Building Cactus: AI That Actually Helps Solopreneurs

Automatic transcript of the recording (English); speaker labels are automatic. Timestamps open YouTube at that moment.

0:00Sacha: Welcome back to AI with Sach. Today, I have the pleasure of having Avinash and Ajith. You guys are amazing. I saw your posts about Cactus and I wanted to get into it with you guys. So let's start with you, Avinash. You've said on LinkedIn that you've always wanted to build a product-first company, solving large problems with simple tech. Where did that mindset come from?

0:24Guest 1: Avinash has been at the top of my mind. Beautiful pieces of software. That's how I look at it. This has been even since the first time I got introduced to a programming language called Logo, where all it did is moved a turtle that went up. You could create patterns based on it using commands. It was so simple and very effective in learning programming. I was in fifth grade and I enjoyed it. I loved it. I'm always curious with technology, curious with building products. And as I progressed in my career, I got introduced to companies like, of course, we have Apple that builds awesome stuff. That's where everyone strives to be. For, in that time, I also got introduced to a company called 37 Signals, which is the maker of Basecamp, Hay and many other products. And their approach to building simple products is amazing. Effective simple products, as simple as that. Shopify is another example.

1:26I love Shopify. So these are kind of products that really resonate with me. And that's what I mean when I say that.

1:33Sacha: I love that. You've also mentioned that you lead and build with empathy. What does that look like for you? That's

1:39Guest 1: a great question. For me, empathy is not just building or understanding what our users want, but also understanding our internal team dynamics and kind of balancing both of them together. It could mean that we get a lot of customers. One way of doing it is just keep crunching, building it, and then shipping. The other way is not to do anything about it, think about it, you know, and then extract certain features that you hand off to your, you know, your team to actually build a meaningful piece of a feature or software rather than a clunk that you build multiple things and then just, you know, hope that it works. I'm optimizing for not just being empathetic to our users, but also our employees to make sure that both of them succeed at the same time. So that's, that's what I mean when I say empathy.

2:40Sacha: I really resonate with that. What's a belief that you hold about software and, and, or startups that most people maybe around you don't agree with?

2:50Guest 1: Don't build too fast. I guess what I mean by that is, and also use, we use unconventional methodologies to build our product that maybe not everyone resonates with or agrees to. One of them is currently for Cactus. We are using Ruby on Rails to, and we're using the monolithic architecture, which people who, people who are technical here will probably know what it means. So instead of building multiple small microservices, writing APIs to connect them all together, the way I'm looking at it is building one big monolithic software using Ruby on Rails, which is again, for most people is unconventional, but I'm optimizing for what we call developer happiness, which means that you end up shipping faster and, you know, kind of, it helps you. It gives you, it gives you the batteries included, so you don't have to worry about the nitty gritties. It's convention over configuration is what, what it's called. So that is one thing. And when, in the beginning, when I mentioned about build slow, what I mean is, what I do not mean is do not iterate and ship fast.

3:57What I mean is, when you get requests, don't just build it and ship it. The way we approach this problem, as I said before, is we, we listen to our customers. We just let slide a little bit. And there are certain, when you talk to a lot of users, a lot of customers, which everyone should be doing, you start to see patterns, right? So rather than building for each customer, you group this patterns together and you might come up with a much better implementation that works for a larger audience. And rather than solving all hundred problems in a mediocre or less effective way, I would rather say solve 10 problems that are more effective and drive that value.

4:41Sacha: And it's very interesting. I heard that Jonathan Ferry, who built Good Food, I don't know if you guys know that, but he was saying that it's very interesting because a lot of people say you have to listen to your customers and, you know, talk to them, et cetera. But he says and thinks that it's better to look at what they do, because sometimes there is a big difference between what they say and what they actually do. And sometimes I feel like not founders, but companies or startups are, as you mentioned, stuck into trying to build everything that customers say rather than really taking the time and building slow. And I really agree with you on that point.

5:22Guest 1: One thing I want to mention in that aspect is every customer has wants. They want a lot of things, right? But do they really need it? That's what you should go behind. If you start building all the wants, I think you'll end up building. I mean, again, I'm not saying there are people who do it effectively, I guess. But from my perspective, from our perspective, instead of building the wants, build the needs, just as you mentioned.

5:48Sacha: Yeah, I love that. Ajit, you've previously raised 20 million plus and worked with pretty big companies. Why pivot to small businesses?

