were you doing there?

Yeah, we kept building. I think me and this one friend I paired up with, we probably built and shipped about 50 apps during middle school and high school. And then after that, I, uh, my parents were also very supportive during this time. So I think there were a couple of years where I just went to school for a day or two every month. And the remainder of the time I would just be home either doing college-level Coursera courses, uh, because I just wanted to learn more, or just building apps.

So I'm really curious, like, you basically were going to school a couple days a month. Like, how did you get away with, like, still getting good grades but, like, spending, you know, 18-20 hours a day on a computer? Like, how does one go and figure that out? Like, and only go to school a couple days, like, a month? Like, I just— how'd you hack the system like that?

I was Continuing to get good grades. And I had a couple of, again, a lot of this is just because I had really good friends who I could call the day before the exam and they would just spend 6 hours teaching me everything. Uh, cause I would do the same for them from other tests and the teachers who were just very supportive when I told them like, hey, I'm preparing for this, this competition, it's called IOI. And I need to give my all. And for the next 8 months, I am going to literally be on my computer 20 hours a day. And for them to be like, that's okay, doesn't matter. We'll take care of the school stuff. You go do your thing. I wouldn't be here without them. So a lot of it, I honestly like give to those people, those teachers, my parents, my friends. Uh, without whom just like none of this would be possible.

I love that. And they say it takes a village and quite literally it did here in terms of like having your support systems and that sort of thing. So ConvertCC, you basically build India's first music discovery platform. You're building it on Google Cloud credits. You get the cease and desist from Google and they come in and they slap you and they're like, we're shutting this thing down. Um, so. At that point, most people could have given up and they would've been like, you know what, screw it. But I wanna kind of unpack 3 things that you did here because I think there's a couple things for the audience to know. First, you built more than 50 apps before they took off, right? So not everything we do is gonna kind of take off. Second, it takes a village, right? And in a lot of ways, and what you said to me just a few moments ago is that you wouldn't be where you're at if you weren't able to sort of skip out in school and only go to school a couple days a month, it was because you had those perfect friends that were helping you out and getting you ready for those exams. So can you talk a little bit about that? Like, how do you get away with only going to school a couple days a week but still get good grades?

One of the things that I really worked on for a lot of my early life, and I think this is big, big kudos to my mom for this, is just learning how to learn things really fast. And what that helped me with is pretty much being able to read textbooks like bedtime storybooks and go through them and really understand everything very quickly, where by the end of it, you could ask me anything on it and I would be able to give you an answer. It wouldn't be perfectly what's written in the book, but it would make a lot of logical sense because everything went and kind of joined back and made sense in my head. So that just let me while I was in school, ace all my tests, and also do that while being enrolled in 14 college courses online at the same time. Wow.

It's amazing. Now again, I think a lot of people might be tuning into this thinking about it for themselves or for their children and saying, well, yeah, that's easy, but clearly Tanay has an off-the-charts IQ and that sort of thing. For folks who might not be as gifted as you, like, what are some techniques that you can recommend in terms of like, how does someone master the skill of learning? Like, just like take me through some of those building blocks because a lot of people listen to the show, they write in every month, you know, they're, they're young entrepreneurs. Like we had a 17-year-old kid write in recently who said, you know, I'm about to start my first business because of tuning into Forward Obsessed. And, you know, I wanted to say thank you. And I was just, it really moved me a lot. So for someone listening to this now, how do you, how do you pick up that, like, learn to learn skill?

You know, I don't like the word gifted, and it's something that my parents never used with me either. Whenever something went well, the thing they would say is like, you worked hard for this. And that, I think, at the core of it is the skill. Like, if you look at 3-year-old Tanay, what you'd see is this, this like short, like 2.5-foot kid who's walking around everywhere with a math book in his hands. 'cause I just absolutely loved it and my mom made it seem really fun. So I would be going to dinners at my friend's house with my math books everywhere and just do that day in, day out. And I think for young kids, right? And I did the same thing with my brother and he then went and did all the Olympiads. He's now doing a double major at Stanford. And so I don't think it's rocket science to get this right with young kids. But what you really want to do is make them just fall in love with learning new things and give them the same spark that they might get from eating a candy or playing video games. Because if you can do that, then you just become good at what you keep doing all the time, and your brain finds ways to just make itself more and more optimized at that. And so at the heart of that, I would say it's the simplest things. and there's not too many tactics or strategies around it.

No, I love that. I think what I'm hearing there is like, it's all about the framing and the modeling, right? If you look at something as being boring, it's going to be boring. If you look at something as being inspiring, it's going to be inspiring. You know, it's like, you know, it's like the old quote, like, if you think you can or you think you can't, like, it's what you think, you know? So I think, um, absolutely. So you get this big cease and desist, Google comes in, threatens to sort of scare the pants out of you and your family. Um, you shut this thing down, you keep going on and building stuff. Like, at what point now do you say to yourself, you know what, like, I want to keep being an entrepreneur, I want to move to the States? Like, when did you move to the States?

