Matthew: Hi, this is Matthew Wise with FounderLY. We empower entrepreneurs to have a voice and share their story with the world, enabling others to learn about building products and starting companies. I’m very excited today because I’m here with Babak Pahlavan, who is the co-founder of Clever Sense. Clever Sense is a platform that is Pandora for the real world. With that said, Babak, we would love for you to give our audience a brief bio.
Babak: Sure, thanks for having me here, I really appreciate it. My name is Babak, I’m the co-founder and CEO of Clever Sense. I’ve always been fascinated by building stuff in general; I’ve always been in the mode of just creating things when there’s nothing. Back in 2000, I immigrated to the United States with my family, and everybody lives in Orange County. I applied for college – most of my cousins went to UCLA and colleges that are in southern California – I got into UC Berkeley where I focused on electrical engineering and computer science.
And then I was supposed to go back to southern California, but then I got into graduate school at Stanford and I said, “OK, I’ll give it another shot for a couple years up here and see what happens.” While I was at graduate school, I ended up being fascinated and getting really enthusiastic about this particular thing that ended up being a company. We’ve been at it since 2008, and we’re just powering through it.
Matthew: What is Clever Sense, what makes it unique? Who’s it for, and why are you so passionate about it?
Babak: Clever Sense is an AI based platform that really finds what matters to the user when you are on the go. Generally speaking, right now we’re at the point that there’s so much noise around us on the web, on mobile phones and what not, that’s it’s just really hard to make decisions, and it’s hard to find things that will match your taste and context – if you want to go get a drink, if you want to find a restaurant or buy a product, anything. And the problem becomes exceptionally harder when you’re on the go, when you’re dealing with screens that are much smaller; you really want somebody else to do the hard thinking for you and then narrow the choices down when you want to do these kind of things..
Fundamentally, we love to do these things with our friends, if it’s Friday night you want to figure out where to go to have dinner; it’s always good to have friends who actually recommend and say, “Yeah, you should definitely check out this new place, it just opened up, it’s going to be match our taste. It’s going to be great.” No one really says it’s going to match our taste, but it’s going to be cool. We’re going to go there, and we’re going to have fun.
The problem exacerbates when you’re traveling and you don’t have friends in the area, especially for business travelers is – I know, I was doing that for a while right between grad school and college – it’s hard to find what matters to you when you’re entering areas that you don’t know at all, or you’re trying to buy a product in the physical world. And you really need this “best friend” who would actually get to know you and understand what matches your taste and matches your context. It would find things that would be relevant to you so you never miss out.
That notion of missing out happens to us all the time because we don’t have somebody, we don’t have these assistants running around and helping us out: you should definitely do this; you should definitely do that or you’re going to miss out. And that’s generally the problem of really having so much information around us in the physical world. The conclusion is we’re definitely at a point where we need to build AI engines to help us out to find what matters to us.
The thing is, I grew up with the Transformers and the Jetsons and all of that, and also I watched a lot of reruns of Knight Rider with KITT, the car. It was one of those things that we had this serene, naive way of thinking about it. What if you just could really have Rosie from the Jetsons follow you around and take care of you? Or, KITT for your car, which would talk to you and say, “Hey, you should really pay attention to this thing you’re about to pass up, you are going to love it if you stop by.”
And the thought process was that we are at a point – with machine learning, AI, and all the data that surrounds us – you should be able to do so much better than what’s the status quo. You should do so much better with that. So, we ended up building this platform that basically can be used for our own products, or eventually developers can use it to get the users to teach him about things that they like. And anywhere they go they never will feel alone again and they’ll never miss out on things. It would do the thinking for them, and it would just find entities and places, or products or events that will match their taste.
Matthew: What are some of the technology and market trends that currently exist in your space, and where do you see things developing in the future?
Babak: Certainly. Generally speaking, it comes down to mobile and essentially connecting people to entities – these entities are places, events or products. The next big trends are figuring out what matters to people without them even searching for it; being quite aware of what would match their taste, your context, and your intentions without asking you to enter in a whole lot of data, without asking you to search or think up the query and what not.
It’s really about having the full understanding of users’ general preferences; what are the things that they like, the information around them and be up to date about that, and the users’ intentions – to predict what would be other things they might be interested in without them expressly telling you over and over again and to be able to learn from those interests. So, if somebody likes these particular sandwich shops, don’t ask it over and over again.
Let it learn from those and be able to save time for the user and connect them to merchants or products or other entities that really match the users’ tastes and also their contexts. That’s certainly the trend that’s going to happen as we go forward. The information needs to get a lot more curated than it is now. Currently, there’s almost no curation when you’re on the go, looking for stuff. That’s generally one.
The second thing that just happened a couple of years ago is that points of interest and the data that’s out there, it’s going to be monetized more and more. It used to be that you have to pay a couple of thousand dollars to InfoUSA and other entities that would actually find this information. Now, there are companies such as Factual and others that would actually give you the data thinking that idea of data is available. Not everybody should be doing these actions of getting the data, curating it, and setting up these points of interest – latitude and longitude and what not – because it’s a waste of energy and time.