5:57Guest 2: It came from a personal need, Saj, right? So both of us are batchmates from college and we work closely in our previous startup as well. We built the product end to end and we saw this journey where we implemented that across 15 or 16 of the top 20 CPGs, including Pepsi, Unilever, Coke, all of that, right? And the reality was that, yes, the impact is there for sure, right? But the impact is for the companies. So we are helping the companies, right? We are helping Coke and we are helping Pepsi, save money, save time. Absolutely. But as we started going into that journey, we actually became dads. And both of us started getting help from private chefs who would come to a house, do the entire meal prep for us. And that's when we started discovering that they have some intense pain points, right? And they're running a great business, by the way. It's not like it's a small business. Some of them are making $20,000, $30,000 a month, which is amazing for a single person or like one or two people business.

7:02But they're very busy. Their entire life is revolving around managing their customers, the admin work. And they, at some point of time, they see their dream fading away that I want, I started this catering business or I started this any small business, right? And I'm not able to really enjoy that because I have to do all this admin work. And it was very clear to us that there is the need for a software which can manage all that admin and let them focus on what they truly love to do, right? So that was the first piece of the insight. The second piece of the insight was when we actually moved to San Francisco in January, we saw a very clear glimpse into the future. The future is very clear, right? Anybody and everybody is going to get access to building software. Now, it doesn't mean that everybody will build like enterprise-level software or like very complicated software. But as you can see from Lovable and all those other, AI is just going to keep getting better and better.

8:07The future is where people have access to software and they can solve it. So it was very clear that that's why we started building Craved, which is like a complete AI business assistant for all these solopreneurs. And the impact is massive, right? We see this daily. They are thankful that we're using this. And we believe SMB, small, medium businesses and solopreneurs are going to be the future. There's going to be millions of them. And the entire market is going to change. And we believe that the first billionaire solopreneur will be built on Cactus, right? So that's the dream that we have.

8:47Sacha: I really hope so as well. How did your experience at Turing Labs shape your approach to building Cactus?

8:55Guest 2: It is very, very pivotal, right? Because I was a first-time entrepreneur and I didn't take a salary for four years, made all the sacrifices possible and the first-time founder mistakes as much as possible, right? Everything. So I had a great opportunity to learn from all of that. Now, the thing is when people go through that, two things happen. One is either they break, right? And they give up on the journey of entrepreneurship because entrepreneurship tests everybody. Like, literally everybody. I haven't met a single successful entrepreneur who said that, oh, it's been beautiful, like paradise. You're going to have a wonderful journey. Everything is awesome. Even today. Even today. Like, it's stressful. It's intense. You have to give up, I would say, almost everything to be successful, right? And that I did that and I learned so much from the journey, which really helped me to set this journey correctly.

9:59And I think the most important thing for me was finding the right co-founder with Avinash. We are batchmates from college. We know each other. Even before we created Cactus, you know, we had this discussion that what kind of company do we want to build? What is our philosophy? What should our partnership look like? We are part of that partnership, truly, right? That's what the world sees. But between us, Cactus is one part of that partnership. The partnership is a bigger thing. So, yeah, that's the high level, basically.

10:32Sacha: That's awesome. For those who don't know Cactus yet, could you pitch me the business in two or three sentences?

10:42Guest 2: Yeah, absolutely. So, Cactus is the AI business assistant for solopreneurs like private chefs, personal trainers, mechanics. And what we do is manage and automate all the admin tasks so that you can focus on what you truly love to do. That's the quickest way to say it.

11:04Guest 1: Business in a box, basically.

11:05Guest 2: Yeah.

11:05Sacha: Where are the, like you touched upon that, but where did the initial spark for Cactus really came from?

11:13Guest 2: I think you should talk about this one. Yeah. So,

11:17Guest 1: you know, when we were, when I was working in the previous startup, in a startup tutoring labs, at that time, we, me and my wife, we had a second kid. And until then, we would just, basically, we would meal prep and we would cook at home fresh always. We do not eat out much. Even if we do, it's like once a month, maybe twice a month, because somebody says, you have to go and eat at this restaurant. We don't enjoy eating outside much. We are foodies in the sense that we love eating home-cooked food as much as we can. But what started happening is with the startup life, my wife being a full-time, you know, software engineer as well, and having two kids, it started becoming a little bit hectic for us, and we couldn't do that anymore. So we started looking at, like, DoorDash, even meal kits we started looking at. Yeah, yeah. But it was, like, it did not work out for us. It was not personalized. It felt very manufactured, and we couldn't, it was not healthy, and it was expensive. So we started trying to look around, and we found one Indian auntie who basically would come to our house once a week, and then she would basically cook the food that we like to eat.

12:26We would put it in the fridge, and then we would just have it for, you know, a couple of days. And it was not expensive. We thought, that is private chef, technically, if you think about it, right? And we thought it was going to be very expensive, but it was not. We really enjoyed that experience, and I was talking to Ajit about this, and that's when he said the same thing. Because his wife is an amazing chef. His mother-in-law is also an amazing chef, right? So he was like, dude, I'm facing the same thing here in Toronto, and going through similar things, right? So yeah, that, I think, was kind of the genesis of this. And then we started looking at what the market is for this. And there was no platform, truly, that connected authentic private chef experience with customers who were looking for it. That was kind of the genesis. We wanted to basically organize this completely unorganized market. I think that was the genesis of Craved. I think you should talk about, you know, how Craved can impact us just recently.

