Speaker 1
Hi. It’s John here. I’m in Los Angeles, where it’s 72 and sunny. Like, it is pretty much every day here in spring. I’m coming to you from the RBC Capital Markets Technology Private Company conference. It’s an annual gathering of tech founders and investors exploring ideas, risks and opportunities in tech land. Now, if you’ve been following Disruptors, you know that the last couple of years haven’t been exactly 72 and sunny, at least for tech. Year on year, the tech heavy Nasdaq index is flat, while the S&P 500 is up 25%. But one of the things I love about tech is its ability to look beyond the clouds, to see opportunity. And that was definitely the spirit of this conference. We covered a lot of themes, but here are five ideas I came away with. Number one private for longer. We heard it again and again. Despite the run up in some tech stocks, the mood continues to be that private tech companies will avoid the stock market a while longer. It’s too volatile, too onerous for compliance, too demanding of profits. And given the abundance of private capital, not as necessary as it once was. But that doesn’t mean there isn’t a lot of potential. One fund manager told us that this year feels a lot like the early 2000s, when there was a seven-year bull run just ahead, and some of that next tech boom may come from Low-code No Code, which was my second takeaway. A lot of startups are showing what drop and drag can do to replace old fashioned code writing, and get ready for even more transformation in software with voice to code and image to code. And that will take more data navigation, which was my third takeaway. Data navigation is all about the plumbing of digital networks. In the hyperconnected age of AI. Many organizations are now realizing they can’t be wired to a single cloud. Regulators won’t let them anyway, so there’s a big opportunity for tech firms that are helping us all navigate multiple clouds and see where our data is going. And with that comes a wave of new risks, which was my fourth takeaway. Cyber defense is the new frontier for tech investing, which is both old news and the new thing. Cyber threats are growing exponentially in an AI powered world, and few of us can keep up with the cyber criminals and their AI weapons. But guess what? Generative AI is also on our side, helping companies, governments, and everyone else stay safe. It’s big business and it’s growing. My fifth and final big takeaway is that the next frontier for AI is strategy, because strategy is about solutions, not problems. To be blunt, too much gen AI has been looking for problems to solve rather than accelerating strategies and liberating solutions for all of us. You can read more of my takeaways on tech by searching RBC and Thought Leadership, or on my LinkedIn page. One of the champions of that last idea of AI for growth is Sachin Dev Duggal, who is our featured guest for this episode. Sachin is the CEO and founder of builder AI. It’s an AI powered platform that uses low code and customizable software to provide flexible, bespoke apps at the speed and cost of an off shelf product. In 2023, builder was named one of the world’s top three Most innovative Companies in AI, alongside wait for it, Open AI and DeepMind. So let’s jump right into that conversation. It’s a good one. This is disruptors, an RBC podcast. I’m John Stackhouse. Welcome to the RBC Capital Markets private technology conference in Los Angeles. I’m John Stackhouse and I’m joined by Sachin Dev Duggal, who is the founder and CEO of builder AI. And carries another title, chief wizard. And we’ll, we’ll get to that in a moment. Sachin is a serial entrepreneur. He started his career at the age of 14, building PCs and moved quickly into other aspects of technology by 17. He was already starting to build one of the world’s first automatic currency arbitrage trading systems for Deutsche Bank. Sachin then launched builder AI in 2016, and had a simple aim to simplify the lives of everyday users and make building software as easy as ordering a pizza. Sachin, welcome to disruptors.
Speaker 2
Thank you so much for having me here.
Speaker 1
Let’s start with your story, Sachin. How did you get hooked on technology?
Speaker 2
You know, it was a really funny story. So growing up, I actually had no interest in technology or computers. I wanted to play tennis, and I managed to blow up my mom’s computer and she said, fix it, or else. And so then I had to start reading this book, understanding how to program DOS. And it was really quite addictive. And so I managed to rewrite the DOS menu so that it could now start working. And then it came full circle. Next year it was my birthday. And I said, oh, I want a PC. And she said, you know what happened the last time you touched one? And I said, no, no, but I want to build it. And she said, this is just such a bad idea. In the end, I actually ended up building it and I thought, it’s not so bad and sort of got into building computers. My first customer was my design teacher, and actually for him, I just fixed his laptop and I still got paid. So that was great and full circle. You know, Michael launched Dell, building PCs was no longer as profitable. And so I started to write code and learn how to write software programs. And I guess the rest is history.