I came here for undergrad when I was 18 years old.

And where did you study?

I was down at Stanford.

Nice. So just for those that might be living under a rock, what was it about Stanford that kind of drew you here? and made you pick the choice to come to America and go to Stanford versus all the other places you could have studied in the world?

I think it was primarily because I wanted to be in the Valley. When I was growing up, I started following Silicon Valley a lot and the lives of the founders here all the way from when Digg or Reddit or Foursquare started back in the day and how this was essentially the place where in my mind, if you want to build a big software empire, you come to Silicon Valley and you do that. And so that was what was driving me. And Stanford to me was the place which made all the online videos that I kept watching. So they had an amazing catalog on Coursera, and I had no idea what Stanford was until the longest time. And I was like, what is this university that keeps making all of these really fun courses? 'Cause the thing I love about Stanford is it's a very, I would say laid back school in a sense where they don't take themselves too seriously. At the same time, there's just an inherent love for learning in the way they teach. That really resonated with me. And when I was in 10th grade, I went and told my parents like, hey, I don't wanna do IIT, which is the standard part for most kids growing up in India. And I want to go to Stanford. And that was a big shock for them 'cause they weren't expecting to send their kid abroad. You know, whole set of issues. They're moving to a new country. The fees are like, the tuition fees are about like 100 times more expensive than they are in an Indian college. And took about a year of conversation for them to like really understand what goes on, figure out the application process. And then 2 years after, I was here.

Amazing. And did your whole family move out there or did you go out by yourself?

No, I came here by myself and my parents are still in India. My brother's here though, which is, which is great.

Now, what was that like for like being a young man and picking up your life and moving across the world? I mean, like, was it like scary for you, like, to kind of make that decision? Just for folks that are thinking about making the big leap.

It was exciting. I think when you're young, you've been at home living with your parents your whole life. I think it's a whole adventure to really growing up is being somewhere and learning how to do everything by yourself. And that's something I think take a lot of pride in. And I think everybody growing up and coming out of high school should really focus on. Because it really changes you as a person. Because at the end of it, you get to a point where you depend on yourself and you know whatever's going to happen, I can take care of myself. And that's a level of self-belief that you can't get otherwise.

Yeah, no, I love that. And during your time at Stanford, I mean, obviously I hear lots of stories. I didn't go to Stanford, but I have a lot of friends who did. Did you have an idea for the problem that you wanted to solve? I mean, you've been obsessed with voice for a long time, but you know, when you founded Feather X, which I'll— we'll talk a little bit about in just a second, um, that was after college, right?

That was right after college, yeah.

So like, did you— when you were finishing up Stanford, did you have like internships and that sort of thing, or did you just have this idea like, you know what, I'm gonna go start a Silicon Valley tech company?

I applied to no jobs. I was just doing user research for this.

And for those that don't know, like, what, what was FeatherX? What was the big idea? And what did you go on to build? I mean, after you— so basically you applied to No Jobs, you graduate from Stanford, you're this whiz kid programmer, um, writing code probably 20 hours a day, um, at that point still, right? And what's FeatherX?

So we built FeatherX with the main vision to build a more personal internet. The way that you and I go to a website and we see the same thing, It makes no sense. And we wanted to change that. And we wanted, when somebody lands on a website, make them feel understood, make them feel like this website is for people like me. So we built FeatherX to essentially be a plugin that websites would install on themselves. And as a new user, you go on a site, say you're buying mattresses, and you're scrolling down and you stop at one place and you're reading a review about how it helps people with back pain. Cool, you scroll down and the rest of the website is just focused on solving that one problem for you. Then you go on a sports website and again, it's lumbar support and all of that that you're really seeing and being able to share that information about you across websites in a way that was very user-friendly. So as a user, you could go in and tweak the things about you that the system is learning turn off access at any given point. So very privacy sensitive, but also being able to make those websites work for you. So that's what we built. And then eventually it was a larger company that noticed us. They were working with big clients like Uniqlo and Forever 21. So they ended up buying FeatherX and now it lives on those brands' websites. If you go there and see things personalized, it's FeatherX under the hood that's powering all of it.

I love that. Now, and you were CEO and founder of that?

Yeah.

Did you raise money or just bootstrapped it?

We got funding offers the same time as we got this acquisition offer. And so we went with the acquisition instead.

Wow. So how long were you at it with FeatherX before— is it Cibra or Cibra?

Cerebra. Yeah.

Like the—

so I think it was actively working on it for about 6 to 9 months, and then another 3 to 6 months of doing user research and early prototypes.

Awesome. Now, one of the things I love about you that I, I really wanna call attention to is just, even though you're a full-stack engineer, even though you're a developer whiz kid, you have a deep appreciation for design. And I think a lot of that probably comes from, like, you had early inspiration from Steve Jobs, right? It's like, design is not how it looks or feels, design's how it works, right? So In a lot of ways, like, is that a fair assessment? Like, you kind of got that love for design at a young age from some of these kind of like, you know, Silicon Valley geniuses, if you will?

