WEBVTT NOTE This file was generated by Descript 00:00:01.650 --> 00:00:03.300 Mike Bifulco: Welcome back to APIs you won't hate. 00:00:03.330 --> 00:00:08.520 My name is Mike Biko, one of the co-hosts of the show, and I am sitting down for an interview this afternoon 00:00:08.520 --> 00:00:10.969 with someone who I'm really, really excited to chat about. 00:00:11.070 --> 00:00:15.630 You, you may know if you've been listening to the show long enough for hearing me p along on the internet for long enough that I. 00:00:15.900 --> 00:00:21.810 Worked for a while on the Google Assistant team in the world of voice tech and voice software and the land 00:00:21.810 --> 00:00:24.060 of assistance and iot and automation, all that stuff. 00:00:24.430 --> 00:00:28.140 That is one of the reasons that I'm super excited to, to chat with this founder. 00:00:28.140 --> 00:00:31.414 And the other reason, of course is that Nikhil Gupta is joining me from Wapi. 00:00:31.414 --> 00:00:31.714 He's. 00:00:32.144 --> 00:00:33.704 Also a Y Combinator founder. 00:00:33.704 --> 00:00:37.814 So we're, we're we have many sort of related parallel lives seemingly going on here. 00:00:37.994 --> 00:00:40.035 I'm super excited to talk about what you're building. 00:00:40.084 --> 00:00:46.925 The pitch on the website, the, the H one on your site says Voice AI for developers, which I think paints a lot of story for a lot of people listening. 00:00:46.925 --> 00:00:48.574 But Nikhil, thanks a ton for joining. 00:00:48.624 --> 00:00:49.764 I appreciate you being here today. 00:00:49.764 --> 00:00:50.245 How are you doing? 00:00:50.839 --> 00:00:51.890 Nikhil Gupta: Thanks for having me, Mike. 00:00:51.940 --> 00:00:52.389 Excited 00:00:52.389 --> 00:01:01.089 to chat about Wapi and so curious to hear kind of the stories of voices like Google Assistant and what's happening there Now, everyone but everyone is 00:01:01.089 --> 00:01:04.640 wondering now too, if you have any needs to share with us that would be great. 00:01:04.644 --> 00:01:10.615 And it's super fascinating the, the whole world around us, just like. 00:01:11.005 --> 00:01:12.235 It's gonna change, right? 00:01:12.235 --> 00:01:16.435 Like the way we talk to com, like the way we interact with computers, the, maybe we interact with like microwaves. 00:01:16.465 --> 00:01:19.495 I think that's all about a change in the next like five years. 00:01:19.615 --> 00:01:21.804 I like the fabric of like interaction. 00:01:22.175 --> 00:01:25.325 And I think that was pretty clear to people when they saw that G four oh demo. 00:01:26.005 --> 00:01:26.664 So it 00:01:26.670 --> 00:01:27.655 decided to get into it. 00:01:27.745 --> 00:01:28.134 Yeah. 00:01:28.485 --> 00:01:28.935 Mike Bifulco: of course. 00:01:28.935 --> 00:01:29.115 Yeah. 00:01:29.115 --> 00:01:34.689 So we're, we're a couple weeks out, out from GBT four Oh which is the O stands for O Omni. 00:01:35.564 --> 00:01:36.764 Channel Omnis Omni. 00:01:36.764 --> 00:01:37.994 Yeah, just omni in general. 00:01:38.184 --> 00:01:45.714 So this is the first like multimodal GPT from OpenAI which I think has really changed the way a lot of people are thinking about the software. 00:01:45.984 --> 00:01:51.444 And, and so we're definitely in a changing landscape, especially since, so I, I left Google Assistant two years ago at this point. 00:01:51.664 --> 00:01:53.464 And at that time things were very different. 00:01:53.494 --> 00:01:56.344 But before we get into that, let's start with you here. 00:01:56.344 --> 00:01:59.284 Give, give me first of all the elevator pitch for, for Wapi. 00:01:59.289 --> 00:02:00.365 What's the, the, the story? 00:02:00.365 --> 00:02:01.474 What do you sell it to people as? 00:02:01.904 --> 00:02:02.804 Nikhil Gupta: Yeah, sure. 00:02:03.084 --> 00:02:06.384 The elevator pitch is, yeah, voice effort, developers. 00:02:06.984 --> 00:02:15.934 The you wanna add voice spots to your phones, website apps, we provide like a modular interface that's easy to use, 00:02:15.934 --> 00:02:18.394 easy to configure, and get you up and running very quickly. 00:02:18.994 --> 00:02:25.984 The, the, the, I guess like the, the, the favorite thing for developers is like the modularity of it. 00:02:26.809 --> 00:02:29.689 You don't like open ai, you wanna use Entropic, you can use Entropic. 00:02:29.989 --> 00:02:30.979 You don't like 11 Labs. 00:02:30.979 --> 00:02:31.789 You can use play hd. 00:02:32.209 --> 00:02:39.739 There's like different options for like transcribers hey, you want to like integrate, like make Zapier, like tools, like multi 00:02:39.919 --> 00:02:47.699 assistant, like all that kind of configurability that I think exists a lot out there for like chat you can bring to voice. 00:02:48.089 --> 00:02:49.979 And the future I guess, is pretty clear to people, right? 00:02:50.849 --> 00:02:56.789 When you call into a restaurant, most likely you'll be talking to an ai, and that's kind of what we're building. 00:02:56.954 --> 00:03:05.909 The, the, the bricks and pieces the shovels to help you get that up and running in a day, in a couple hours instead of like months. 00:03:06.854 --> 00:03:07.244 Mike Bifulco: sure. 00:03:07.304 --> 00:03:11.504 Something that would've sounded like an absolute miracle just a few years ago, maybe even a few months ago. 00:03:11.509 --> 00:03:12.109 If you think about it, 00:03:12.609 --> 00:03:18.390 Definitely compelling for, for anyone working in a world where your end users are especially consumer focused. 00:03:18.390 --> 00:03:22.470 I'd imagine there's lots, lots to be said there, but in service industry, that's really interesting. 00:03:22.470 --> 00:03:26.190 And customer service of any sort myriad use cases for that. 00:03:26.370 --> 00:03:27.510 That's, that's super cool. 00:03:27.835 --> 00:03:32.695 Nikhil Gupta: I mean, I think one thing I want to highlight there is like, I think everyone dreads like calling into United, right? 00:03:32.725 --> 00:03:40.286 Like, like at t, like tax like IRS, but I, it's pretty conceivable that we would love calling them in the future. 00:03:41.065 --> 00:03:49.234 Because they, they'll be like, they have like such great ai, they're so patient, they have so knowledgeable and you just call them in and like 00:03:49.234 --> 00:03:51.154 they can immediately answer like, what's happening with their flight? 00:03:51.154 --> 00:03:53.284 Like, can they like cancel, move it around. 00:03:53.795 --> 00:03:57.795 So there is an interesting kind of change there too. 00:03:57.795 --> 00:04:03.114 Like one, the experience improves, but does it also mean like the, the world went from like phone calling to like chat 00:04:03.385 --> 00:04:09.814 because like these, it, it's very expensive for these companies to like host voice experiences because it, it involves humans 00:04:09.814 --> 00:04:12.454 in the background where it's like chat could be more structured. 00:04:12.844 --> 00:04:16.144 And now it's like what is the inherent desire for people? 00:04:16.204 --> 00:04:19.084 They wanna do, use a voice or chat. 00:04:19.384 --> 00:04:24.814 But that I think is also gonna be interesting in the future where you can just call into at t and it'll be like a great experience. 00:04:24.874 --> 00:04:26.314 Which is weird to think, but it. 00:04:29.185 --> 00:04:29.724 Mike Bifulco: Sure. 00:04:29.815 --> 00:04:30.025 Yeah. 00:04:30.025 --> 00:04:33.205 I love the one word you used there, which is really interesting to me, is patience. 00:04:33.235 --> 00:04:38.640 Like having an operator on the phone who's patient is kind of also a pretty big game changer. 00:04:38.879 --> 00:04:43.499 I, I don't think people think of it this way, but usually if you call in customer support somewhere, or even if it's 00:04:43.590 --> 00:04:46.169 a restaurant for example, you call in and speak to someone. 00:04:46.710 --> 00:04:52.439 They're a complicated person with a busy day and probably lots going on, and you're almost certainly not the first chat they've had that day and. 00:04:53.309 --> 00:04:58.474 I would imagine many customer service people have had negative experiences just about every day with someone who's mad about something. 