Speaker 1
So mothers are great teachers, so is failure. And you had a couple of early failures, which I think allowed you to help doing what you’re doing now, but give us a quick sense of what you learned.
Speaker 2
There’s a natural velocity at which things are meant to run. If you broach that law of physics, things break. So that was a huge learning lesson for me, because I was this really restless teenager where everything should have been done yesterday, could not understand why anything took more than a day, and had no real sort of perception. Some things is like wine that just need time to ferment. I think the second is high the right sort of the people around you. When you’re 21, 22, you kind of don’t know what the right or the wrong person is, and you’re just very trusting that everyone’s the right person. And I still go into discussions when I meet people for the first time with 100% trust, but I’m a little bit more cautious of what to look for. And I think, and this was probably the hardest one, was celebrate the failure and the things that didn’t work. You know, for many years I would just say, hey, you know, that happened. And it was some broken glass and we had an exit and we moved on. And and it was because I was on to my next part of my life. I was quite young when that happened, and I didn’t know how to deal with explaining. Well, you know, I screwed up or this happened or this was the broken glass, because to some extent, I was just like my mind had already gone forward, and I felt that people would look at me differently if I say, oh, what if she didn’t work? And I had this issue and I didn’t know how to explain it? I guess I feel so much more comfortable now because with builder, we built a tremendous business, and I don’t want to ever have the narrative that, oh, you know, there weren’t problems and there weren’t issues because I don’t think it’s fair to the entrepreneur that was in my shoes 20 years back to give them the confidence.
Speaker 1
Tell us about Builder AI and what you’re trying to build.
Speaker 2
For me it’s always the root of the problem that is really interesting, right? And in this case, I was in San Francisco. I was trying to build a photo sharing app, but a really robust backend kept trying to find front end developers, getting rogered and trying to get it done and then realize it’s so difficult. So that was sort of the seed. Started to see more and more people trying to build software businesses or trying to become software entrepreneurs and going through very similar struggles. And then I think it just struck me as saying, well, if we take a long enough view, there’s going to be no more traditional business. Everything is going to be software powered or software native. And if that’s the case, then 95% of the people that are running businesses, running departments, running ideas, running large corporations, they’re ill equipped because they’re not technical, they’re not product managers. And yet they need to be able to use software and build software to unlock their potential. And that’s really what got us started. The Genesis actually was really simple. They said, well, the world’s going to build software. Most of the world doesn’t know how. When you look at what they’re building, they’re quite similar. But what do we mean by quite similar? The ingredients or the features that make up most software applications? They’re the same. 80% of the features that make up most applications are the same time. Just look at the phones and applications you have, right? I’m sure you’ve seen a login view or profile view, a map view, a payment, a chart. They’re all the same features. Like I still don’t understand why there are a thousand people in the world today trying to build a login feature. It’s the same login as the last thousand yesterday. And so we said, well, that’s a lot like Lego. And you know, I have two kids. They play with Lego a lot, and you can use the same Lego blocks to build different things. Why can’t we use A.I. to put those Lego blocks together and organize them? And really, the dream for us is how fast do we get to, we call it 77 apps an hour, which is a million applications a year, which is only 1% of all the new businesses being started every year. And that way, I think we’ve made a small step towards allowing entrepreneurs of all sizes, companies of all sizes and professionals from across domains to be able to unlock the potential.
Speaker 1
We’ll come back to the business model, but tell us first about Chief Wizard.
Speaker 2
So this goes back to my learnings. I was very keen when I was younger. I’m CEO, sort of the brash version of your younger self, and I realized that I actually had none of the experience. The second was the title carries with it like a huge amount of weight. Not so much a weight for me, but a weight for when I speak to folks. It’s too serious. And so you don’t necessarily always find things out. But I say the third thing is, if I really think about what my job is, it’s to help magic happen. And so whether that is magic between teams, magic in product, helping customers unlock potential, which is also quite magical, hence wizard. Plus it gives me something to grow into. Eventually I will become a good CEO and I will carry that title.
Speaker 1
So helping your customers understand or find the wizardry in the economic and business opportunity here is a big challenge for AI. How are you approaching that?