The question is: what can be done with this data? If we have all of this data, why don’t we just build something of far more value on top of it? So, we can get this data a lot of times for free, and the question is what can be done with it? And you’re going to see a lot of innovation happening in this space because now the data is far more readily available, and the question is how can you serve and create value for the users when they are on the go?
Certainly a big trend that is going to happen is about the curation and personalization of information when you’re on the go. So it saves you time, it saves merchants and other entities a lot more hassle of finding users that would actually become fans of their business much, much faster. And it creates the unique situation for everybody there in the market.
Matthew: What inspired you to launch Clever Sense? Was there an ‘aha’ moment? How’d you conceive the idea?
Babak: The general issue that we saw was the issue of information overload, trying to find something. You and your buddies in the city want to go have dinner together, and unless you have planned these things before and done enough research and what not, it’s difficult to do when you’re on the go. It’s just really, really hard. These computers are not learning, these smart phones that we carry in our pocket are not learning from us; it’s a very repetitive process and that was very frustrating for us when me and my friends were just going out.
And that was the general hypothesis that this information around us needs to get a lot more curated and a lot more personalized to match our taste and context because the information overload is far more palpable when you’re on the go. It’s a far more daunting task to actually find things that would match your taste.
There is actually a back-story. Even before this, what was really agitating me – which has to do with, again, information overload – was these ValPak coupons that I kept getting in the mail. I couldn’t stop receiving those. So, the question was: I get all this data, I get all these coupons and what not, but then none of them matched me. I don’t know what to do with these things. Then, the oil change ones that actually I could use, I would usually think of them after I’d already done my oil change and it was already expired. It was a clear example of information overload. If somebody, if I had a friend who would actually know what I needed and if I’m passing by the shop or I’m passing by the mechanic shop, they could tell me that they have a coupon that you don’t want to miss out on because it’s a 50% off oil change and what not.
So, originally the thought process started off for personalizing coupons. This is back in 2008 until we met up with Randy Komisar at a Perkins. It was a very friendly meeting, brainstorming over things. Randy Komisar is the CEO of TiVo and Lucas Arts. He’s the guy who really opened up our eyes and said, “If you’re thinking about really personalizing coupons, why don’t you think bigger? Why don’t you think about personalizing and curating everything that matters to business users on the go.” And that’s a far more, bigger scope.
That was the ‘aha’ moment, but also it’s a much, much bigger problem to solve, which ended up causing us to really sit down and do the R&D for the company for two years, from 2008 to 2010. Me and my co-founder, we basically didn’t have weekends or nights, we were so excited about this whole thing and we kept going over and over again until we solved it; we got the technology down and then raised the money, and we built the company.
Matthew: And this was an out-spring from your research in grad school?
Babak: This is something that me and my co-founder were working on at Stanford on our own time. Back in 2008 I was really good at data mining and he was really good at machine learning and AI. The thing was what do we do? We know we’ve got to solve this problem, so do we just want to sit on it or just work at it on nights and weekends? He and I saw so many sunrises because we were just working throughout the night. That’s one of the good things about these schools, such as Stanford. They foster this kind of thinking that if you have this idea and technologically you think you can take a big stab at it, go for it. We’ll support you. We’ll help you out. And my advisor had a lot to do with that, he was a great inspiration and helped us out quite a bit.
Matthew: Who is your co-founder? How did you meet? What qualities were you looking for, and how did you know they’d make a good fit?
Babak: I’m going to write a book about this at some point in the future. Nima Asgharbeygi, he’s my co-founder and CTO. We met up late 2007, beginning of 2008. There’s a club at Stanford called PSA, Persian Student Club, and there’s a division of that which is a business alliance. There was a gathering of people who were trying to think of other ideas for entrepreneurship to foster, to bring more speakers to campus and do more things that would actually encourage more entrepreneurship.
I had done two other smaller companies before and I was talking to a lot of people. It was a room of 25 people at a round table about what do you want to do and if I was going to join a company after graduation and I kept saying, “No, I’ve been working on something, and I am definitely going to do a startup”. It was very interesting, after that meeting was done, this guy which I hadn’t noticed in the room at all, he didn’t say a word; very quiet. He had a very reserved personality. He just came up to me and said, “My name is Nima, I’m really, really good at machine learning, I have a lot of publications and this kind of thing. If you are doing a startup and it has something to do with this, call me. Here’s my number.” And that was it.
Two weeks later, there was a business planning competition on campus. I called him up two days before the deadline and said, “I have this idea that I’m working on. Do you want to do it together and see what happens?” He said, “Yeah, I’m game. Let’s do it.” So, we met up at his office, we worked throughout the night to put together the presentation and submitted it the next morning, and we won the competition. And then we went in front of these VC’s and what not, and that’s literally how it got started.
And we were – knock on wood – just the perfect match with the multi-tasking, doing 20 things at the same time, tend to speak fast and do a lot of things quickly. He’s a reserved, vertical thinker, and he solved absolutely crushingly difficult problems. Personality-wise, he’s almost the brother that I’ve always wanted. Hopefully, we’re going to be doing this for many, many years to come. We already have plans for what we’re going to do in our 40s and 50s, so it’s really cool.