I think so. That was a big inspiration. And the second one was positive reinforcement. Whenever I built something that had good design, people had a different spark in their eyes.

How do you sell that idea to people when they see design and they don't understand it as an important investment? Because obviously great design is not cheap, right? It really drives up the cost. How do you sort of position that as a leader?

I think it really comes down to getting people to experience it. So this is an experience that we had at Whisper and other companies as well, right? Where you bring on some engineers and they're very excited to build new features. So they go build this thing up and I was like, okay, this is what we're going to ship. And then you got to really pause it and just be like, hey, why don't we let this loose in the company and see how people feel. And it's like, okay, cool, let it run. No one uses it. Like, okay, why don't you talk to the people and see why they don't use it? So you get feedback and then the engineer themselves would go and rebuild it with the feedback that they got and then ship it again and then ship it again. What they're really learning this whole time is design. And design is all about user experience, not less so about making things pretty. That's a separate part of it. But once they go through that cycle, they realize the essence of that. And what that has let us build now with Whisper is a very design-first company, not designer-first company, but a design-first company where everything that we build, like our product docs, start off with Here is what people might be feeling at this point. This is why they might discover this feature. That is what is going on in their head when they're doing this. This is what might be going in the head of an 80-year-old person who barely uses software. How do you make it really intuitive to them as well? And you list all of it out and then you go and build the feature.

This episode is sponsored by DS, a team of makers, marketers, and entrepreneurs that help businesses transform their brand experience and unlock growth like never before. Are you a CMO, founder, or business leader looking to level up your brand and marketing? Head over to werds.com and let's explore what is next for your business. That's werds.com. Visit us to learn more.

So picking it back up, so I wanna, I wanna understand more about For you, there's been a journey of what probably feels like serendipity for most people, but it just sort of feels like your life, right? And not a lot of people go, don't have a plan, start a company, and then 9 to 12 months later it gets acquired, right? So literally you're like, hey, I should probably think about raising some funds for this company. I'm, you know, I build out Feather X, I've got a demo. Like, were you thinking, hey, I'm going to go raise venture capital, I'm in the Valley, or, or were you just like, I'm going to bootstrap this thing? Like, where was your vision before Obviously things happened the way that they did.

I wanted to first make it cashflow positive, which we did, and then I wanted to go raise money because at a smaller stage we probably get a much lower valuation than we could otherwise, and I wanted to not dilute us as much.

Nice. Now when you were first building Feather X, like how many people were on the team? I mean, obviously you probably were doing a lot of the heavy lifting on the engineering, but who Who else did you recruit alongside you?

So initially we had one engineer who was based out of India, one guy I went to high school with and we worked on some projects together. And then about 6 months later, I hired another one of my classmates from Stanford who joined as CTO and he now runs another company called Featherwire.

Amazing. Amazing. So now what were you thinking about raising? Like what, like what amount of capital at that time?

We were looking to raise about $2 to $3 million.

And so did you put together the pitch deck? Was that kind of getting ready, like, to go out there? And then, like, talk me through, like, the play-by-play, like, how does Cerebra show up to the table and, you know, really kind of circumvert all that?

Yeah. So we put together a pitch deck, we applied to YC, uh, we get into YC, end up rejecting them because the, we were getting better terms from, from VCs. And then one of my mentors, Barish from Human Capital, puts me in touch with the CEO of Cerebra. He says, hey, you guys might have some partnerships. He's working in a similar space. They were building AI tools for large retailers like Uniqlo, Forever 21, like you name it. And as we were chatting, me and the CEO there just formed a really, really strong bond. We both kind of had a similar upbringing and at some point it's like, let's work together because I think this is one of the cases where the whole is greater than the sum of its parts. And a couple months later, we closed the deal and I ended up joining Cerebra as their head of product and engineering.

So let's talk about that for a second. Again, don't share anything that, that you feel uncomfortable sharing, but you know, I always like to ask the question is like, what was that process like in terms of like the acquisition, that sort of thing? Like how did that all come to life? You know, a lot of folks that are on the show, you know, at this point typically we go through the whole journey of how they, whether it's investment bank or how they went through the journey, you know, how, how did the transactions happen, that sort of thing. But this is a very organic type of acquisition, right? Like you, you had a relationship with the CEO. Clearly they saw something in you. What was that process like when it's like, you know what, like, why don't you just bring on and roll this thing into ours?

You know, it was very, very non-traditional because while this conversation was happening with Cerebra, another company reached out and they wanted to hire me to lead their team as, I believe, CTO. And I was just thinking, hey, you know what? Let me sell FeatherX to Cerebra and then I'll go join this other team. And as those discussions were happening, this other team also put in an offer for FeatherX. And so we had a bidding war going on between these two companies. And then eventually working with Cerebra just felt right. I was closer with that team. And then eventually decided to just go work with them.

Nice. Now, when that transaction happened, was that sort of an all-stock deal, mix of stock and cash? Like, how'd that all come about?

It was 50/50.

And are you able to talk about the amounts, or that's something you kind of keep under wraps?