13:28Guest 2: Yeah, so that was the start. And we started off as a marketplace, right? Where we connected private chefs, caterers to busy families who need home-cooked food. But as we started doing that, we started seeing some patterns, right? We started seeing that everybody has admin challenges. Now, if you think about a traditional marketplace, most of the marketplaces are good at connecting the two sides of the marketplace, right? So you build one side and then you connect them. But what we realized is that that's actually not the problem for private chefs and caterers. Like if you're a private chef or a, like, you know, a person who does cooking, if you just go to Facebook marketplace, you literally get 50 plus inquiries in a single day. That's the amount of, like, need that is there in the market. And we started seeing that with our chefs. So the challenge was what happens after that, right? How do you manage all that? How do you, you know, do the scheduling? How do you do the payments? And me and Avnash were meeting with the chef. We asked her, like, how much did you earn last month?

14:29And she was like, she gave us a number. And we said, okay, can we just take a look at actually your transactions? And it was half of that, technically. So she was shocked. She was like, oh my God, this is crazy. You know

14:40Guest 1: what's funny? We just talked to a chef who said the same thing. Exactly. Even today. Yeah. We asked him, oh, how much is your revenue in general? How much did you make last month? Like, um, I do not know actually. I have no idea. You have

14:52Guest 2: no clue actually.

14:53Guest 1: This just happened like a couple of times. This morning. Yeah.

14:56Guest 2: So that, that kind of triggered a whole thought process. Okay, actually we should be building tools which help them with all these pain points. Right? So that is the first kind of big insight. But then as we came to SF, we started looking for a personal trainer. And I was on this call with this personal trainer. They started asking me questions like, how many times in a week do you need the personal trainer? What's the budget? Uh, you know, XYZ. And it became so clear to us that actually this is not just a problem for private shares, but it's for personal trainers. Oh, by the way, it's also for mechanics. Oh, by the way, it's also for coaches. By the way, it's also for dietitians. Like, it's very big. Like, it's for any solopreneur or small business owner. So that's how the whole transition happened. And that, I believe, is the insight that YC also liked really a lot, Y Combinator. And we got into the batch and our partner Tom just doubled down on that. And we never looked back from there. We have been growing rapidly from there.

15:58Sacha: I was just going to say, you guys are part of YC and congrats on that again. YC is big on talking to customers or users before you even build anything. How did you go? How did you guys go about that? And how did you ensure that this was actually needed and validated the idea?

16:17Guest 2: Yeah, it happened very organically, right? Like we said, Avinash was the first user technically with private chefs. Same with me. As we started talking to more and more private chefs, we understood the challenge. Both of us have this thing that we would not go to bed without talking to at least three users every day. And we still have that, right? So that's a task which we have on our to-do list for every day. And that gets done. Talking to users is massive, right? And just going back to the point about everybody wants a lot of things, right? Like when you talk to them, they'll want a lot of things. But the reality is they actually don't know what they need unless you put it in front of them. And this is exactly what Steve Jobs figured out as well. Not a new insight, but it's really, it's very true. And when we start talking to them, then we know that, okay, this is what they are going through. And this is what they actually need versus what they say, what they want, right? So there's a big difference there.

17:13Guest 1: To add to that, you know, as Ajit mentioned, one is a personal lead. Both me and Ajit had the personal lead. I needed a chef. He needed a chef. On the other hand, we're talking to the first user. I believe your wife, your mother-in-law, they were the first users that we spoke to who also went to the same pain point of being a chef, but not being able to earn enough. Their operations were so chaotic that we were able to help basically streamline everything. And then they were able to earn much more. Yeah.

17:45Guest 2: The other thing that happened is that we witnessed, I would say, like an insane word of mouth. So we started getting so many inbounds for this, right? And our customers started telling other customers. Chefs started telling other chefs. And it was a very clear signal to us that the need is so large. And even one more layer about that, some of the customers started asking, can we invest in your company? Here's a check. And that really gave us like very big conviction that there is a need and it is not something that, you know, that is just for us, basically.

18:21Sacha: I often hear when I talk to founders and entrepreneurs that one of the biggest reasons for startups to fail is not having a good relationship with the co-founder. And I know that you guys take that very seriously. And I'm curious, how did you guys divide responsibilities early on between the two of you? And how is that going for you guys?