Speaker 2
Maybe the preamble to this question is where exactly are we in the hype cycle? If you remember a Real Player back in the day, it was a promise of video streaming. You’d hit play, you’d wait 10 minutes or 20 minutes, and then you’d get a 32nd clip. How many of you use Netflix or Hulu or Apple TV today? So where’s Real Player? How many of you have a dial up modem today other than sitting in the cupboard or in a museum? Right. You don’t. Right. And and so that’s my point when we’re at this really early evolution of the technology. So that is quite interesting because you have a really early technology that has this massive ability to cause disruption. Then the question is then why is everyone becoming so hyper? It’s been here for about a decade. What changed? So what changed was design, the thing that no one actually talked about. Suddenly you had the most complex software system and it looked like WhatsApp. So the design change suddenly meant everyone reacted like people on the tube were talking about large language models. The world is talking about it. Every boardroom, every CEO saying, where is our AI strategy in many cases? Where is our generative AI strategy? And yet they haven’t found all the use cases. So you have many pilots that aren’t going into production. And this boils down to, to me, the last part, right. Which is what is AI. Because it’s surely not just generative. If artificial intelligence is in some respects trying to emulate human intelligence. We have fundamentally three ways of thinking. I’m sure many of you swam as kids. If you have that visual in your mind, that video that you’re playing, you’re not generating the video, by the way, that’s a real video. You’re retrieving it. That’s a knowledge graph. If I said to you, please help me complete my sentence, what do you think I’m going to say? Right? You might have said, do this. Probably a million other words you could have used, but you inferred say or do. And so that’s inference neural networks. And then if we say, hey, we’re going to write this poem about us on stage, known entities. Now we’re generative. And but the thing is, we had all these other systems that had to come together for us to be able to do that. And I think that’s what is now dawning on CEOs and companies as I meet them is they’re realizing, well, actually, I got to be very specific with the problem I’m trying to solve. You know, today’s AI is really good for removing human variance. It’s really good for removing tasks. It is not removing jobs. It’s allowing people to move more upstream. And my favorite analogy it’s the cape you put on humans to make them super human.
Speaker 1
I get to talk to a lot of CEOs, including about AI, and I find the most successful ones see it as a growth opportunity. This is not about an efficiency play. They’re looking to add productivity, but essentially value per worker per hour worked. That comes down to skills and enhancing skills. You’ve talked about helping people go from good to great. Maybe that’s the superhero cape that you’re putting on them. Give us a sense of how that actually plays out in a real company and in the real economy.
Speaker 2
Broadly speaking, what are company’s most interested in? How do I build a better customer experience in what I’m selling? How do I build a richer customer experience when I’m servicing, and how do I make sure I’ve optimized my cost base so that I’m getting the right leverage to be able to scale? And so this is where AI becomes really powerful. So, you know, imagine a call center. We often think in call centre calls, every conversation is unique but it’s not. I remember my own experience in the sort of rough data points, we ask too many questions last year about customers, but only 1200 were unique. The rest of it was the same question asked in different ways. So that tells you actually the conversations the brand has with its customers aren’t that unique. So you can actually be quite controlled around it. And if you had an AI that could allow an agent to be successful straight from the beginning without having to wait three months for training, that’s really quite powerful. Well, what about the customer experience? This is an example I actually had with the customer last week. They said, I have all these people calling our call center. They’re clueless individuals that are trying to buy an engagement ring. We waste all our time answering the same question. What if we could actually have a conversation with them synthetically through an AI and we talk to them and say, hey, what kind of things does she like? Is she a sporty person? Does she like to go down Beverly Hills wearing Chanel? Like, what is her raison d’etre? And on the back of that, you can then start to say, well, actually, these are the kind of stones you might like. This is the kind of designs you might like. You could even generate by saying, it’s a bit of this and a bit of that and something like this. And so now suddenly you have a really rich, fluid experience to buy something that otherwise was entirely manual, and you had to keep going backwards and forwards. And it was really complex to buy.
Speaker 1
And what’s the role of builder AI in that conversation?
Speaker 2
So we’re helping customers build really that soup to nuts. So whether it’s using Natasha as the conversational platform, whether it’s how we’re allowing you to bring your core insights and your data up, realizing that actually 90% of communication is repetitive, of the 7% that’s actually unique. So figuring out what that seven is and then building the software stack, right. So it can be a conversational experience. It could be a call centre experience. It could be a web or a mobile app experience. Our job is really simple. We’re here to help people build software and do it in a way where it’s really voice to software or text to software, and you don’t need to worry about anything in the middle, as long as you can answer questions and you can describe what’s in your mind. You can build it.
Speaker 1
Tell us a bit more about what that means voice to software or text to software.