That is under wraps.

No worries. The question I do like to ask is, was it a life-changing acquisition type of moment for you, or was it sort of like an affirmation, but you're going to keep going type of thing?

As a recent college grad, it was life-changing.

Awesome.

Yeah.

So at this point now you're, you've sort of graduated beyond ramen noodles to a nice, nice place to live out there. Yeah. Love that. So what was that like? I mean, obviously you've been a bootstrapper and a hacker your whole, your whole life, quite literally since you were like 13 years old, right? When you become like head of product, I think it was head of product AI and engineering was your remit there, right?

I had initially started off as head of product and then slowly got promoted to lead and build the engineering and AI orgs as well.

Nice. So as someone who loves to code and it's a big part of your identity, again, you're a young guy, you just graduate. Like, what was harder in your mind, like building the product itself as like an engineer or leading the teams that build the products?

It was definitely people.

Why?

I think building products is what I've done my whole life. I can probably code entire things blindfolded and like that, that isn't too big of an issue for me. It's just fun, but a very solvable problem. But managing people was harder, right? And especially like, imagine me, I'm 21 years old, clearly don't know what I'm doing with life, right? Just graduated from college and I have to start managing a lot of people, some of whom are as old as my parents. And like, to be honest, I was I was bad at it for the first few months. And it was one of the things where you don't know what you don't know. And I had this mentor who was our chief business officer at that point, who kind of took me under his wing and would give me a few books to read and gave me just some tips on how to work with people who are older. And then I made it my mission Or over the next 6 months, I want to be one of the best managers that any of these people have ever worked with. And that was again, a huge learning opportunity for me to do that. And eventually 6 months later, reached that point where people would organically come up and say that to me. And I was really happy with it. But I'll say at Cerebra, that's what I grew in is being a manager, leading a team, building a hiring organization and just scaling that up from, I think it was 5 people when we started to 15 people over the next year.

Amazing. Now, clearly, you know how to crush books. We've talked about that a lot. What were some of the most influential management books and leadership books that you read that you think other engineers can get a lot of benefit from reading?

There's one which is called The One Minute Manager. It is a book that we've also given to every single employee at the company, both to be good managers and know how to be managed well. So that's one I highly recommend, like a 1-hour, 1-hour read. But it changes your perspective on how management should really feel like.

I love that. What do you think sets apart a great engineer from a great manager?

I think they're two like completely different skill sets. As a great engineer, the singular thing that I think is necessary to be a great engineer, even in this world of AI, is being able to take a big complex problem and then break it down into small pieces. And then anybody can take care of those small pieces. That doesn't matter. But the first part is what is essentially an art form that you learn. And that is what I think true engineering is all about. And that, I think, regardless of how good Cursor and AI and any of these systems get, this is going to continue to be true and continue to be what I look for in great engineers when we're looking to hire one.

I love that. And how about a great leader? What sets apart a great leader?

That one I would say is harder to define, and that is primarily because I think also leaders and managers in my mind are two completely different people.

That was a trick question. That's why I was hoping you took us there. Yes. Take us, go deeper there, please.

Yeah. So when we're talking about a leader, right, you are generally looking for somebody who may or may not have any direct reports at all. And what that means is they're not responsible for like taking care of anybody's promotions, having one-on-ones with them, assigning them specific tasks. But the main goal of a leader is to provide everybody who follows them with a sense of certainty and give them a higher purpose to go towards. And I think if you have those two things you're doing those two things, then you are a really good leader. Now I know there's a lot of words that get floated around of being very charismatic or being like pushing people really hard. And I think all of those other traits come down to just being true to yourself, right? And there's this framework to think about leadership, which is called the DiSC framework, where it puts leaders in this category of There's dominant, conscientiable, there's inspirational, S is something else. And what the best leaders do is they just figure out what's true to them. And for example, for me, I can't push people to beyond their physical limits. I can't push people to spend less time with their families. That just doesn't align with my values. And so how do you give people certainty? How do you make them inspired? How do you give them purpose to go for while still sticking with your values? I think that, that second part really shapes you into the best leader that you can be. And so one thing that I learned along the way that was really helpful for me is when you're thinking about the kind of leader you want to become, try to find other people who have similar values as you. And it is hard because you often see the most controversial leaders get the most amount of screen time. Like you would just hear all these stories about Elon Musk. Like, yeah, he's the most out there. You don't hear that many stories about how Satya Nadella is as a leader because like everybody's just generally like pretty, pretty happy and it's not as controversial. But I think in some of the lesser-known leaders, you will find better role models than going for the most popular figure.

I love that. One of the quotes that came up in the research that I really loved was like, in a report card once, it describes you as 99% perspiration, 1% inspiration. Can you tell us a little bit about that story and like where that came about?

You know, this was the teacher that said this to me. I was in kindergarten. And a lot of other teachers would call me gifted at that point, and that's how they would refer to me. And this one teacher was like, you know, Tanay, I don't care if you're gifted or not. I just love how hard you work, and I see you outwork everybody else. And that is the thing I admire about you the most.