18:48Guest 2: Yeah, I think I'll talk a little bit about this because I have been a founder for like nine years. So the first thing I completely agree, by the way, that the most important thing for a startup is a co-founder relationship. And it's not a coincidence that literally every successful founder that we talk to says exactly the same thing. So if you ask them, what is the most successful thing in a startup? Brian Chesky said the same thing. Yesterday, we talked to a YC partner. He said the same thing. Literally, everybody says the same thing. And I 100% believe in that. What is most important is trust, right? So the responsibility is definitely, yes, there is need for the division of responsibility. And there is need for, you know, everything to work. And so talent is a base for the relationship. And every tech startup needs somebody who can really build an awesome product and somebody who can sell the product really well. Right. But that's just a very broad category.

19:50Right. That the foundation is trust. You need to be able to say that, OK, I can trust this person with my life. And you also need to know what is the common direction or journey that you both are going towards and what is important for both of you in life. Right. So if those things are very well aligned, then the co-founder relationship works. Right. And this is the challenge that happens. And this happened with me and my co-founder in the previous company. Right. We didn't really spend much time understanding each other. Honestly, we didn't even meet when we became co-founders, actually. So it is a big mistake. Right. And it became very clear to me that over a period of years that the alignment was the base foundation was not there, basically. Right. And of course, when the base foundation is not there, then the building doesn't work. So that's the way that I think about this. And I think, Avinash, you should talk about how you think about this.

20:48Guest 1: For me, from my perspective, it is my first time being a founder. I've always wanted to start a company, but I never knew what it meant. Right. Again, I would never have done Crave or Cactus if it wasn't for Ajit. Like, there is no way I could have done it.

21:05Guest 2: Same for the other way as well. And

21:06Guest 1: it's the same other way as well, I think. And that's the kind of level of trust and level of understanding we have for each other. Regarding how we divide responsibilities, as Ajit said, we need two kind of people for startups. One is a person who can build. The other is a person who can sell. And both Ajit and I have distinct qualities where I can build and he can easily sell. The add-on to this, which is very unique, I think, is that Ajit always was a programmer from the beginning. We did computer things together. We have known each other since 2006. We have known each other. We are like, you know, college mates and college buddies. And we have the understanding of how you build something. So we are not just selling in the ether. We are not just going and selling something for the sake of selling, but with an understanding of why we are doing it. Right. So that, I think, is the add-on that is making it much more powerful for both of us to, you know, do it right, basically.

22:09Yeah,

22:10Guest 2: and I think just to complete the thought is that, see, there is something to say about accountability versus responsibility. Right. So both of us are responsible for everything. What does it mean? Like, if I need help in selling, Avinash will help me for sure. And if I need, I mean, he just took over customer success last two weeks because we are in YC. We are going through a lot. Right. Like, and he just, he just wanted to do it and he's like crushing it. Right. So similarly, if he needs some help, I will help. But we clearly know who's accountable for what that, okay, if selling is not happening, then it's me. If product is not working, it's Avinash. Like that's, it's as simple as that. So, yeah.

22:53Sacha: I love that. I recently heard that. And Avinash, you mentioned it as well. The companies that work often the best is when co-founders have a lot of admiration for each other. And it sounds like you both have that. So I'm really wishing that this works out. Something that I've read also recently is that people think that lean startups should now hire generalists first. I'm curious to get your take on this.

23:22Guest 2: I think you should answer this being an engineer.

23:24Guest 1: What do you mean by a generalist?

23:27Guest 2: It's a good question.

23:28Sacha: So the way I understood that is that generalist is someone that can help in many different aspects of the business. Meaning coding and what I'm hearing is you also have this ability to not just sell, but also help Avinash with the coding and probably the other way around as well. Right. So generalist meaning being able to do a lot of different things, but also most likely go very deep in one aspect. Right. I'm curious to hear what you guys think about this.

23:58Guest 1: So from my perspective, I kind of agree to it, but there are certain places where you need more deeper skills. For example, in engineering, I don't know if we could hire a generalist who maybe the way to look at it is that you're hiring a generalist within a particular domain. If that makes sense. Right. Yeah. For product side of things for engineering, you hire somebody who is an amazing programmer, right? Who is willing to basically learn if you need to new things and just build. But at the same time, what I have also realized is that you, for in our case, we are building on Ruby on Rails, right? So you need some kind of expertise in this Rails ecosystem. One. Or you need somebody who has built platforms before. At least something. The third thing is, or you have somebody who is willing to go deep and actually learn these things.