00:04:58.844 --> 00:05:03.074 And the nice thing about building a voice agent programmatically is that they don't get tired. 00:05:03.104 --> 00:05:08.145 They might not experience that, and if they're kind and polite and all that, it really changes the way that people interface with things. 00:05:08.150 --> 00:05:09.049 That's super interesting. 00:05:09.084 --> 00:05:10.389 Nikhil Gupta: I, yeah. 00:05:10.449 --> 00:05:10.839 Yeah. 00:05:11.319 --> 00:05:19.259 And maybe I can talk a bit about, like, you know, right now, if you were to kind of try to create this future maybe you wanna build a company, maybe you wanna 00:05:19.259 --> 00:05:26.569 build, like a small agency or maybe you're a small business, like there's kind of different personas here of like who would, who would wanna use, like why ai? 00:05:27.109 --> 00:05:28.009 There's a couple options, right? 00:05:28.009 --> 00:05:33.739 Like you can take 11 labs deep Graham and open AI off the shelf and like try to switch them together. 00:05:34.229 --> 00:05:38.369 That process will take you at least a couple weeks, like if not months. 00:05:38.729 --> 00:05:47.254 Once you have stitch it together you would have to kind of figure out how to like scale that system because these are what's interesting about like 00:05:47.254 --> 00:05:57.924 multimodal and like, I guess voice specifically as soon as you enter like rounds outside our text is these are stateful long lived jobs and maybe this 00:05:58.374 --> 00:05:59.034 technical center, but. 00:06:01.869 --> 00:06:06.009 When you hear stateful long lived, I think that's like alarm bells go off for people. 00:06:06.009 --> 00:06:07.119 It's like, oh, that sucks. 00:06:07.119 --> 00:06:13.789 Like, it's like, but that's the reality, like where these GPUs are running, which contain the context of the conversation. 00:06:14.269 --> 00:06:20.089 So ultimately when a person is calling, they have to be pinned to something like a resource in the backend. 00:06:20.299 --> 00:06:29.329 So scaling that out as a system is like a very complex undertaking, and that's kind of the value we bring in where if you harken back to like. 00:06:30.049 --> 00:06:31.499 Twilio or Stripe, right? 00:06:31.499 --> 00:06:37.939 Like early days payments could be done where you just had to fill like this ENT form with like Wells Fargo and like you 00:06:37.939 --> 00:06:41.089 could get set up like to accept payments on your website. 00:06:41.929 --> 00:06:45.679 Same for like Twilio, like if you wanted to send a text, like it's totally possible you could. 00:06:46.084 --> 00:06:48.004 Had a contract with like Verizon at and t. 00:06:48.574 --> 00:06:55.204 But Twilio just offered like, here's an API give me the number you wanna call, gimme the text you wanna send them done. 00:06:55.774 --> 00:07:03.934 So that kind of ease of use is what we are trying to bring to like voice AI world where here's the number I wanna call, or 00:07:04.084 --> 00:07:07.984 here's my number and here's the position prompt I wanna use. 00:07:08.344 --> 00:07:08.764 And that's it. 00:07:08.854 --> 00:07:09.124 Done. 00:07:09.566 --> 00:07:13.906 Mike Bifulco: I, what I love about this is that it represents such a big change for, for this world. 00:07:13.906 --> 00:07:19.706 So clearly I'm gonna go pretty, pretty deep and nerdy on the, the things that have changed in voice tech in the past few years. 00:07:19.706 --> 00:07:27.006 But thinking back to the dawn of the first voice assistants the I, I'm reticent to say their names on a recording because 00:07:27.006 --> 00:07:29.226 I'll kick off actions on people's phones around the world. 00:07:29.226 --> 00:07:37.666 But you know, Apple's, assistant and Amazon's and, and Google's are all the initial programming interfaces for those were almost quite literally 00:07:37.666 --> 00:07:44.476 like, imagine a tree structure for the types of conversations you want to have and build that tree for everything anyone could possibly say. 00:07:44.581 --> 00:07:46.261 And have escape hatches for everything. 00:07:46.261 --> 00:07:53.181 And that's changed so dramatically by now that AI is way, way better at picking up with the context of the conversation and inferring things. 00:07:53.181 --> 00:07:56.151 And even like if you stutter into an ai, it won't freak out. 00:07:56.181 --> 00:07:57.891 It can understand that as well. 00:07:57.891 --> 00:08:03.801 It's, it's super interesting and you really are at the dawn of something where people can build very compelling things without 00:08:03.801 --> 00:08:06.651 a lot of work without having to catalog the whole conversation. 00:08:07.396 --> 00:08:07.726 Nikhil Gupta: Yep. 00:08:07.996 --> 00:08:08.236 Yep. 00:08:08.806 --> 00:08:11.836 It's exactly like, that's, I think, the common thing we hear. 00:08:11.836 --> 00:08:19.246 So there's it, it's, it's interesting because like when we started Vai about a year ago, we were doing something else before, and when we 00:08:19.246 --> 00:08:27.131 started Vai, we were expecting that there's a new platform being built kind of the same way that iPhone kind of ushered in this like apps era. 00:08:27.131 --> 00:08:28.961 And then no one predicted like Uber would be the thing that like. 00:08:29.726 --> 00:08:30.626 Was a killer app. 00:08:30.986 --> 00:08:34.976 We were expecting like voices capability, that thing unlocked as a platform. 00:08:35.456 --> 00:08:37.376 There would be like some really cool apps. 00:08:37.766 --> 00:08:46.146 And I still believe that, you know, like there'll be completely crazy apps, infrastructure like applications out there that are voice experience based. 00:08:46.146 --> 00:08:48.066 And I think we are seeing that in learning a little bit. 00:08:48.766 --> 00:08:51.676 They're, they're hitting multimillionaire or I know they're hitting like. 00:08:52.471 --> 00:08:57.901 They're on track to hit like, you know, a billion dollar companies or multi-billion dollar, even ars. 00:08:58.231 --> 00:09:01.921 But that was like our kind of pieces where we wanted to. 00:09:03.121 --> 00:09:09.506 I'll be this kind of infrastructure for this new platform where a lot of like new experiences are gonna be formed around voice. 00:09:09.896 --> 00:09:17.586 But the, to tie in what you were saying, oh, a lot of use cases and surprising amount of like volume, like an insanely 00:09:17.591 --> 00:09:20.106 large volume exists in like the old world of telephony. 00:09:20.796 --> 00:09:21.456 And. 00:09:22.521 --> 00:09:30.531 In like systems that people are built, like people have like teams of like 5,000, like they're designing like these like flows, these conversations. 00:09:30.941 --> 00:09:36.616 And clearly that's like shifting to like prompt, like here's the system prompt. 00:09:37.301 --> 00:09:45.461 And making better bots faster is like a, is a thing that resonates with like this crowd of like, telephony, which we are seeing a 00:09:45.466 --> 00:09:48.326 lot of adoption is I, I feel like, I don't know, I'm like hopeful. 00:09:48.331 --> 00:09:51.026 I really want, like, new experiences, new interfaces. 00:09:51.031 --> 00:09:58.056 That's why we have like, for example, like a Python, SDK, but you can like, set up like a AI toy with, it's not used 00:09:58.636 --> 00:10:03.011 as much as like the polyphony stuff, but like I would love for that stuff to come. 00:10:04.841 --> 00:10:05.756 You know, as things happen. 00:10:06.751 --> 00:10:08.671 Mike Bifulco: Some of this I feel like people need to, I. 00:10:09.391 --> 00:10:14.311 Come to the understanding that there is this like creative space where, where you can build things. 00:10:14.311 --> 00:10:17.021 Now that used to be much harder to do. 00:10:17.026 --> 00:10:23.361 And the fact that someone could pick up a Python, SDK and in a few hours build a pretty impressive demo is like not necessarily obvious. 00:10:23.366 --> 00:10:29.641 Even if you've used chat GPTs, you know, new, new audio features in the interactivity built into that from their app, it, it 00:10:29.641 --> 00:10:32.011 still feels kind of like magic is happening on your device. 