I want you to go deeper on that, Tanay, because you have a tenacity, you have a work ethic. Still to this day, that blows my mind. I mean, we connected in the middle of the night one time about your product early when you were just first getting going. And I want to just unpack for the audience, like, you manage a lot of people, you lead a lot of people now as a CEO, right? People say a lot that the new generation doesn't want to work hard. And, you know, I don't know if you agree with that or disagree with that, but what I'd love to hear from your perspective is like, debunk that for us. How do you find people that want to work hard? Because people that work at Whisper, they're not building the product that Whisper has become today and will become in the future by not working hard. You're solving real hard problems here, right? We're not talking about like ShadCN and Tailwind here, right? This is like hardcore engineering. So how do you find the people that are going to bring that level of work ethic to make sure that the culture that you're creating and setting, people are keeping up with?

That part is actually really simple. One is we look for examples of that in their past where I want people to describe how they got so engrossed and passionately involved with a project that they couldn't think about, couldn't stop thinking about it day and night. And the second is in my first call with all candidates, I explicitly tell them this is going to be the hardest that you will ever work in your life. And not necessarily in terms of even the number of hours, but more so in how much you're going to have to be pushing yourself every single day. And if this is, this is not what you're looking for, this is not the place for you. And I just want you to know this beforehand. And that really helps filter people out.

Model that. So like, if I'm on an interview with you and I'm listening and I'm in your back and I'm a fly on the wall. We all know what the thing, thing would say that would clearly screen them out, right? But what do you typically find that screens people in? What are those signal words for those that are trying to recruit this amazing talent that you've done?

They lean in and I can see them get excited. And that's when I know we, we found the person, because you can tell from their expressions that this is exactly what they're looking for.

Love that. So how, once, once, uh, Cerebra came in and acquired Feather X and you go on to become head of product and head of all the things. How long were you there before you're like, you know what, I'm ready to step out and start my own thing now?

Exactly a year.

You move fast. And like, what was it that hit you? Like, did, did the idea for Whisper come back to you? Were you like side hustling on it, or did you just be like, you know what, I'm ready to move on?

No, it did. I was— when I joined Cerebra, I planned for it to be the thing that I do for the next 5 years of my life and then grow it into a really large enterprise. And that was when I was chatting with Sahaj, who is now my co-founder and was previously my college roommate all through undergrad. So we go way back and he just left his job. And so he called me one day and he was like, Tanay, I just left my job. I think I want to start a company. By the way, this was a total surprise to me. This man never wanted to start a company ever before and he wanted to go do a PhD. So him saying this to me was like, got me really excited. And he said, I would love to do this with you. I know you're at Cerebra, so tell me what you think. And then I was kind of left in a pickle because I was just halfway through Cerebra, right? 6 months in and just settling into my new job, starting to get a lot more responsibility. And I told him, how about we spend some time working together? And going through ideas, see what makes sense. And I told him like, hey, you know what? I've had this vision for what a very personal voice assistant could feel like that really understands you, that moves people away from being on their screens all the time to having something that just exists in the background and gets everything done for you. And I want to, like, if there's one thing that I would love to build is this. I don't know if it's possible to build this right now, but this would be my dream company. And so as Atesh does, he goes and he reads a bunch of research papers and then we start talking about it. We start doing this thing on the side. It slowly goes from like 5 to 10 hours a week and I'm now spending all of my weekends working on this thing with him. And eventually we start to more seriously talk about starting this company. And I think a big thing was for about 2 months, we were having discussions about what our values were, what we cared about, how would we think about if we got an acquisition offer at this deal, what kind of people do we want to bring on our company, how do we want to think about the finances and the vision and the ambition and the scale of what we're doing. And I think that set up, set us up for a really strong foundation. I've been working with him for a little over 3 years now, and I think it's maybe one of the strongest relationships I do have.

I love that. And I want to talk about the founder, co-founder dynamic, because that's a big one that you get that wrong and shit hits the fan. You get that right and magic happens, right? What made you realize, like, you have this belief that screens are obsolete. Right? It's something that you believe like deep, deep-seated inside you. Like, what was it about that eureka moment that kind of led to this idea that we're going to build a voice-first interface company? Like, I want to sort of unpack that for more because, you know, for those listening, it's like, dude, we've had transcription services forever, right? Dictaphone, Dragon, Siri. Like, why do we need another voice company? Like, what's so different about what you're doing that everybody else isn't already doing?