24:59So I think those three are important from my perspective today. If you put me in a startup where let's say I need to use a new language, Haskell, let's just take it as an example. Right. I've never built with Haskell before. I've never built with Go before. But if I'm put in an environment where I should use Go programming language and build a web application from ground zero to production, that I can do it today, right? I need that kind of people. Those are the kind of generalists I'm looking for. And yeah, so that's what I feel about generalists in the programming side of things. Now, on the other side, you know, we do love working with people who can wear multiple hats. For example, I love coding, but I'm doing customer success as well, right? So we have people who have, and at some point I'll be doing sales as well. Oh, absolutely. Absolutely. So I'll be learning all aspects or all facets.

26:00Marketing, you know, I love that pieces as well. But there is, as I just said, accountability of, you know, building the product. So in general, yes, I agree to that. But I think there are nuances to that statement, I would say.

26:14Guest 2: Definitely. I mean, I just want to add my thought into this, right? So if you look at the graph of how important experience is versus how important the rate of learning is, right? So in 1990s, the experience mattered a lot, right? Your rate of learning was not that important. And I think in the 2010-15, it was like kind of equalized. But I think we are now really truly finally in the place where I think experience matters way less than how quickly you can learn. Because everything is changing by the week, not even by the month or by the year. Things are changing so fast. But that doesn't mean that you can say that I'm just a generalist, right? I don't think a great startup can be built today without a top 1% engineer, co-founder, CTO, right? I don't think a great startup can be built without a top 1% CEO who has technical background.

27:19I think it's needed, right? Of course, we will learn, right? We will learn everything. But you have to be amazing at one or two skills. That's the way that I think about it, at least.

27:30Sacha: Yeah, fully agree. T-shaped to me seems like the right way to go in the future. Talking about building the product, what parts of the product were hardest to get right? Was it the voice agent? Because I know you guys have that. The scheduling, the payments. Can you give me a little bit of an insight into this?

27:48Guest 1: I don't know if there was a hard thing to build, in my honest opinion. We are not building rocket sense. Let me put it that way, right? We are building a web product, a product that customers will use. We have built features, random features that we have deployed a little bit and saw that there's no value in it and then shelved it. We basically got rid of that feature. I think the hardest piece has been actually finding what exactly is it that our customers need. Again, not what, need. So what is it that our customer, what is it that we can provide to our customers that gives them a lot of value? I think that has been more harder than actually building it because these days with AI, you can literally build anything. People do wipe coding and I think it's not hard to build. We are not building space. We are not launching satellites to moon.

28:50We are building a web tech product. I don't see it as rocket science. I don't know how long it is to put it. We have

28:56Guest 2: the luxury though, right? So we have the luxury of like Avinash having that capability. I would say we're in the top 1% who don't have to worry about the building part of it. But yes, it's truly very hard to figure out what to build. That's the hard part, right? That's the iteration that we keep doing.

29:13Guest 1: But I can answer one thing in that is that, so out of the kind of features that you mentioned, right? Like the voice versus, you know, all of these things. I think the hardest part was finding the right kind of technology to use or partner with. For example, for voice, we are currently using WAPI, right? And it is so good. Again, shameless plug for WAPI. It's really good because what, when we are demoing our product to other YC companies in our batch, I've heard many of them come to us and say individually that we love the way your voice is handling. What are you guys doing, right? And we have seen many companies switch from what they used to use to WAPI. And I'm very proud of that, I would say. What I do is that I'm not building too many complex systems. What is my, I'm kind of lazy. So what is the easiest way to build something? And that's what I do. And I think that has helped us at least so far. Let's see how it, how it translates to in the future.

30:13Sacha: How are you guys thinking about the user experience? I know that this is a big thing for you. And what does Jarvis for small business actually look like day to day? So

30:21Guest 2: right now, as we talk about it, it does three things, right? The first is it helps any solopreneur, small business to have a branded, their own branded website within a few minutes, right? So let's say we want to onboard a private shift. We could do it in the next five minutes as long as they have the content and everything. So that's really helpful for them. Second is a voice AI qualification. What we realized after talking to and working with hundreds of small businesses that two insights. First is they get a lot of inquiries every week and only 10 or 15% of them are actually going to be paying customers. So they waste a lot of time talking to all of them. And the second is that when somebody inquires, you have to get back to them within five minutes. If you don't, they are lost. And that's the hardest part for them, right? So that's why we built the voice AI, which does that, right? Which does both inbound and outbound in terms of inquiries. And of course, we have a booking scheduler, right? So right now, that's the experience that we have for the users.

31:24We are obsessed about, you know, giving the best experience, user experience. Avinash is like, that's the highest priority. And from customer success point of view, we want them to get value for almost from day one, right? So that's something that we really kind of imbibe from Brian Chesky. So we love that.

31:42Sacha: Where does AI stop and good product design take over?

31:47Guest 2: Question.

31:47Guest 1: That's a great question, I would say. What do you mean by like AI stops and design, good product design take over? Like, does it mean like, are you asking like, is it when you're coding or when you're designing something? Where do you see it?