Speaker 2
Yeah. So today what we’re seeing in the industry is a lot of tools for developers. Cognition Vercel GitHub Copilot. They’re all about allowing developers to code, complete, or write code more efficiently or faster. And so you’re seeing like 30, 40, 50% performance gains at the developer level. But writing code is not building software. It goes so much before that there’s the design. There’s even further. What are the features needed to solve the problems? What are the journeys needed to solve the problems? Now what can everyone do universally, globally? They can talk, they can chat, and they can explain what’s in their mind and the problem they’re trying to solve. And so the question is how can we use that common user interface? Show you options. Ask you questions. Drive you through a path, and then ultimately start to build a software application with abstracting away all of that complexity.
Speaker 1
Are you actually putting the Lego blocks together, or just giving the Lego set to the customer, and maybe giving them a user guide or one of those maps that they can play with?
Speaker 2
Great question. So no, we’re actually putting the Lego together. We’re customizing the Lego for them. We’re deploying the Lego. So we’re putting it in the showcase. And then we’re saying the lights are on. Now you can invite people home to see the Ferrari you just built. And that’s actually really important because for a customer that’s not technical, which makes up 95% of the audience, they don’t want to be given tools and canvases. No one wants to be given a white page and saying, what do you want to do. When you buy things that you don’t understand, you always buy with options. So I want a bit of this and a bit of that, and I really like this. It’s the pizza analogy. You never go in saying, I would like a pizza and say, this is how much dopugh I want you to put. You say, I want this pizza or this space with these toppings. And that’s generally how we work.
Speaker 1
But don’t customers want a bit of customization or a lot of customization?
Speaker 2
Absolutely. So what we found is at the volumes that we now operate, 80% of the software is out of the box. 20% is customized, sometimes 30%. And that 20 to 30%, we have now built the technology to generate half of it. And then we have an expert network, which is humans in the loop that are doing the last mile. And I don’t think that’s ever going to change. And there’s a very specific reason for that, because there are new things that are being built and you need that human creativity around it. But here’s the flip side of it is today, for every developer being available, there’s eight developers being demanded, nine developers meanwhile. So we just have a shortage of supply anyway.
Speaker 1
You’ve talked about this being a new age of Da Vinci. Explain a bit more your thinking there and where the creator economy is going to take us, but also builder.
Speaker 2
So look, you know, I give you this analogy of imagine, imagine you have a skill, right? And let’s say your skill is you’re a good salesperson on the shop floor. We now know through what we can do with AI, we can empower other people to be really good salespeople because we can see your attribute. We can see how you did it. We can see what you said. We can see how you closed and we can empower everyone else. Now that’s great for the company. It’s not so good for the salesperson because they went from being the best performing salesperson. But now everyone is really good performing. The evolution of skills is actually learning new skills. And that opens up another point, which is from an educational framework perspective, Stem doesn’t work anymore. Kids shouldn’t be learning how to code. What really is important is kids are doing music and architecture and and the arts. And that polymath is why I talk about the da Vinci era, which is for us to now coexist with this technology, to really be the superhumans. We need to open our minds to disciplines that probably never percolated together before.
Speaker 1
At a more practical level, a lot of companies are doing POCs proofs of concept. Do you see that shifting significantly in the next couple of years?
Speaker 2
So I think it goes to my point. You know, you did the video streaming POC when Real Player was here, but it wasn’t until we could stream 4K or we could stream ten ADP that actually it worked. Right. And so the POCs are actually a double entendre in some respects. On one hand they’re testing the technology, but actually what they’re doing is they’re testing the problem. And so what you’re finding is that, well, it’s not really a problem I need to solve, because what I’m still seeing a lot of is a solution seeking a problem. And the POCs are the conversion of the solution, seeking a problem. What we will see over the next 12 months, 24 months is the problem starts to seek a solution because people now understand what the art of the possible is. And that kind of makes sense, right? Because until this wave came along, we never had a concept of what the art of the possible was.
Speaker 1
So shifting more to solution identification and description, what are the best companies you deal with, getting about AI that maybe others are not?