00:10:32.011 --> 00:10:33.601 It's, I feel like it's that new, you know. 00:10:34.166 --> 00:10:37.256 Nikhil Gupta: it could be, it could be, it could be the, the, just the newness of it. 00:10:37.376 --> 00:10:42.166 I also wonder like how much of it is like there's a bit of like stasis, I think because open air is moving so fast 00:10:42.556 --> 00:10:45.886 that people are like not sure what to build because AJ is around the corner, right? 00:10:45.886 --> 00:10:49.126 Like, what would be, what even is the point of like building anything. 00:10:49.576 --> 00:10:50.806 It's totally fair. 00:10:51.126 --> 00:10:54.186 But like at the same time, everyone knows there's like so many huge opportunities. 00:10:54.216 --> 00:10:59.666 I'm just like not seeing them being capitalized as I guess as much which is, which is interesting. 00:10:59.701 --> 00:11:00.476 Mike Bifulco: let's take a step back. 00:11:00.626 --> 00:11:06.836 You mentioned a minute ago that, that you're only about a year old into building this product, or you've only been building it for about a year. 00:11:07.046 --> 00:11:10.496 Can you tell me about how Vapi got started, like what you were doing beforehand? 00:11:10.496 --> 00:11:12.956 Where and, and what was the genesis of all this? 00:11:13.266 --> 00:11:13.686 Nikhil Gupta: Yeah. 00:11:13.686 --> 00:11:13.956 Yeah. 00:11:13.986 --> 00:11:18.466 So we were building ai, meaning note taker and it was called Silver Power. 00:11:19.456 --> 00:11:20.146 It was great. 00:11:20.476 --> 00:11:27.706 Hit profitability and we kind of realized the next milestone for us was like to get it to like 10 mil error, which is the 00:11:27.706 --> 00:11:31.136 usual, you know, milestone when you hit that kind of number. 00:11:31.886 --> 00:11:35.756 And we just, like, we didn't wanna spend our twenties like selling meeting software. 00:11:36.776 --> 00:11:37.016 It 00:11:37.016 --> 00:11:37.346 was. 00:11:38.126 --> 00:11:42.296 Not the most challenging not the most like forward-looking impactful thing we could be doing. 00:11:42.906 --> 00:11:50.406 We just felt like there's like a bigger technical challenge we could take on and that like we were engineers and like, we just love technical problems. 00:11:50.766 --> 00:11:53.976 So started stumbling around on like finding what we could do. 00:11:54.316 --> 00:11:59.476 The, this my co-founder one day in his. 00:11:59.836 --> 00:12:08.286 Depression of like pilot pivot, depression made this thing because we over like like over the course of like our startup, we have spent 00:12:09.666 --> 00:12:12.816 around like 30 to 50 KI think on like coaching, like founder coaching. 00:12:13.186 --> 00:12:13.336 We 00:12:13.336 --> 00:12:17.566 started at the company when we were like right outta college immature. 00:12:18.526 --> 00:12:28.036 Didn't know how to work with emotions, communicate the, the usual thing of like founder complex and worked with a coach who was saved our company. 00:12:28.486 --> 00:12:32.566 Like he helped us articulate what we need from each other and 00:12:32.986 --> 00:12:38.416 worked through ourselves and our own mind, and that that coaching was transformative. 00:12:38.416 --> 00:12:39.016 So transformative. 00:12:39.016 --> 00:12:47.146 We were like, oh, that, like when, so my co-founder Jordan, he created that same experience basically with an ai where that AI could like help him. 00:12:47.761 --> 00:12:51.571 Reflect back his thoughts, his emotions, kinda like a Yeah. 00:12:51.661 --> 00:12:52.231 Therapist. 00:12:52.571 --> 00:12:54.731 The experience got better and better. 00:12:54.971 --> 00:13:01.061 But ultimately we realized there's still like, like the company that's focusing, that's focusing on like, making this experience better. 00:13:01.541 --> 00:13:03.791 It needs to be that needs to be the company itself. 00:13:04.151 --> 00:13:09.701 Like we cannot take out the problem of like making a great voice experience and making it like a great therapy. 00:13:10.436 --> 00:13:12.196 But, those have to be separate companies. 00:13:12.196 --> 00:13:17.206 And we are like, oh, that's actually kinda great because we are, we love technical, we are technical, we 00:13:17.206 --> 00:13:20.216 love technical problems, we love technical customers. 00:13:20.526 --> 00:13:23.836 Like we, we just wanna work with startups and like really technical people. 00:13:24.406 --> 00:13:31.866 That it was just like a perfect alignment where oh yeah, like this is, this feels right, like making APIs that feels right for people. 00:13:32.346 --> 00:13:33.886 And that's how we got into it. 00:13:34.841 --> 00:13:35.351 Mike Bifulco: sure. 00:13:35.771 --> 00:13:36.131 Yeah. 00:13:36.131 --> 00:13:43.601 I think, I think especially if you've had some coaching around like trying to reach scale and get to a point where if, if at that point you had taken 00:13:43.601 --> 00:13:47.621 on investment, like you really need to tackle some gargantuan problems. 00:13:47.871 --> 00:13:51.771 And to your point before about not wanting to spend your twenties selling meeting software like. 00:13:52.021 --> 00:13:56.851 You would need to sell all of the meeting software to really hit the scale that a lot of investors are looking for. 00:13:57.101 --> 00:13:58.061 And that has happened, right? 00:13:58.061 --> 00:14:03.021 Zoom certainly had a, a bit of a rocket launch over the past few years, but you know, how many times can that happen? 00:14:03.021 --> 00:14:05.901 And almost the meta problem here is, is also really interesting too. 00:14:06.291 --> 00:14:07.611 Yeah, that, that's really fantastic. 00:14:07.616 --> 00:14:07.641 I. 00:14:08.341 --> 00:14:09.241 Nikhil Gupta: Yeah, you're right. 00:14:09.241 --> 00:14:14.581 Like there's like, you know, something about like the space you're in which limits, like, what you can do as a feeling. 00:14:14.921 --> 00:14:20.466 I, I do find like, I, I, I go between my, in my mind between, like, I think the, the tam is never the market. 00:14:20.646 --> 00:14:22.806 The tam is the founder, which is like, 00:14:23.226 --> 00:14:29.376 but you wanna just keep finding new ways to like, go, like expand Tesla, you know, like robotics now it's like, it says random as hell. 00:14:29.776 --> 00:14:31.906 Like SpaceX became starlink. 00:14:32.776 --> 00:14:33.346 Cool. 00:14:33.646 --> 00:14:41.656 So I, I do find like the tan isol as a founder, but yeah, like focusing, like for us it was just like, we wanna spend our time solving hard 00:14:41.656 --> 00:14:45.266 technical problems and offering great experiences for developers. 00:14:45.868 --> 00:14:47.548 Mike Bifulco: There's a lot of wisdom in your words. 00:14:47.858 --> 00:14:52.208 It sounds like you, you've probably been through, like ingested the coaching in a way that was really healthy and 00:14:52.208 --> 00:14:55.088 have lived through a, a hard pivot with another founder. 00:14:55.138 --> 00:14:56.368 Having been down this. 00:14:56.368 --> 00:14:57.718 Startup wrote a few times myself. 00:14:57.718 --> 00:15:03.618 I've had the good, bad and middle experiences of, of you know, many lifetimes I feel like wrapped up in building startups. 00:15:03.688 --> 00:15:08.998 It's always interesting to hear someone who, who acknowledges that challenge as part of building the, the company too. 00:15:09.058 --> 00:15:11.998 That's, that is a big thing that we don't talk about a lot in general, I think. 00:15:13.238 --> 00:15:13.958 Nikhil Gupta: Oh yeah. 00:15:14.048 --> 00:15:19.388 I mean the, we've been trying for like four and a half years now. 00:15:19.478 --> 00:15:20.108 Like at it, 00:15:20.448 --> 00:15:21.438 We were Winter 21. 00:15:21.563 --> 00:15:22.093 Mike Bifulco: Good for you. 00:15:23.063 --> 00:15:30.108 Nikhil Gupta: It is kind of, I, I, I mean, I think there's something to, like, something to, like, I think PJ and Michael, 00:15:30.108 --> 00:15:38.188 lets to say this where pg where it's like what sometimes like founders are founders because they don't have anything else to do. 00:15:38.248 --> 00:15:39.658 Like I, I don't know what else I would do. 00:15:41.388 --> 00:15:41.778 Mike Bifulco: Yeah. 00:15:41.828 --> 00:15:42.273 Agreed. 