A Product Market Fit Show | Startup Podcast for Founders
00:07:11 · Pablo Srugo, Rahul Shivkumar
[00:07:11] S0: You started Assured with the same co-founder as the company before? Yeah. And so timeline's interesting here. Like you start the last company 2021 where like COVID remote work, you know, so telehealth, right? That's like probably the big thing. And as you're doing it, ChatGPT comes out kind of, I guess, in the tail end, right? So end of 22 and you kind of sold it in 23. Did that, by the way, I'm curious, like that drive at all, the exit, you obviously weren't running an AI native, company. Was that part of the math? How did that play into that and then what you decided to do later? [00:07:39] S1: Yeah. I think a big part of the narrative of starting Dawn was this kind of shift from needing care in a clinic to being able to not just get like continuous care, but even diagnosis and things like that over telehealth, right? That was what happened in 2020 through 2021. What we started seeing in 2023 was actually people wanting to get back to in-person care. You know, they're like, hey, I've done this virtual thing. We want to actually see our provider in person. We want diagnosis in person. We want treatment in person. So I think there was a little bit of headwinds there. The classic public market example of that is Teladoc. If you look at their stock, the pop in 2020, 2021, and then the kind of fall in 23, I think was probably the best signal for what was happening in that market. AI wasn't a huge reason either way for us like exiting or continuing our business. I think it was very nascent in 2023. GPT3 had just come out.
00:07:39 · Rahul Shivkumar, Pablo Srugo
[00:07:39] S1: Yeah. I think a big part of the narrative of starting Dawn was this kind of shift from needing care in a clinic to being able to not just get like continuous care, but even diagnosis and things like that over telehealth, right? That was what happened in 2020 through 2021. What we started seeing in 2023 was actually people wanting to get back to in-person care. You know, they're like, hey, I've done this virtual thing. We want to actually see our provider in person. We want diagnosis in person. We want treatment in person. So I think there was a little bit of headwinds there. The classic public market example of that is Teladoc. If you look at their stock, the pop in 2020, 2021, and then the kind of fall in 23, I think was probably the best signal for what was happening in that market. AI wasn't a huge reason either way for us like exiting or continuing our business. I think it was very nascent in 2023. GPT3 had just come out. [00:08:33] S0: Yeah, we were all scared that AI was going to take over. I guess some people still are, but yeah, that was a big thing back then, right? [00:08:38] S1: I think like it was able to do like summarization well. It was able to generate text somewhat well, but it was like fairly obvious that like something was AI generated and something wasn't, which I don't know if that's true anymore with these newer... [00:08:50] S0: Getting less and less true, man. Opus 5.5 is pretty crazy.
00:08:50 · Pablo Srugo, Rahul Shivkumar
[00:08:50] S0: Getting less and less true, man. Opus 5.5 is pretty crazy. [00:08:53] S1: Oh man, it's getting less and less true for sure. So I think the thesis of Assured was actually built around GPT-3, which was like, hey, if you believe that the next version that's going to come out is going to be a step function from what GPT-2 to 3 was, all of a sudden, the old school like robotic processing automation and some of these other browser automation tools that were built, like it was like pretty obvious that they were going to be far in the past, right? Like the new capabilities were going to be significantly better. So the bet on Assured was, hey, where we had people doing things, you're actually going to have models that can reason, think like a human being and then actually be able to complete tasks end-to-end, right? Without needing a human in the loop or an engineering team to constantly kind of like tweak things, which is what the UiPath and the RPAs and all of those were in the past, yeah. [00:09:44] S0: You sold it in 23, like early 23, mid-23, when was it? [00:09:46] S1: We were mid-23. [00:09:48] S0: And did you have to stay there for an amount of time working after you sold it and you kind of moved on? [00:09:52] S1: We were there for a little bit, mostly through like a transition period, but our plan wasn't to stay on it, yeah.
00:09:52 · Rahul Shivkumar, Pablo Srugo
[00:09:52] S1: We were there for a little bit, mostly through like a transition period, but our plan wasn't to stay on it, yeah. [00:09:58] S0: So what I want to go a bit deeper on is like once you kind of do that transition, whatever, now it's like second half, late 2023, were you like already preset, we're going to obviously start something new just no matter what? Did you guys start thinking, like how did that, how did you get from there to, okay, we're launching a short in March 24?
00:10:13 · Rahul Shivkumar
[00:10:13] S1: Yeah, both Varun and I actually took a little bit of time off and by time off, I mean maybe like three or four weeks where we were like, hey, we're not, we're not going to talk about work or maybe not even like going to like talk to each other that actively about anything other than sports or, you know, nothing work related. And the main reason for that was like, I think both of us wanted to really think through, hey, one, do we actually want to go do another company, right? Like I think the, Dawn was like the, obviously the first company that we did and it was a reasonably like intense like experience, right? So it was like, hey, are we, are we ready to go do this again? Do we maybe want to look at even industries outside of healthcare? Do we want to, you know, like interview for jobs, right? Like I wanted to kind of like have all options open and like truly think through like, hey, what was it that I wanted to do next up? And I think when we reconvened, it was, it was sort of funny that we both came to the same conclusion, which was like, there were a set of problems that I think we understood really well because we did Dawn and it makes sense to like kind of double down on, on knowledge that maybe we had that other people didn't have. It seemed like that there was a huge technology wave that was about to happen. I actually think we were late. It had already kind of like started happening with GPT-3. Like, so all of those led us both to the conclusion that we should stay in healthcare. We should start a company around, you know, this new wave of technology that's enabling value creation at maybe a level we've never, at least I haven't seen in my like career before. It kind of made sense to go do another company.
00:11:36 · Pablo Srugo
[00:11:36] S0: And how do you get from that to the specific problem you chose? What did you do? Like a rigorous, like here are all the problems that we've noticed, which ones we go after? Was it more of a gut instinct that led you to just the specific thing you're solving?