I'd say that the simplest thing to that is you don't use the other things as much. What my vision for Whisper is, right, that for it to be something that is powerful enough to replace what you do with screens all day, and by screens, like your computer, your phone, everything else, it needs to be a voice interface that you trust. That requires 3 things. Number 1, it needs to understand you. If you don't feel like this thing understands you, this thing gets you, this thing knows small nuances about you, you're not going to trust it. The second thing is it needs to be competent. And the third is it needs to be reliable. Having a really smart voice assistant is very similar to having an actual real-life assistant. And these are the 3 traits you would look for them. Look for in them as well. And if you have somebody like that, you'd be very happy delegating tasks to them, right? And having them access to more and more of your life because you know they're there for you and you're essentially on the same side of the table. And that is fundamentally how I think about what we're building at Whisper. So I was saying earlier, everything is driven by human emotions, including these ones. And so this shows up in the design, this shows up in how we're thinking about the different features and product lines and messaging to users. Because once we have people feel this way, that's when you get to a point where people then start saying, hey, actually I don't need to review what Whisper has written, just send it. And that is kind of the step one to reduce your emphasis on on screens and then more and more where it just starts to feel like an assistant that's, that's with you all day long.

No, I love that. It's a transformational product. I would, I would almost ask you to share your screen and use it, but we're definitely going to cut some footage in later to show people doing that because it's such a powerful product. Let's talk about the funding journey. Did you do this again? Were you like a— was it a cash flow positive company again before you went out and raised funding? Or given the size of the problem that you're solving, did you have to go out and raise?

No. For this one, we realized that it's going to require a lot of R&D. 'cause we're gonna have to go in and build a lot of things from scratch 'cause solutions that work don't exist right now. And so we went out, we raised our first round of funding, that was about $4.6 million and it was led by NEA and 8VC. And then after that, we had some more interest, so we raised another $5 million and then most recently we raised a $12 million round. And so overall, it gives us a lot of capital to get the best designers, the best engineers, the best machine learning researchers to really push the frontiers of what AI can do.

In your most recent round, if I read it, it was a $12 million raise led by Matt Kraning, right?

Yeah, that's right.

What was it about that? Like what struck Matt? Because Matt's sort of a notorious investor in a lot of ways. Like what struck him as saying, you know, well, I want to come in and sort of take a bite of this apple.

So I got connected with Matt totally serendipitously during our first round. And so Matt was actually one of the first checks into the company.

Oh, wow.

And he was the one who put us in touch with NEA and everything went from there. But we were having a lot of discussions about what the future of Whisper would be like. And the way that he saw it, and he had a lot of conviction in, is if you can build a voice interface that people trust, you are existing on the input layer. And what that means is you can build a whole platform where users interact with everything else through you. And that is essentially the new browser, the new search engine, the new iPhone to access this new internet that we're building today. That is going to be very natural language focused. And that makes this company have a scope for being a, not just a billion dollar company, but a multi-hundred billion dollar business. And that is what, I mean, I think we were Matt's biggest checks at that time. And for an individual to lead a round of this size is like usually unheard of. And so honestly, really grateful for that word of confidence from Matt. And he continues to be one of my most trusted advisors and investors that I spend a lot of time with.

So like, I'm a tech guy, you're a tech guy.

Yeah.

Why do voice assistants still suck, right? You got companies like Apple out there, Google, they've got the best engineers in the world. Obviously OpenAI has launched a product called Whisper, which they've open sourced in a lot of ways. Like, you got Deepgram. There's a lot of tech out there that have tried to solve this problem. Like, in your opinion, let's get technical for a second. Like, why do voice assistants still suck? And how is Whisper and what you're doing fixing it?

They're solving the wrong problem. They are— every single voice transcription service today, they optimize for this metric called word error rate. Of the words that you said, how many of those words did we get perfectly right? And they're like, we're 98%, we're 99%. It doesn't matter. Being 99% good at the words you get right means that in every sentence, about one word is going to be wrong. And that actually is problematic because that means you can't trust a single sentence that comes out of them. The metric we optimize for is what we call zero-edit messages. What percentage of your messages are perfect to send? And so it's a lot about understanding the person and what is ready to send instead of getting everything down word for word. And it's very human for voice to work that way because we think a lot differently than we write. And so you don't want to send your rambles to any person. You want to send a coherently drafted message over to the other person that conveys your thoughts in the right way that they can understand as well.

So when you talk about like conveying thoughts to people, you guys were just recently selected in 2024 for the AWS Generative AI Accelerator. So that's a big deal because a lot of people trying to chase the generative AI stuff and really you know, as we've seen with the memes, it's basically everything's just ChatGPT wrappers, right? Or Claude wrappers. Um, what is it that you're doing so differently? Clearly the way you're framing success, the metrics, like the way you think about everything is different. Like, what do you think led to being able to get selected for that 2024 AWS Generative AI Accelerator? Like, what are some of the engineering problems you're solving or have solved that qualify for you to be breakthrough?

So I'll answer this question in two parts. One is, one of the things that's very different about Whisper is we treat models as just starting points. And my co-founder, for example, was one of the people who was the pioneers behind diffusion models, which is now what Midjourney, DALL-E, all of these are built on top of. And the other people on our machine learning team are two of possibly the most talented ML PhDs that I've ever met in my life. And what that means is whenever we're working with a model, we are going in and tuning knobs that most people don't even know exist to get them to do what we want. So one example of this is we used to have Flow hallucinate quite often earlier. Sometimes when you ask it a question, it would just respond with the answer instead of typing in the question, which is likely what you want to say to somebody. And really by going in and figuring it out, we made that hallucination rate, which is a huge problem for all large language models, drop by about 1,000x. And this is just one of the many problems that we have to solve along the way. To make it the system that is very reliable, works across people, and at the speed that it does. But when I was talking earlier about having that incredible focus on design, it's all that this comes down to, is really knowing how to make these models do what we wanted. So that's something that AWS saw. A lot of them at Amazon are users of Whisper, and the entire judging committee was using it by the end of the process. And so it was a, uh, kind of a good, good relationship with them.