00:15:42.733 --> 00:15:46.663 Nikhil Gupta: yeah, so, so we just gotta figure it out. 00:15:47.023 --> 00:15:47.683 That's, we 00:15:47.683 --> 00:15:51.163 had another choice but to figure it out and that's where it's been a lot of fun. 00:15:51.473 --> 00:15:58.073 So I think for people, I guess kind of wondering, like develop, I'm guessing obviously very technical, wondering like what to do. 00:15:58.563 --> 00:16:05.378 I, it is, I guess like the, the, the seed, the seed of like, that feels so good. 00:16:06.788 --> 00:16:10.748 I know I'm spending my time serving people I like a lot. 00:16:11.198 --> 00:16:13.898 And that is enjoyable in itself, like no matter what happens to Wapi. 00:16:14.328 --> 00:16:20.848 So that like the feed of advice that people give around, do keep doing the things you really like, that you feel like especially passionate about. 00:16:20.853 --> 00:16:25.018 And like that the moat or competitive advantage will come from that because you 00:16:25.018 --> 00:16:27.808 just have like so much experience, like so much love for that thing. 00:16:28.218 --> 00:16:29.838 So much love and serving that thing. 00:16:30.408 --> 00:16:31.008 So that's been 00:16:31.118 --> 00:16:36.518 Mike Bifulco: It has to be a, a problem you can care about for 10 years, you know, not just a, a 18 months or something like 00:16:36.798 --> 00:16:37.938 Nikhil Gupta: yeah, a hundred percent. 00:16:37.938 --> 00:16:38.328 And like. 00:16:39.648 --> 00:16:46.128 Like the, when I think about like, the common thing I hear, and maybe this could be a bubble, is that people love our API. 00:16:46.728 --> 00:16:47.298 It's like people 00:16:47.298 --> 00:16:48.738 love, like the design of it. 00:16:48.738 --> 00:16:49.938 It's like it's easy to use. 00:16:49.943 --> 00:16:50.778 It makes sense. 00:16:51.078 --> 00:16:52.188 The validation is great. 00:16:52.458 --> 00:16:57.603 And that's just like, I'm like really in know about like the little details where like it 00:16:57.603 --> 00:16:58.803 has to be like, for example, like. 00:16:59.733 --> 00:17:00.753 Singular, not plural. 00:17:00.753 --> 00:17:03.213 Like it has to be like consistent in this format. 00:17:03.243 --> 00:17:06.273 There has to be like the same crud interface to all of these things. 00:17:06.273 --> 00:17:07.083 It has to be a noun. 00:17:07.293 --> 00:17:12.033 Like all these best practices that I'm sure people that I'm talking to here know about. 00:17:12.453 --> 00:17:18.783 But it's like, you know, it's surprising to me when I look at other APIs, it's like, just like, what was the design process here? 00:17:18.783 --> 00:17:20.373 Like, what the fuck was going on in that head? 00:17:20.373 --> 00:17:24.758 Like making this interface like, like makes no sense to me. 00:17:24.763 --> 00:17:25.148 Where. 00:17:25.778 --> 00:17:31.178 Having this opinion of like a good developer experience, like really caring about that gives me joy. 00:17:31.388 --> 00:17:31.898 So that 00:17:32.198 --> 00:17:32.948 I've been a lot of fun 00:17:33.488 --> 00:17:34.088 building. 00:17:34.573 --> 00:17:38.683 Mike Bifulco: you and I are just meeting for the first time, but I came into the world of APIs by way of studying 00:17:38.893 --> 00:17:41.333 user user experience and human computer interaction. 00:17:41.613 --> 00:17:46.023 And so being, being deeply embedded in the world of design, one of the things that I often say to people is that 00:17:46.203 --> 00:17:48.783 learning design is interesting because anyone can learn it. 00:17:48.783 --> 00:17:50.073 It's not something you're born with, 00:17:50.373 --> 00:17:53.823 but the curse of it is as you learn these rules and as you learn the, the. 00:17:53.973 --> 00:17:55.203 Things that break the rules. 00:17:55.293 --> 00:17:56.853 You see it everywhere, right? 00:17:56.853 --> 00:18:04.083 So like by focusing on building an API that follows all these rules and uses plurals correctly and is thoughtful about the way interfaces are presented, you 00:18:04.083 --> 00:18:07.563 start to see all of the inadequacies of your own product, also other products. 00:18:07.563 --> 00:18:11.493 And they become more and more like little red irritations across your life. 00:18:11.498 --> 00:18:11.673 And, 00:18:12.128 --> 00:18:12.418 Nikhil Gupta: Yeah. 00:18:12.453 --> 00:18:13.533 Mike Bifulco: it's a blessing and a curse I think. 00:18:14.263 --> 00:18:16.153 Nikhil Gupta: What's your favorite API you think out there? 00:18:16.768 --> 00:18:21.228 Mike Bifulco: So I'm, I'm definitely biased here because I also used to work at Stripe, so I was on the Derell team there. 00:18:21.283 --> 00:18:22.003 Nikhil Gupta: Yes. 00:18:22.063 --> 00:18:22.663 Fair. 00:18:22.878 --> 00:18:25.608 Mike Bifulco: obviously Stripe is the gold standard in many ways to work with. 00:18:25.918 --> 00:18:31.888 And I've, I've, you know, through the course of my career, especially working in developer advocacy for a while, played with many APIs, small 00:18:31.888 --> 00:18:35.468 and large to kind of build demos and show people how to do things and. 00:18:36.458 --> 00:18:38.618 I think a a So Stripes, API is wonderful. 00:18:38.618 --> 00:18:39.998 The documentation is incredible. 00:18:39.998 --> 00:18:45.698 The team is top notch, like super cool to see, but there are also smaller companies making things that are really interesting to use as well. 00:18:46.008 --> 00:18:51.003 Lately I've been looking at weather APIs for various purposes for my current company for craft 00:18:51.283 --> 00:18:52.289 Nikhil Gupta: Yeah, yeah, yeah. 00:18:52.353 --> 00:18:57.603 Mike Bifulco: the most thoughtful APIs that I encounter are ones where I don't have to become a meteorologist to understand how to use it. 00:18:57.943 --> 00:18:59.623 And that's, that should be the bar for everything. 00:18:59.623 --> 00:19:03.748 Like my, my lowly self should be able to get to your API and understand what's going on, you know? 00:19:04.088 --> 00:19:05.168 Nikhil Gupta: Yes, yes. 00:19:05.228 --> 00:19:06.068 Yeah. 00:19:06.068 --> 00:19:06.069 Yeah. 00:19:06.223 --> 00:19:07.658 And like yeah, like that's. 00:19:09.053 --> 00:19:10.583 Also would love to hear from people. 00:19:10.583 --> 00:19:14.163 It's like, like I always wonder like when people show up, like like we have a swagger, right? 00:19:14.163 --> 00:19:14.913 Like they show up to this. 00:19:15.243 --> 00:19:16.293 Does everything just make sense? 00:19:16.293 --> 00:19:17.013 Like it just, if you 00:19:17.013 --> 00:19:21.243 just to look at the top level concepts that exist in the API, does everything come together? 00:19:21.643 --> 00:19:21.823 Yeah. 00:19:21.828 --> 00:19:28.393 Strip is great because you don't strive is very painful, but it rewards you like you go through the pain 00:19:28.783 --> 00:19:30.793 and because it's so complex. 00:19:31.063 --> 00:19:33.433 But everything kind of does make sense. 00:19:33.943 --> 00:19:35.053 I think that's like where. 00:19:35.923 --> 00:19:42.983 A lot of like teams like lose, I guess, like you, the more complexity you add can everything still seem coherent? 00:19:43.353 --> 00:19:44.883 Like as, as you learn? 00:19:45.303 --> 00:19:49.803 Does the world, does the world of your, does your world make more sense or does they make less sense because there's 00:19:49.833 --> 00:19:52.803 contradictions everywhere, like different ways to do things? 00:19:53.193 --> 00:19:59.043 Or is it just like there's one way to do thing and like, yes, it's very complex, but like if you learn everything, it'll make sense. 00:19:59.943 --> 00:20:00.363 Mike Bifulco: Totally. 00:20:00.408 --> 00:20:01.578 Nikhil Gupta: does a great job. 00:20:01.638 --> 00:20:02.928 Great job of that for sure. 00:20:03.378 --> 00:20:11.198 I, I think there's something about like probably like, I don't know if like AWS counts, but I think a Ws 00:20:11.198 --> 00:20:16.248 is like is also kind of the same thing where, pretty complex, like very hard to work with. 00:20:16.498 --> 00:20:19.588 But like once you learn it, it's pretty okay. 00:20:19.678 --> 00:20:20.968 Everything makes sense. 00:20:21.568 --> 00:20:24.108 Actually, you know what my favorite API is Kubernetes. 