32:03Sacha: So to me is a lot of companies or not a lot of companies, let's say a lot of startups could look like it's a great product. But the reality is it's mostly AI wrappers, right? Or API wrappers. So I'm curious, how do you guys position yourself, you know, and how do you make sure that AI is used in the appropriate place and you use good, you know, product practices where it's actually needed?

32:37Guest 1: Now it makes a lot of sense. I mean, this is awesome. The way you put it is beautiful. To be very honest, I am not a, I don't say not a big fan of, I love AI. We use AI, but every time I think about we are an AI for something, I always think about just as a set wrapper, right? So that's how people potentially are thinking about us as well. But there, I think is a fine line of overusing AI and not actually delivering value. So for me, for us, what that could mean, again, I'm thinking out loud, feel free to jump in and correct me at any time or add to it. So the way I'm looking at it is that, yes, we have AI features, right? For example, this voice AI agent, and we will eventually have text AI agent. We'll have something to handle emails as well for our customers. We'll have something to do booking management as well, right? To automate those. But at the end of the day, we want to give control to our customers. We will give them all the knobs to actually tune it and then, you know, make use of these features that we are building.

33:46Through good product practices, I would say. Not just build as a, again, I think it goes back to that initial conversation. You're not just building it for the sake of building. We're not building it just because we want to build an AI product. We are building it because we want to deliver value to our customers. We are very passionate about, one, building an awesome product. Again, awesome product is very subjective to everybody. Delivering value and empowering small business owners and entrepreneurs, solopreneurs. That's our main goal, right? How do we do that is the most important and not just overusing AI for the sake of it. I think that is the best way I can answer.

34:25Guest 2: Yeah, yeah, absolutely. A hundred percent.

34:27Sacha: Awesome. You guys promised to handle 90% of the admin tasks for solopreneurs and small businesses. What's the 10% that you won't automate and is there a reason why?

34:42Guest 2: This is a great question actually. And this is something that we realized from, you know, our customers. So, every solopreneur and a small business brings a very unique energy. So, if I'm talking to them, right? And you know, okay. So, this is Chef XYZ. They have this catering company for like 30 years. And they're really, their motto is XYZ. And it's a very unique energy, right? You cannot replicate that energy through software. You cannot replicate that energy through a technology, right? And that's the 10% which we don't want to automate. Because that is the soul of your company. And it's usually the founder, right? And that's something that their customers love to interact with. Now, the problem is there is limited, you know, hours. There's like limited bandwidth, right? So, you have to protect that.

35:43And basically, give pieces of that to customers. And of course, to the customers who would, you know, are the right fit for their service. And want to, you know, collaborate. So, that 10% we never want to automate basically.

36:00Sacha: I love that answer. And talking about the soul of an entrepreneur, building a brand myself, I find it extremely interesting, extremely important that everything that I put out is exactly as I want it to be. Yes. How do you guys ensure that using like voice, the voice assistant, that it still represents the brand of, you know, that entrepreneur?

36:23Guest 2: You should go for this.

36:25Guest 1: Yeah. So, the way we are looking at it is that we are not trying to take over or replicate the founder or the soul of that particular, the essence of that particular owner of the company, right? The way we are saying it is that, hey, we are onboarding an AI agent for you. And that AI agent is the assistant that will take, that will help you through at least 90% of your task. And the rest 10%, you know, you handle it, right? And very transparent about this. And that has been helping a lot. Yeah. Now, that being said, we also have gotten many requests for people, from people who say, can you guys just replicate my voice? That's it. Like, I don't, I don't, I don't care. I don't want an AI agent. I don't want, you know, I don't want an assistant. I just want my voice to be replicated. And we are not able to do it. I mean, it's possible to do it, but we are not focusing on that right now. So I think the way our superpower has been that we are being very transparent about how it works.

37:31And I think that helps a lot, you know, in getting trust to the customer.

37:36Guest 2: Yeah. So think about this, right? You're running a business, right? It's much easier to hire an assistant who will do everything for you versus cloning yourself. Right? So we want to be on the assistant side versus the trying to clone, basically.

37:51Sacha: Makes sense. Now, shifting our focus a little bit to YC and fundraising. Why did you guys apply to YC? What were you hoping to get out of it?

38:02Guest 2: Yeah. So I went through YC previously as well. It has been, it was the most pivotal point of my entrepreneurial journey when I went through the batch back in winter 20. YC gives you everything you need as a startup, as an entrepreneur, right? It pushes you towards product market fit. Awesome partners who have done it with thousands of companies. Of course, funding is a big part of it, right? You get to do, they fund you. But apart from that, there's also demo day where you get to talk to thousands of investors on a single day. And I recently talked about this on LinkedIn as well. Like YC gives you something else which is highly underrated, which is belief, right? So every founder needs belief that this is going to work. We are going to make it. This is going to be the next big Airbnb. And YC truly, truly injects that belief in you. That is the reason why we went for YC. And I think we owe a lot to YC. And we truly are massively grateful for that.