Speaker 2
From a customer perspective, where we have seen this be lightning is when the use case is really well specified. And so we have a customer that says we need our sales teams to be able to answer any question on the back of this data set, and that data set included PDFs. That data set included conversations that had been recorded. And so that was solid, right? Because it was very well defined. We’ve had customers that said, our NPS in the call center is 15, which is not so bad for a call centre, but still terrible. We needed to be 40. And so we said, okay, well, what are the problem? Well, here are the problems we’re seeing, you know, a good agent and a bad agent, we don’t know until six months in. And so now you can use AI, you can use the full gamut from knowledge graphs to machine learning to neural networks to generative to be able to help them solve that problem. So the more defined the problem is where there’s a real business case behind it. And then the third one I was like, is there are folks exploring, but they’re going in with a really open mind on what the future of a user interface might look like. And that makes much more sense when it’s a brand to a B2C audience or a B2B to C audience. And so you’re seeing a lot of people reimagine. What does a shopping app look like? Sephora did this, for example, with makeup, but like, how does it look for trying clothes on? And can we make that whole thing feel like a concierge, almost like a personal shopper experience for people?
Speaker 1
We’ve got just a couple of minutes left. I wonder if we can turn to builder and some of your challenges, but also dreams and ambitions in taking it forward. What are the biggest challenges you’re up against?
Speaker 2
We’ve had the good fortune and the challenge of growing very quickly, so we started the company in 2016. We came out of beta in 2020 during COVID, disastrous time to come out of beta, and we were all nervous. I was very bullish and the board kept saying, Sachin, this is not the time to be optimistic, the world is falling apart. And and we grew from 17 million revenue to almost 180 million revenue in three years. And along the way we put on a lot of weight. And it was akin to having lots of cheeseburgers and ice cream for three years. We’ve now realized we need to go to the gym. It’s a bit painful to go to the gym after having eaten pizza, ice cream, burgers for three years. And so a part of this is trying to figure out what is the right size of this org. Where do we put on weight, what was necessary, what was an issue, where have we not invested enough? The second, I think, and this is why I use the wizard versus CEO, is companies go through story arcs. We’ve completed story arc one. We completed that when we crossed 100 million revenue. We’re now on story arc two. The things that got us to story arc one will not be the things that get us to story arc two. So the biggest challenge is trying to figure that out and unpack that. And, you know, it’s scary because the genie’s out of the bottle, right? And so to a certain extent, you’re trying to learn this precipice. And then this goes back to sort of I think the other part of the question, which is, what’s the plan? If you ask any of the leadership team at builder, we have only one mission in life, which is how fast do we get to a million applications a year? That’s about 90,000 applications a month, 77 applications an hour. The only thing we’re interested in now is what do we need to do to get to 77?
Speaker 1
Where are you now?
Speaker 2
Probably around 3 or 4. So it’s an exponential step change. And that is the excitement right. Because you have to unpack it backwards. And one of the things I learned, and this goes back to challenge and the direction of travel. We did this exercise a year ago and said, what do we need to do to get to 77 apps am hour? And everybody projected today’s problem seven years forward. I said, guys, if we’re solving the same problem that we’re solving this year, in seven years, that’s the clinical definition of insanity, because we’ve woken up every single morning saying we’re going to solve the same problem. So I said, now I want you to think about this differently. I want you to go seven years into the future and write a story. This is what I see around me. This is what my team is doing. And I’m going to move back every half life since then, which are the two story arcs. And suddenly people changed what they thought the problems were. They were now being creative around what the problem could be.
Speaker 1
If we’re here a year from now where will Builder be?
Speaker 2
We will still be delivering dreams around the world.
Speaker 1
And on that journey from 3 to 77.
Speaker 2
Oh, gosh. So we need to grow at 0.26% a day. So I haven’t done the math of that, but hopefully we’ll be close to six.
Speaker 1
And as a Chief Wizards last question, what sort of wizardry do you need to apply over the next year to achieve that?
Speaker 2
You know, I think that that’s probably a lot more personal for me because I love detail and I love understanding detail because it helps me explain the narrative, but it’s become really big. My biggest struggle and the challenge I need to overcome is how can I still be involved in the detail at the right time?
Speaker 1
It’s a great conversation. Really inspiring. Thank you. Saching, thanks for being on disruptor.
Speaker 2
Thank you. Thank you so much.
Speaker 1
I really like Sachin’s take on the future. That is both fabulous for tech, but also not for the faint of heart. Three descriptions will be key to the year ahead. Strategically focused, operationally lean and results oriented. And for those with that approach, the outlook is pretty sunny. This is Disruptors an RBC podcast. I’m John Stackhouse. Talk to you soon.
Speaker 3
Disruptors, an RBC podcast is created by the RBC Thought Leadership Group and does not constitute a recommendation for any organization, product or service. For more disruptors content, visit RBC.com/Disruptors and leave us a five-star rating if you like our show.