00:20:24.498 --> 00:20:25.818 That's, that's Kubernetes? 00:20:26.208 --> 00:20:26.838 Yes. 00:20:27.048 --> 00:20:27.708 Kubernetes. 00:20:28.168 --> 00:20:30.508 I, I might even say I like it better than Stripe. 00:20:30.838 --> 00:20:31.678 Mike Bifulco: Oh, that's interesting. 00:20:32.068 --> 00:20:33.448 That's a new take for me, for sure. 00:20:33.448 --> 00:20:33.538 Yeah. 00:20:34.283 --> 00:20:40.643 Nikhil Gupta: It is, I find like the documentation when you read through it, like they, they read your mind. 00:20:40.673 --> 00:20:43.313 I think maybe that the problem is that Stripe has too many customers. 00:20:44.658 --> 00:20:44.778 Mike Bifulco: Hmm. 00:20:44.933 --> 00:20:52.553 Nikhil Gupta: So when, when you, when you enter Stripe documentation, like they're trying to walk you through how to handle like subscriptions, 00:20:52.553 --> 00:20:54.653 they're trying to walk you through how to handle the invoicing. 00:20:55.013 --> 00:20:57.233 They're trying to walk you through how to handle usage based billing. 00:20:57.773 --> 00:21:00.293 And they try their best and it's great. 00:21:01.148 --> 00:21:04.598 Kubernetes, you enter the doc and they'd already know what you're trying to do. 00:21:04.988 --> 00:21:05.708 They know that you're 00:21:05.713 --> 00:21:10.388 trying to deploy something that can scale, so they talk you through different concepts. 00:21:10.388 --> 00:21:14.018 Like the first thing they'll talk about is here's pods, here's services. 00:21:14.528 --> 00:21:19.028 If that's, it's just the documentation gives me joy reading it, 00:21:19.298 --> 00:21:22.628 like I, it's a conversation with like, the person who wrote it is 00:21:22.883 --> 00:21:23.043 awesome. 00:21:23.528 --> 00:21:26.173 Mike Bifulco: I, I wonder if you'll think of this as a hot take. 00:21:26.393 --> 00:21:32.243 But I, I see Kubernetes documentation and the, the way that they describe things in stark contrast to the way that 00:21:32.243 --> 00:21:35.363 Docker's developer experience is and was for a very long time. 00:21:35.733 --> 00:21:41.943 My, my feeling on Docker and figuring out how to make Docker work for, for a really, really long time was Docker can't tell 00:21:41.943 --> 00:21:45.273 you what Docker is, and you have to figure it out by trial and 00:21:45.308 --> 00:21:45.528 Nikhil Gupta: No. 00:21:45.973 --> 00:21:50.983 Yeah, I think Doctor has excellent like dev interface, like, you know. 00:21:51.933 --> 00:21:52.203 Mike Bifulco: Yeah. 00:21:52.288 --> 00:21:52.578 Yeah. 00:21:52.888 --> 00:21:55.318 Nikhil Gupta: Docker form makes it really easy to like, get things up and running. 00:21:55.318 --> 00:21:57.598 Like Kubernetes is harder to get up and running. 00:21:57.958 --> 00:22:05.458 But yeah, like, and Docker logs for example, like everything sort of works, but yeah, like it's, they have no clue what they're doing. 00:22:05.518 --> 00:22:07.198 It's kind of the vibe that you get for sure. 00:22:07.463 --> 00:22:07.943 Mike Bifulco: Right. 00:22:07.943 --> 00:22:08.153 Yeah. 00:22:08.153 --> 00:22:12.953 You, you kind of have to know that Docker compose and docker's form exist to be able to figure out what they do. 00:22:12.953 --> 00:22:16.583 And it's not obvious what either does, you know, from the name alone. 00:22:16.873 --> 00:22:19.653 So let's talk about your, your use case. 00:22:19.653 --> 00:22:21.873 So you've, you've spent all this time building a, a great API. 00:22:21.873 --> 00:22:23.253 What is it like to use wapi? 00:22:23.253 --> 00:22:24.983 Like, what, what does hella world look like? 00:22:25.413 --> 00:22:25.653 Nikhil Gupta: Yeah. 00:22:25.653 --> 00:22:28.803 The hello world is and I'm curious to get your thoughts on it. 00:22:28.863 --> 00:22:32.163 The hello world is you go to a dashboard, you don't even touch the API first, right? 00:22:32.453 --> 00:22:38.573 You go into the dashboard and the first thing you'll see is like and prompt to create an assistant. 00:22:39.533 --> 00:22:42.948 And you, you select like there's some templates options there. 00:22:42.948 --> 00:22:47.688 Like, Hey, I wanna make like a game NPC, or I wanna make like a customer service. 00:22:47.688 --> 00:22:48.828 I wanna make a, on my scheduling. 00:22:49.188 --> 00:22:56.178 You click one of those assistants, you get a template, and after that one click you should be able to, there's a 00:22:56.178 --> 00:22:58.698 button, right, that will show which like talk to assistant. 00:22:58.988 --> 00:23:00.938 And that is like kinda the hella world there. 00:23:00.938 --> 00:23:02.798 It's like you have quit an assistant that you like. 00:23:03.368 --> 00:23:04.268 And you can talk to it. 00:23:04.658 --> 00:23:09.068 And the, what you'll see front, right front and the center is like, what's the configuration? 00:23:09.308 --> 00:23:10.388 Configuration of the assistant. 00:23:11.168 --> 00:23:14.768 It's like, oh, like my name is not Domino's. 00:23:14.768 --> 00:23:16.298 My name is like Pizza Hut. 00:23:16.688 --> 00:23:16.928 You know? 00:23:16.933 --> 00:23:18.608 It's like, welcome to Pizza Hut. 00:23:18.728 --> 00:23:19.418 Like that's changing 00:23:19.423 --> 00:23:25.553 that, that, that should give you the first like satisfaction of like putting, cutting the system and being able to talk to it. 00:23:26.183 --> 00:23:29.693 Then you, you know, the dev experience, which is like, okay, where do you wanna put this assistant? 00:23:29.693 --> 00:23:31.193 Do you wanna put it on a phone number? 00:23:31.193 --> 00:23:32.783 Do you wanna put it on your website? 00:23:32.783 --> 00:23:33.833 Do you wanna put it in your app? 00:23:34.413 --> 00:23:38.103 I guess like that's where APIs and the experience, like dev experience enter. 00:23:38.503 --> 00:23:43.238 I haven't figured out the hello world for it, but that then you would go to our documentation and then they're kind of like. 00:23:44.378 --> 00:23:45.668 It's like, you know, choose your own journey. 00:23:46.028 --> 00:23:48.128 It's like I'm gonna create a phone number. 00:23:48.398 --> 00:23:51.878 Then you go to this doc, you wanna create like a put on a website, go to this doc. 00:23:52.248 --> 00:23:57.648 Mike Bifulco: I think it's an interesting case because there's a lot to be said for being able to just poke around with a thing 00:23:57.648 --> 00:24:00.198 and try it, especially with the case of what you're building. 00:24:00.258 --> 00:24:01.488 Kubernetes is a great example. 00:24:01.518 --> 00:24:07.218 If I know that when I get the API calls correct for Kubernetes, that I'm going to have something that does all this nice orchestration 00:24:07.218 --> 00:24:10.728 for me, and the, the outcomes of that are clear from the onset. 00:24:11.068 --> 00:24:13.168 But for the product you're building, the outcome is like. 00:24:13.793 --> 00:24:19.963 A little more ambiguous and perhaps less clear in that, okay, if you execute these calls correctly, you now have a voice service. 00:24:20.383 --> 00:24:21.373 Is that any good? 00:24:21.433 --> 00:24:21.673 Right? 00:24:21.673 --> 00:24:25.243 Like the first thing you need to sell is, is the outcome of all this dev work going to be valuable? 00:24:25.483 --> 00:24:29.503 And I think inverting the experience in that sense is honestly probably a really good choice. 00:24:29.863 --> 00:24:30.913 Nikhil Gupta: That's like smart. 00:24:31.183 --> 00:24:31.393 Yeah. 00:24:31.393 --> 00:24:32.623 That's pretty smart observation. 00:24:32.653 --> 00:24:33.943 I've never thought about it that way. 00:24:34.133 --> 00:24:40.553 All I know is like as a developer, I think the thing that probably like now I know like I have to do Stripe, right? 00:24:40.883 --> 00:24:46.853 But like, I, I, wish Stripe just gave me like more joy, like before I even get started, like, do I, will 00:24:46.853 --> 00:24:47.723 stripe even work for me? 00:24:48.143 --> 00:24:52.853 I think it takes me a long time to figure that out, that answer to that question for most dev tooling. 