39:04Guest 1: Yeah, nothing to add to this. I mean, if the only thing I can say is that when we build the next thing, we would happily go through YC again. That's what I can say.

39:12Sacha: Abhi, what's one thing that YC has taught you that you didn't expect to learn? I

39:17Guest 2: think you should answer this.

39:18Guest 1: That's a great question. I have not really sat down and deeply thought about this. We are just going through the batch. I think as of today, YC has... What YC does best is... It's not... People... I think people think that YC is a school, right? Where they will basically give you a training program or... I don't know what it is, but maybe people think that. But what YC does really amazing is that they give you guidance, but they don't hold your hand and then, you know, kind of take you through everything. They will give you all the material. And at the end of the day, you as founders are responsible to take that information and basically do the best out of it. For example, every YC company has a group partner, right? They have a partner that works closely with you. And I think that is a hidden gem for YC. Paul Graham recently told that you should treat your YC partner as a doctor, right? You should go and talk to them about every single problem you have.

40:21And you should simply blindly implement whatever... I mean, just implement what they are saying because they see a lot of startups going through exactly what you are going through and why you are going through it. And how can you basically get out of it? And this, I think, is completely underrated. And I was really amazed by the partner that we have and how he is guiding us, basically. Let us take a small example of the pivot. Oh, yeah. The name change, right? It was amazing. We applied to YC with the name of Craved. And we got in. We were slowly transitioning to... When we were applied to YC, we were transitioning from the marketplace model to the AI copilot model. But we had not changed the name yet. We're like, okay, we will change it, you know, in a couple of months or weeks. Even before the YC batch started, we had our office hours with our partner. And basically, we got conviction that we have to change the name right away because that's what makes sense for us, right? We have the timeline to change the name.

41:24Within two days, we rebranded everything, found a new name and rebranded. That would not have been possible without YC by the way.

41:30Guest 2: Yeah, absolutely. I think YC teaches you speed and focus, right? I think those two things are... In the batch, you deprioritize everything else other than the company. And the reality is that you need to do that if your startup needs to be successful, right? So this is the reality of it.

41:48Sacha: Yeah, this is great. Great. And again, having this conversation with other people, it seems like focus is the most important thing, right? When you're building. How do you guys ensure that you're staying right on track and focusing on the right thing, you know, at the right time?

42:07Guest 1: Talking to users. Yeah,

42:09Guest 2: I think it's talking to users, getting feedback. And see, the reality is that every startup needs to pick a metric which says that are they going in the right direction or not, right? For us, it is the monthly recurring revenue, which is also the ARR, right? So as long as that goes in the right direction, we know that we're doing the right things. We're building an awesome product, right? And there'll be some lag as well. So it's not just that metric, but also like what impacts those metrics, right? How do we make sure growth is happening? How do you even more important for us retention, right? How is retention happening? Are people happy? So we do Sean Ellis kind of customer surveys. So I would say we have a very strong feedback system in the form of like a company dashboard, which is how we go through. And of course, we regularly talk to our partners. We have some fantastic advisors like Neil, who has been there with us from day one, even day minus one, actually.

43:06Sacha: What do you guys think or what signals do you guys think investors are really looking for in AI tools for SMBs?

43:15Guest 2: Right now, so there is a trend of investors looking for traction, right? And I don't blame them because you hear stories like Lovable, where they've gone from zero to, I don't know, X, 15 million in a few months and things. That's awesome. Good for them. But that's an outlier, right? If you're expecting every company to do that, then I think there is a challenge. And the reality, at least what we have seen so far is that companies take time, right? Airbnb took a long time to get there. And here's a fun fact, which YC people know is that all these companies actually struggle to race during demo day, like even Airbnb, DoorDash, all those companies. So investors are looking for traction, which is fair. But I think YC is doing the right thing. They're looking for who these founders are, what is their mission, right? And are they well equipped to do this? So, yeah, it's a very interesting market.

44:17And I was just reading that non-AI companies are not getting any funding, which is also very, very weird, right? So, yeah, it's a very interesting time.

44:26Guest 1: Yeah, I have a controversial answer to this as well, which is don't build AI tools for or tools in general. If you're optimizing for investors and not talking to users, I think you as a founder have already lost, unfortunately. Yeah. And I think that's what we love about Ajit and me together. We're not optimizing. So all of this, you know, we're not optimizing for investors. We're optimizing because we're enjoying this, because we're talking to our customers, because we are growing organically. We're building a product, talking to the customers. I think that is, I think, highly underrated. If you're trying to optimize for investor with AI, I think you have lost as a founder.