I love that. And is it safe to say that you're on the AWS stack primarily?

We primarily are. Yes.

Smart choice for them, clearly. Um, so you also nailed something I wanna, I wanna unpack it too. Like you're a hacker, you're a designer, you're a builder. You got number one product of the week for Product Hunt last year. Like that's a big deal for a brand new company. How did you do that? Like, for— because folks listen all the time, they're like, I know I got to have a successful launch on Product Hunt or else I'm, I'm dead in the water. What are some tips that you'd give to somebody that would help them get to number 1? And more importantly, how the hell did you do it?

There's a few things. So number 1, we actually spent a lot of time in preparation for it, for that launch and every launch that we do. It's about 2 months. Of heads-down preparation that we do, which is nailing the story, nailing the features, nailing the assets and the narration and the video that we're going to put out because it's all part of the package that entices people. The second thing is in the product, our goal is to get the user to an aha moment as soon as possible. Which for us happens when you're onboarding and you do the first dictation. So that gets us to have a lot of very excited users. And when we launched our Product Hunt, a huge percentage of our users actually went, made accounts, and then went and uploaded it just because they loved the product so much and they had that excitement feeling from the first minute of using it. And The third thing I'd say is we're just very in touch with our users. So even now we have a 20-person company at this point. I probably talk over email to over 100 users every single week, spend 2 to 3 hours on calls with them. And it could be user research. Somebody has a new feature idea. I hop on a call with them to understand what they would want to do this for. And that bond really, really shows and drives the engagement and helps us know what is truly gonna resonate with the people.

That's awesome. Now when you're doing Product Hunt, did you like bring on agencies or did you just do this all internally?

Did it all by ourselves.

And had anybody on the team previously done a Product Hunt launch or did you guys just have to figure it out from scratch?

It was all figuring out from scratch.

Oh man, we definitely gotta have you back on for a part 2 for sure. Can you talk a little bit about just like user growth metrics? Like what, what, what you're comfortable sharing in terms of just how do people are using Whisperflow every single day or how do people have used the product? Just like anything you're willing to share.

Yeah, a couple of things that I think are very novel outlier metrics about the product is about half the users who have ever downloaded the product are still using it every week, 3 months later. And so this is really important because for a lot of AI tools, you end up having AI tourists who come sign up on your platform, you try it once, and then they bounce off, never to return again. And it is really hard in this market with so many tools coming out. To be something that is that retentive. So that's something that we worked on really hard to get that number as high. And you start to see the results of it because people who use it as much are also going to be proponents of you and like you're one of them. And now we're doing this podcast, right? So the value for it definitely compounds. The second thing I'd say is about 10% of the users who have ever downloaded the app are paying for the product right now. This number usually for a lot of other products is about 5% or lower. And these are companies like Dropbox and so on. And so for us, that number is really important, which means that people very early on are finding a lot of value with the product where they're willing to pay for it. And we actually have a lot of students for whom they don't pay for ChatGPT, they don't pay for anything else. And Whisper is the only product that they're willing to pay money for. And I think that is, again, it's a sign for us, like we're adding genuine value in people's lives and like genuinely changing how they interact with technology and sets us up with a really good foundation to build and scale even more.

Speaking of not being able to live without the product, um, are we going to be able to let the Ford Obsessed listeners get early access to the iPhone app? Because I have to say, whenever I go away from my computer and I lose out the ability to push that button and trigger Whisper and I have to go back to Siri dictation, I want to throw my cell phone across the screen. So tell me, like, how soon until the iPhone app happens?

We are going to ship the iPhone app to all of the Pro users by the end of March.

Oh my God, I can't wait. So definitely, definitely want to be on that list for sure, because, uh, of course, a pro user, I couldn't live without the product. Um, so we've got a revolution in front of us, right? And you're leading the revolution, and, and I would call it the, the voice-first revolution, right? So for those of, of you that have kind of heard about this big cursor wave, we talked about our most recent episode, all these vibe coders out there that are using LLMs or using Whisper and talking to their computer and just having it spit up in CodeApps. I want to talk about, like, if vibe coding is how people are building software today, how people interact with computers in the future, interfaces in the future, using something like Whisper.