00:24:53.303 --> 00:24:55.673 Whereas I think we are trying to answer the question like, yeah. 00:24:56.108 --> 00:24:57.008 Bobby, will this work for you? 00:24:57.008 --> 00:25:02.078 This is, here's a assistant, and then it's your, the, the job is to just figure out how to like, get that assistant in the 00:25:02.078 --> 00:25:04.988 right place that you need, which we have all the AP ideas for. 00:25:05.408 --> 00:25:14.843 But that I, I think about the law too where like, it's like, what is like in my, like where dreams, like what would I love as a developer from this experience? 00:25:14.843 --> 00:25:19.433 You know, it's like, what is like the, the, the most beautiful, amazing experience coming into Wapi? 00:25:19.773 --> 00:25:21.243 I don't have an answer to that question, but like, yeah. 00:25:21.248 --> 00:25:23.313 The hello world is don't code. 00:25:23.613 --> 00:25:24.153 Just go in, 00:25:24.543 --> 00:25:28.413 try something and then if it looks good, get, get, get cracking. 00:25:29.783 --> 00:25:33.233 Mike Bifulco: I think it's pretty compelling, and I, especially for this audience that we're talking to. 00:25:33.233 --> 00:25:37.973 So for API developers, you get two things out of the box before writing your first line of code. 00:25:37.973 --> 00:25:39.863 One is you can go and try the thing, right? 00:25:39.868 --> 00:25:42.773 You, you go to the dashboard, do that, do your own experience. 00:25:43.023 --> 00:25:47.493 But naturally, I think most of the people listening to the show will also go to your API. 00:25:47.538 --> 00:25:48.438 Reference, right? 00:25:48.438 --> 00:25:50.148 Go to the docs themselves and see what's there. 00:25:50.148 --> 00:25:54.108 And like, there's a lot of good, juicy stuff in there that's, that in itself is exciting. 00:25:54.108 --> 00:25:57.918 And you can kind of see some of the thought and, and usability that comes outta that straight away. 00:25:58.308 --> 00:26:04.673 And to your credit too, there's a handy little button on a lot of pages on your site, maybe all of them that says ask ai where you can maybe 00:26:04.673 --> 00:26:07.343 get some answers, some questions answered pretty directly too, right? 00:26:07.848 --> 00:26:09.348 Nikhil Gupta: Yeah, that's awesome. 00:26:09.348 --> 00:26:18.058 Like too, where I think now this like dev tooling, like, like there's, so, I think there's gonna be so much, like, so much rich experience around like 00:26:18.058 --> 00:26:21.868 dev tooling because a lot of it more, a lot more can exist thanks to like ai. 00:26:22.758 --> 00:26:23.048 Like, 00:26:23.383 --> 00:26:30.643 like before, I think like if I was to think about like integrating a new API, it'd be like, oh man, I don't wanna figure out the documentation. 00:26:30.643 --> 00:26:31.543 It just seems like insane. 00:26:31.543 --> 00:26:32.593 Like, that sounds like a lot of work. 00:26:32.833 --> 00:26:36.223 Now I can go to the documentation, talk to the AI and be like, okay, like I need to do this. 00:26:36.223 --> 00:26:37.153 Can you gimme the code for it? 00:26:37.543 --> 00:26:39.193 It gives me the code I just put in my code. 00:26:39.193 --> 00:26:40.003 Like that 00:26:40.723 --> 00:26:42.013 is transformative. 00:26:42.013 --> 00:26:46.168 I think to like composing, like composable APIs, I think we're gonna see a. 00:26:47.233 --> 00:26:52.293 So many APIs, I think in this like next, like two, three years it's pretty, pretty exciting for me. 00:26:52.653 --> 00:27:01.503 But yeah, I think for people like, who are considering, I guess like yeah, like what would for API builders, I guess, like the 00:27:01.503 --> 00:27:04.233 question you, the reason, the way you would, the reason you. 00:27:05.643 --> 00:27:08.853 It's like, maybe you should ask yourself if, like, what are you trying to build? 00:27:09.033 --> 00:27:12.963 Like, are you, A lot of our users are startups, right? 00:27:12.963 --> 00:27:23.113 Because they're trying to explore prototype, whether they wanna sell into like a small, like a particular niche, like Vertical sa or I guess 00:27:23.113 --> 00:27:26.023 like the, like everyone knows Activity four Oh will be everywhere. 00:27:26.443 --> 00:27:30.643 Like that experience that this showed, I think that's going to be in every phone. 00:27:31.213 --> 00:27:33.423 Every microwave, every car those 00:27:33.423 --> 00:27:34.113 assistance. 00:27:34.503 --> 00:27:37.773 So now the question is do you wanna bring that experience? 00:27:37.773 --> 00:27:48.153 Do you wanna help get that experience deployed in different places and you can wait for four to come out, or you can use to get started on that right now. 00:27:48.753 --> 00:27:55.503 And then when we do have gbd four, we'll just plug it in and then you'll just have, won't have to do any work to switch it over. 00:27:55.923 --> 00:28:01.893 And, and then I guess it is for us to figure out like what other value we can add, which will be like observability, tooling. 00:28:01.893 --> 00:28:11.823 Like you feel like a million calls that you're doing, are you gonna really use like the raw for opening API or are you gonna use like. 00:28:12.333 --> 00:28:16.693 Some production ready tooling around to make sure that your calls are going well. 00:28:16.843 --> 00:28:23.263 There's a visibility, tooling, but that kind of thing is I think people, what people would use this for and that's what we are here to provide. 00:28:23.713 --> 00:28:27.223 So yeah, I think I'm drawn to people like if you think voice cool. 00:28:27.223 --> 00:28:29.983 If you think multi model is cool, give it a try. 00:28:29.983 --> 00:28:31.663 Like make something prototype. 00:28:31.663 --> 00:28:32.953 It's so easy to prototype nowadays. 00:28:33.813 --> 00:28:34.983 Mike Bifulco: Sure, yeah. 00:28:35.403 --> 00:28:40.233 The imagination should run wild and, and go build the things that you definitely wouldn't have been able to, 00:28:40.263 --> 00:28:43.053 you know, in 2022 or, or however you wanna look at that. 00:28:43.483 --> 00:28:46.273 Can you tell me about maybe an interesting use case or two that you've seen? 00:28:46.303 --> 00:28:47.108 With api, I. 00:28:47.423 --> 00:28:47.753 Nikhil Gupta: sure. 00:28:47.803 --> 00:28:49.633 I can tell you the UCSA. 00:28:49.693 --> 00:28:49.993 Yeah. 00:28:50.083 --> 00:28:53.293 WC this is our first customer that we worked with. 00:28:53.558 --> 00:28:55.023 They do sales training. 00:28:55.153 --> 00:28:56.053 They're called hyper bound. 00:28:56.473 --> 00:28:59.013 And you can go to their website. 00:28:59.013 --> 00:28:59.733 It's really fun. 00:28:59.783 --> 00:29:00.593 It's called hyper bound. 00:29:01.193 --> 00:29:07.853 I don't know the actual domain, but so hybrid, it will show, it should show up and you can call these boss to practice selling. 00:29:08.333 --> 00:29:13.553 So the founder, you can imagine, like, it's pretty learning how to sell is like a pretty key skill and pretty hard. 00:29:13.943 --> 00:29:17.453 You can practice on real humans, but then you're leasing losing real deals. 00:29:17.883 --> 00:29:21.393 So with practicing these bots, and these are bots are really hard to sell to. 00:29:21.633 --> 00:29:23.793 Like they would like hang up you on you all the time. 00:29:24.213 --> 00:29:30.908 So that, that's an interesting kind of role play is a very interesting use case that we see where like having AI to represent 00:29:30.908 --> 00:29:34.208 like hard conversations, having represent like sales training. 00:29:34.578 --> 00:29:35.748 So that's like one category. 00:29:36.088 --> 00:29:42.058 The other one that's pretty obvious is customer service, where people call into this, this restaurant, it's 00:29:42.058 --> 00:29:46.198 like, I want pizza here, or I wanna book a reservation. 00:29:46.198 --> 00:29:48.178 Like, here's an availability for the reservation. 00:29:48.538 --> 00:29:55.088 That kind of like function calling, tool, calling use cases are another, what are other popular ones? 00:29:55.448 --> 00:29:55.568 Mike Bifulco: I am. 00:29:55.628 --> 00:29:59.528 I'm curious you mentioned your founder created a coach bot early on. 00:29:59.778 --> 00:30:02.208 Do, do you or your co-founder have any that you use for yourselves? 