45:11Guest 2: Yeah. Yeah. That's

45:12Guest 1: my contribution. No, it's

45:12Guest 2: a very, very good take. And, you know, the reality is that you know the right investors when, when for the right investors, they will invest in you for the right reasons at the right time. And they, they do believe in you with full conviction. We have, you know, some, one of the best investors in the world are on our cap table already. Right. And they believed in us way before, you know, we were like an AI company or we had like traction. Yes, we have like crazy traction, but they believed in us before that. And they are, they are, many of them are operators, right? They are investors who, they became investors after being founders and operators like us. We, we love those people. That's those, they exist. They will help you out. So don't worry about investors.

46:00Sacha: Slowly closing our conversation. What does success look like for Cactus in a year, three years, five, 10 years?

46:08Guest 2: So I think we have a shared vision, right? Our shared vision is that we want to make tremendous impact in millions of solopreneurs and small business people, right? We are entrepreneurs at heart. We want entrepreneurs to succeed, right? And we have many, many metrics of success, right? We want to be in five continents within five years. We want to have millions of people using this and getting like tremendous successful outcome in their lives and also their customers' lives. And I think the single most defining feature is being that trusted software, which they'll be like, okay, we have Cactus, we are good, right? And Cactus is a symbol of resilience, right? We want them to never fail, always be alive. Also, if you look at our logo, we have like four colors, which mean unconditional love, resilience, persistence, and empowerment, right?

47:08So all of these things mean everything to us. And we want to, we want all our customers and entrepreneurs to feel that basically. That's the end goal. And I would love Avanash.

47:19Guest 1: Yeah. What I want to add to that is we also want to give back. That's very important to both me and Ajit personally for both of us, you know, coming from India, migrating to the US, migrating to Canada, and now being here in San Francisco, we really want more entrepreneurs to be born from the university that we came from. So giving back and then empowering them, I think is also a huge win for us. We are going to, you know, double down on that for sure. I think that's the other thing we want to do. Absolutely. To our university, but also to the society back as much as we can, however we can.

47:59Sacha: I really love that. Question for the both of you. What's one thing that keeps you up at night in AI? And what's one thing that really excites you in AI right now?

48:08Guest 2: I think you should go for this.

48:10Guest 1: It's a question for both of us.

48:11Guest 2: Why don't you take the, which one do you want to take? The night thing or the other one?

48:16Guest 1: Either. You can go.

48:18Guest 2: I'll take the, what keeps me up at night. So what really keeps me up at night is will AI really change our lives to an extent where human relationships won't matter at all? Right? That's very scary to me. For example, can everybody have like an AI boyfriend, girlfriend? And can they live in this virtual AI life where you'd never have to talk to anybody? I think that's extremely scary to me. And, you know, it almost feels like we're going to be in the matrix. Right? So that's scary. For sure. Let's

48:54Guest 1: unplug you right away.

48:55Guest 2: I know. So that's my thing. And then I think you can talk about what's really exciting.

49:00Guest 1: What exciting is today you can, as a solo founder, you know, without having much of technical knowledge, you can build production ready apps very quickly. That is super exciting. I think that's amazing. Right? Again, it's not all the way there, but it's much better than even just one year ago. Forget one year, a couple of months ago. I'm guessing at the rate, at such a rate that in a couple of months, in a year or two, I'm 100% sure that, you know, we'll have many, many more, you know, companies that might run without a technical person. Possible? At least small

49:45Guest 2: businesses. Small businesses.

49:47Guest 1: So I'm excited for that. I would say. Personally, also, it helps me code better, be a better programmer as a sidekick. Vibe coding is a new thing. And yeah, that excites me a lot.

50:04Sacha: Amazing. Are we getting to Vibe businessing? I feel like it's going to be the next big experiment. Guys, I want to give you the floor for any last words, maybe questions that I haven't asked and you wanted to share something specific. So yeah, the floor, the floor is yours.

50:20Guest 2: No, I think this is awesome, Sacha. And, you know, we want to thank you for this opportunity and for talking to us. We believe that, you know, you're going to be an amazing podcast person, host. And thank you so much for having us.

50:36Guest 1: Yeah. And what I want to also add is that if you, who is watching, if you're a solopreneur, a small business owner, or if you know somebody who is a solopreneur, small business owners, please do send them our way. Founders at oncactus.com. Send them to us. Please come try out a product. Call the number that's on our website and you will see what it does. You will see the magic yourself. That's what I would like to add.

51:06Sacha: Absolutely. You guys are amazing. Thank you so much for joining me today. It was a pleasant discussion.

51:11Guest 2: Thank you so much. Thank you so much.