It's going to look very different for people, how people are imagining it right now. And the reason I say that is, is right now we live in a display-first world, right? You expect to have app icons. You expect to go and click on places to get things done. You expect to navigate across these screens and scrollbars and tabs and buttons. And I think that is going to be elements of a bygone era, the same way we feel about command line interfaces from way back in the day where that was what people thought computing is going to really be like. And so when I think about the future, and it's really coming really, really soon, it is a lot going to be based on language and what I call contextual AI. You're going to have things come up, tools be custom built for you as and when you want them, what you want it for. We're going to have computing that understands you. The hardest part about interacting with computers is communicating what you want. And it happens all the time with the assistants, right? You tell ChatGPT to do something, it doesn't get it right the first time, and it's frustrating because of course it doesn't know what's going on in your head. It doesn't know your preferences. It doesn't know so much about you. How will it ever do a good job?

When I prompt it, it always gets it right. Just saying.

Pete, you're just good at it then. You're one in a million. And the hard thing is gathering that context about you and personalizing these systems so they are very custom built. And so when you are interacting with your system versus somebody else with theirs, it's going to look like a whole new world for each of you. But for you, for selves, it's going to be the most intuitive way to interact and The biggest part is it's going to feel like a very core part of your personality, and without it, you're going to feel like something is missing in life. And that is a change I don't think most people are expecting to happen.

I'm curious, when you look at the younger generations that are buried in their phones, um, you know, I know when I send voice notes to people or those kinds of things, sometimes people are like, okay, millennial. When you look at the younger generation, like Gen Alpha, Beta, etc., Yeah. Do you think that they're going to be as open and excited by Whisperflow as someone like me is? Or do you think that they're, because they're so used to not talking, they're very nonverbal in a lot of ways, they're not going to glom onto it the same way that I have?

The younger generation actually loves voice. If you see, for example, the people who are the biggest voice users, It is the people who are 35 and above and people who are 10 and below. Interesting. Because they grew up in a world where Alexa was a thing. Like, that is how they turn on their bedroom lights. So to not do that and press buttons, that is weird. And that is where I think the revolution is really going to be driven from. And the older and younger generations are going to be the biggest adopters. I think Millennials are going to be perhaps the latest adopters of voice, if I were to make a guess.

Yeah, no, I love that. It's been an awesome episode. I know we had a couple of technical glitches today, which was pretty funny. I want to just kind of switch over to like a lightning round really quick, rapid-fire questions, just because I've been really curious and, you know, this is all about me. No, just kidding. So we'll just do a quick lightning round and then we'll wrap this up and definitely want to tell folks where and how they can get their hands on Whisper, because let me tell you, folks, when you use this thing, You can't live without it. And I'll make sure to put links in the promo codes and stuff for you guys so you can get all hooked up as being a Forward Obsessed listener. So rapid fire, here we go today. So some books that have shaped your thinking that like literally the same way that, that Whisper changed my life for how I use my computer, books that change the way you think.

First one I would say is Design of Everyday Things. And the second one would be Siddhartha by Hermann Hesse.

Ooh, good one right there for sure. That's a pretty powerful. Daily habits and routines that shape you.

Waking up every morning and having a really calm first 30 minutes while I drink my morning coffee, and it puts my body in a place where it's ready for the day and I can focus on having very calm, composed energy.

Love that. Most underrated skill as an entrepreneur?

Smiling. You are going through ups and downs every single hour. And it's one of the things my yoga teacher once told me when I was scratching my face when doing some yoga poses. He was like, is your face hurting? I was like, no. Then just let it smile. And kind of a philosophy for my life now.

It feels good when you smile, doesn't it? It's like even just right now I'm smiling, like it's putting me in a better mood too. So I couldn't agree more. One of your big tech predictions for the next 5 years. I feel like I know the answer given knowing you, but just go ahead and share it anyway.

Tanay Kothari. Your personality is intertwined with technology.

Advice to your 20-year-old self?

Don't be too hard on yourself.

And the hardest business lesson that you've ever had?

Great products are all about habit building.

Amazing one for sure. This has been an incredible episode. Um, one of the reasons I love doing what we do, and again, just want to say thank you so much to all of you tuning in on the Forward Obsessed show. Um, today, this has been an amazing episode. I want to say thanks so much for coming on. Where do you folks go to learn more about Whisper? How do we spell it? What's the website we go to? Hit us with all the promos and do it in the dark. Yeah.

So you can go find the website at whisperflow.com. That's W-I-S-P-E-R-F-L-O-W and whisperflow.com. That's the website for the product. You can go download it for Windows and Mac, and we're soon going to have it on mobile as well. And we'll have a promo code that is going to be shared. I'll send it over to you, Pete, and get some, get some good offers for all of our listeners.

Awesome. Well, again, as always, to the audience and to you, Tanay Kothari, for coming on today and dealing with our fun tech glitches. Thank you so much. You embody Ford Obsessed in every sense of the way. So There's another episode of Ford Obsessed. Thanks, everybody. And please make sure to leave us a review. Keep tuning in so we can do another 100 episodes that are as awesome as we've been so far. Thanks so much, folks.

Take care, Pete.

Thank you for listening to Ford Obsessed. Please share this episode, subscribe, and leave a review on your podcast app. If you'd like to contact our team, visit us at fordobsessed.com.