00:30:02.208 --> 00:30:03.888 Maybe internally, externally, personally, I. 00:30:04.523 --> 00:30:07.048 Nikhil Gupta: I, I use for testing my, my Kill bot. 00:30:07.438 --> 00:30:17.238 But I don't know if I do, I have like I mean, there's one thing which is I don't like the experience of like CGBT, like voice. 00:30:17.743 --> 00:30:27.023 As much, 'cause I mean, right, right now they haven't deployed it before, I think in production for voice and being able to kind of 00:30:27.023 --> 00:30:31.043 brainstorm because like I'll get stuck on a problem and I just need to 00:30:31.043 --> 00:30:32.033 like, talk to something. 00:30:32.473 --> 00:30:32.983 It's like. 00:30:33.553 --> 00:30:40.473 I'm, I need to think out loud with someone that's like intelligent and used, I used to do this with my co-founder all the time, and now I need to do 00:30:40.473 --> 00:30:42.378 less with him, just with the ai. 00:30:42.723 --> 00:30:47.753 I dunno if that, what that speaks for human connection, but it's just like, yeah, just, I just loop it out and then I talk 00:30:47.753 --> 00:30:53.123 to it, but like I'm just holding it down because I'm still processing and I just like pick up a thing and it response. 00:30:53.513 --> 00:30:56.483 So brainstorming is a very common use case for me. 00:30:56.883 --> 00:30:58.773 And yeah. 00:30:58.943 --> 00:31:03.503 Mike Bifulco: have a I have a weekly newsletter that I write on my, my personal site under my own name, Mike, by full code.com. 00:31:03.503 --> 00:31:06.513 I publish a newsletter for startup founders and JavaScript developers. 00:31:06.783 --> 00:31:13.663 And often I'll sit down in the morning and just brainstorm, like talk, paddle through some ideas and try and you know, expand the 00:31:13.663 --> 00:31:15.883 idea to something that kind of has a beginning, a middle and end. 00:31:16.243 --> 00:31:17.713 And yeah, I, I agree with you. 00:31:17.718 --> 00:31:23.123 I don't think 4.0 is deployed yet to that audio experience, and it's a little irritating when it doesn't work well. 00:31:23.373 --> 00:31:28.413 The, the biggest change and the most valuable thing for that is that now it has memory context where I had previously been like, 00:31:28.413 --> 00:31:31.143 fine tuning things over and over or re-uploading transcripts. 00:31:31.143 --> 00:31:34.838 And now, now at least that part is getting better and that's been a big helpful change. 00:31:35.283 --> 00:31:35.763 Nikhil Gupta: That's awesome. 00:31:35.768 --> 00:31:36.003 Yeah. 00:31:36.003 --> 00:31:36.753 Yeah, exactly. 00:31:37.023 --> 00:31:44.013 I, I think my co-founder and I have a bet around like whether as like years go by or months go by, people like. 00:31:44.838 --> 00:31:49.488 Will there be a wide widespread love for AI or widespread like skepticism around ai? 00:31:50.298 --> 00:31:57.528 My, like, it's open, open question because we will be like, oh, AI's getting better, taking my job, or a lot of negativity or 00:31:57.528 --> 00:32:03.848 responsible, like, I just think like when you look at activity four, oh, like talking of the thing is like magical like that demo and 00:32:04.218 --> 00:32:08.658 it'll be hard not to like that experience and like want that thing. 00:32:09.468 --> 00:32:10.098 Everywhere. 00:32:10.488 --> 00:32:16.843 So very curious to see that, that sentiment shift on ai what will happen. 00:32:16.903 --> 00:32:17.283 But thanks. 00:32:17.553 --> 00:32:17.843 Four. 00:32:18.973 --> 00:32:24.468 Mike Bifulco: It is hard to tell, and I think that the mindset is very split right now in many directions, old and young. 00:32:24.528 --> 00:32:26.353 I think men and women have different feelings on this. 00:32:26.358 --> 00:32:30.233 I think different generationally, you know, gen Z and millennial and, and all that. 00:32:30.283 --> 00:32:32.413 There's, there's lots of different opinions on this stuff, and I think 00:32:32.473 --> 00:32:33.463 we'll figure it out over time. 00:32:33.493 --> 00:32:33.823 Yeah. 00:32:33.938 --> 00:32:40.508 Nikhil Gupta: Yeah, but I guess like, I don't know if you run into this, but like, for example, my mom, she's like not very technical savvy. 00:32:40.513 --> 00:32:47.168 And for like, for example, like she needs to, wants to learn how to, like, she's figure this out now, which is great. 00:32:47.428 --> 00:32:48.718 Like how to post something on Instagram. 00:32:49.678 --> 00:32:57.238 And the process for that is like you sharing your screen and you're just kind of pointing her to a button, like press this. 00:32:57.818 --> 00:33:01.358 Talking about it, like, to me it just seems like, like yeah. 00:33:01.358 --> 00:33:02.318 Would be so good at that. 00:33:02.318 --> 00:33:03.938 Like, I think all people would love it. 00:33:03.938 --> 00:33:10.518 Like, it's like it, it's patiently just kind of looking at her screen and like talking you through like how to like, do something 00:33:11.363 --> 00:33:11.653 Mike Bifulco: Yeah. 00:33:11.658 --> 00:33:13.518 Nikhil Gupta: that is pretty compelling. 00:33:14.438 --> 00:33:21.038 Mike Bifulco: Yeah, a very a big potential for a tender and loving experience that helps you grow at your speed which is super interesting. 00:33:21.043 --> 00:33:21.333 Yeah. 00:33:21.638 --> 00:33:21.968 Yeah. 00:33:22.208 --> 00:33:22.868 So cool. 00:33:23.178 --> 00:33:25.358 Tell me a little bit more about how you're building Matthew. 00:33:25.358 --> 00:33:27.608 Like what, what is underlying your your software? 00:33:28.458 --> 00:33:33.818 Nikhil Gupta: So, yeah, we decided to use Node because the thing, so what's that? 00:33:33.818 --> 00:33:35.768 Like, why is voice harder than chat? 00:33:36.128 --> 00:33:36.428 Right. 00:33:36.428 --> 00:33:44.188 And like open A is now running into this, like getting, trying to deploy multimodal systems is because the chat it's synchronous like, 00:33:44.188 --> 00:33:50.588 it's like one thing after the other thing that after the other, the voice the person is talking, they might stopping stop again at any 00:33:50.588 --> 00:33:52.718 point and then you start talking, but they might interrupt again. 00:33:53.243 --> 00:33:59.913 So there's like a very dynamic environment which makes it hard to write like code for it. 00:34:00.333 --> 00:34:06.603 And so for us, like given that there's so much like, like there's, you're getting audio every 20 milliseconds. 00:34:07.213 --> 00:34:12.133 You given that there's so much activity, you clearly need the event loop, right? 00:34:12.133 --> 00:34:15.953 You need like event loop, like audience coming in, like process that is there something you wanna say? 00:34:15.958 --> 00:34:16.853 Like send that out? 00:34:17.303 --> 00:34:23.723 If you need an event loop now for an event loop, you could try to build, use Python, you could use something else. 00:34:24.033 --> 00:34:25.648 You can make custom event loop in rust. 00:34:26.353 --> 00:34:32.458 But, but I was like, no, no is such an obvious answer for me because. 00:34:33.958 --> 00:34:38.758 I'm standing on the shoulders of Giant, like this event Loop has been optimized the heck out of it. 00:34:38.758 --> 00:34:42.208 Like, because the entire world browsers running this thing. 00:34:42.718 --> 00:34:45.268 But the event loop is really good, basically like that. 00:34:45.298 --> 00:34:48.298 I cannot, I can't imagine all to do anything better. 00:34:48.808 --> 00:34:55.418 So we use Node to manage the asynchronous nature of our workload, which is like you're getting on audio base all the time. 00:34:55.748 --> 00:34:59.348 You, you're sending out audio all the time, and there's like so many checks between those two. 00:35:00.158 --> 00:35:05.718 So the a harass nature node, these stack for the API interface, we use nest js. 00:35:06.088 --> 00:35:12.928 Because Nest has like an excellent experience around like structuring, like making sure they get very opinionated interface 00:35:12.933 --> 00:35:17.368 around like how to get data moving and get, get, get operational. 00:35:17.658 --> 00:35:23.538 And then also exposing that, like, exposing that in docs, exposing that in like exposing, having validators on it. 00:35:23.988 --> 00:35:30.518 So Nest js is like kind of our choice for an opinion and framework in Node, like we can run in node. 00:35:31.068 --> 00:35:35.508 And then for the database side of things, like I love Postgres. 00:35:35.568 --> 00:35:36.588 I've always loved Postgres. 00:35:36.588 --> 00:35:38.028 It's just awesome. 00:35:38.478 --> 00:35:39.588 Never had any issues with it. 00:35:40.008 --> 00:35:45.238 And then so we use base for it, I think has an awesome managed offering that. 00:35:46.363 --> 00:35:50.083 Kind of, I think, gets to the heart of it where you have nest node. 00:35:50.153 --> 00:35:54.693 And then of course we have like, kind of a we have many different like GPU services that need to run. 00:35:54.973 --> 00:35:55.573 Like there's 00:35:55.573 --> 00:35:58.663 things that you can often turn on, like, for example, you wanna turn on emotion detection. 00:35:58.783 --> 00:36:00.133 That's a good GPU service. 00:36:00.553 --> 00:36:02.983 You wanna turn on Denoising, that's some other G service. 00:36:03.323 --> 00:36:08.913 So we, our cluster is like this main API gateway. 00:36:10.698 --> 00:36:14.948 Sharding to like different workers and different services for different stuff. 00:36:15.348 --> 00:36:18.438 And we use Kubernetes so as everything because That's great. 00:36:18.678 --> 00:36:22.608 Oh, one thing I think that's kind of underrated that we love is Lummi. 00:36:23.178 --> 00:36:30.088 So if you use, you've probably used Terraform before to spin up everything. 00:36:30.398 --> 00:36:33.338 Lets you do that in your programming language of choice. 00:36:33.728 --> 00:36:40.088 And we, I love it where it's like, you can imagine like, so for example, we have many different reds that we use. 00:36:40.718 --> 00:36:45.483 We specify, we specify a list of them and then we can just literally write that code. 00:36:46.293 --> 00:36:48.543 Or cons, Redis of red. 00:36:48.933 --> 00:36:51.813 Like, it's like, before I do this, like it's like, and it'll actually do it. 00:36:51.818 --> 00:36:52.983 It will, it just works. 00:36:53.043 --> 00:36:55.573 It's crazy to me, like that, that is possible. 00:36:56.113 --> 00:36:59.953 So Lummi for orchestrating things yeah. 00:37:00.284 --> 00:37:01.274 Mike Bifulco: That's an unsung hero. 00:37:01.614 --> 00:37:05.464 And so I, I think you mentioned before you publish an open API spec with swagger. 00:37:05.464 --> 00:37:06.064 Is that right too? 00:37:06.329 --> 00:37:07.619 Nikhil Gupta: Yes, yes. 00:37:07.679 --> 00:37:08.544 Do you want me to link to it? 00:37:09.449 --> 00:37:09.869 Mike Bifulco: Sure. 00:37:09.869 --> 00:37:10.049 Yeah. 00:37:10.049 --> 00:37:11.069 I'll make sure it's in the show notes. 00:37:11.069 --> 00:37:11.249 You, 00:37:11.369 --> 00:37:11.659 Nikhil Gupta: Yeah. 00:37:11.659 --> 00:37:11.859 Yeah, 00:37:11.999 --> 00:37:12.929 Mike Bifulco: welcome to drop it in here. 00:37:13.604 --> 00:37:13.974 Nikhil Gupta: Yeah. 00:37:14.069 --> 00:37:16.709 Mike Bifulco: That's something that our audience will be super interested in. 00:37:16.714 --> 00:37:19.834 It's we talk a lot about open API in that world here for 00:37:20.009 --> 00:37:27.419 Nikhil Gupta: I love open a PII like, I think this is the concept of like, like I just envision, you know, everything to live in my code. 00:37:27.989 --> 00:37:33.959 I editor, like, I just create this thing and then this open API like is like kind blossoming into like this types everywhere. 00:37:34.724 --> 00:37:39.204 Kind of the, the dream world for me it doesn't exist right now as much. 00:37:39.264 --> 00:37:44.094 I don't think there's like as yeah, it, it getting there, but open a open API is awesome. 00:37:45.029 --> 00:37:46.619 Mike Bifulco: Yeah, it's, it is super helpful. 00:37:46.619 --> 00:37:50.479 We we have lots of hard opinions on, on when and how and why to use it around here. 00:37:50.479 --> 00:37:54.404 And this is one of the things I like about this community is I'm always learning from people smarter than me about, 00:37:54.409 --> 00:37:56.624 you know, tools and things available in that world too. 00:37:57.644 --> 00:37:57.824 Cool. 00:37:57.829 --> 00:37:59.884 So, Nikhil I'm, I'm curious what's next for you? 00:38:00.384 --> 00:38:00.894 Nikhil Gupta: Yeah. 00:38:00.954 --> 00:38:10.334 For us, I think more observability, more monitoring kind of like going into Wapi and then being able to see how 00:38:10.334 --> 00:38:12.984 your calls are doing knowing that they're doing well. 00:38:12.984 --> 00:38:19.884 That is kind of like what's next for us, like giving people security that like, like, their voice, AI is doing the work for them. 00:38:20.424 --> 00:38:24.774 I think that that would be a big thing going forward too, where like, you have all these agents, but like how can you 00:38:24.779 --> 00:38:26.574 be sure that they're actually doing what you want them to do? 00:38:27.169 --> 00:38:32.619 Mike Bifulco: Yeah, making the black box a little less black is a very, very valuable feature especially when your business relies on it. 00:38:32.619 --> 00:38:37.689 If I'm standing up a customer service endpoint of some sort, I want to know that it's doing great customer service. 00:38:38.719 --> 00:38:39.349 Nikhil Gupta: So for us it's 00:38:39.349 --> 00:38:44.419 like all about like that, that satisfaction of knowing it's, it's working well. 00:38:44.741 --> 00:38:45.851 Mike Bifulco: Super, super cool. 00:38:45.911 --> 00:38:47.406 so we've talked about how you build things. 00:38:47.411 --> 00:38:48.876 We've talked about yourself and your co-founder. 00:38:48.876 --> 00:38:49.771 How big is the team right now? 00:38:50.761 --> 00:38:52.681 Nikhil Gupta: We are a team of like six right now. 00:38:52.981 --> 00:38:56.686 And definitely always looking for the next great hire. 00:38:57.206 --> 00:39:04.831 If you love making, if you love voice ai and are curious about like this whole world of modality and getting the deployed. 00:39:05.476 --> 00:39:09.916 Which we feel really passionate about and like making great experiences for developers to make it easy. 00:39:10.216 --> 00:39:11.266 We get it deployed everywhere. 00:39:11.606 --> 00:39:11.936 Yeah. 00:39:11.996 --> 00:39:12.741 I would love to chat. 00:39:13.756 --> 00:39:14.086 Mike Bifulco: Yeah. 00:39:14.086 --> 00:39:14.536 Perfect. 00:39:14.536 --> 00:39:15.136 That's great. 00:39:15.376 --> 00:39:18.706 And so just, just to wrap things up then, where can people go to find. 00:39:19.381 --> 00:39:24.796 Nikhil Gupta: Ai, we, we are kinda everywhere where you can find us on Twitter, LinkedIn, ai. 00:39:25.201 --> 00:39:26.161 We have a blog. 00:39:26.371 --> 00:39:30.571 Again, if you're a developer and you love docs, you know, this is actually, this was a learning for me. 00:39:30.811 --> 00:39:31.891 People don't go to our website. 00:39:31.891 --> 00:39:33.331 Like they'll just go straight to go to our docs. 00:39:33.571 --> 00:39:35.491 Like, they don't go to the dashboard, they don't go to anything. 00:39:35.521 --> 00:39:37.741 They, there's like, oh yeah, just like read through the docs. 00:39:37.741 --> 00:39:38.491 I'm like, oh, cool. 00:39:38.491 --> 00:39:39.931 That's that's, that's interesting. 00:39:40.026 --> 00:39:46.386 So docs, we have ai, just Wapi V be 00:39:46.386 --> 00:39:46.446 the. 00:39:46.616 --> 00:39:51.386 Mike Bifulco: have definitely found yourself in the right place to get to people who want to go straight to the docs and read. 00:39:51.416 --> 00:39:54.566 Of course, will make sure that there's links to everything you just mentioned in the show notes. 00:39:55.131 --> 00:39:55.461 Nikhil Gupta: Thanks. 00:39:55.871 --> 00:39:56.201 Mike Bifulco: Nikhil. 00:39:56.261 --> 00:39:57.131 Thanks so much for joining today. 00:39:57.131 --> 00:40:02.671 It's been a real pleasure chatting with you, and I'm, I'm really excited to get in and poke around with some voice assistance myself. 00:40:02.921 --> 00:40:09.261 Please, please feel free to join us anytime if you have more news to share or things that you're interested in chatting through and sharing with the audience. 00:40:09.266 --> 00:40:10.221 We'd love to have you back. 00:40:10.251 --> 00:40:11.271 Thanks a ton for joining today. 00:40:11.271 --> 00:40:11.721 I appreciate it. 00:40:12.611 --> 00:40:13.031 Nikhil Gupta: Having,