Pełna transkrypcja odcinka PK Skip navigation Create Avatar image She Built an AI Agent Team to Bootstrap Her Startup | Emily Gold

you're wearing multiple hats and having this style is worse it is worse right but but i i think how i rise above it is i know my stuff i'm the subject matter expert here right like so if you're not ready for now i'm not going to force it's like dating if you're not ready for now i'm not going to date you if not my icp whatever i'll just move on and the right person will come and the right fit Yeah, you get to know as fast as possible, more or less. And the longer it takes, the more likely it is to be a yes, because you've eliminated the nose early on. And yeah, it's interesting. I mean, I feel like that's one thing. I meet people from different parts of Asia all the time, right? And it doesn't matter if it's Pakistani, Indian, Thai, Chinese, Japanese, Malaysian, Indonesian. filipine filipinos they'll all have that mentality that look we we we're always passed over or we're not we're not outspoken that is our biggest mistake or our big regret of ours. And it feels like it's sad when it comes out this way, right? When it comes out as a mistake, because a lot of times maybe it's like you're known for being humble or well-mannered or down to earth and very kind and caring. But at the same time, the people look at that as it's maybe it's held them back in a way. Isn't that weirder? It is. It really is. And some people mistake your kindness as like, hey, this is an opportunity to bully you. Or you know what I'm saying? Yeah. I mean, like, how do you, so how do you balance it? You said it yourself. You want to be kind. You want to be caring and respectful and nice, which you also don't want to be taken advantage of and run over, right? There's got to be some give and take there, isn't there? Yeah, yeah, yeah. A hundred percent. And I know my value. So here's what I've come to terms with. I've been really into Ryan Sirhan this this this few months but not everyone's going to like you and it's okay I think in fact it's even better right you if there's too many people liking you and it's not it's not really necessarily a good thing so what I've come to learn is I'm just going to be my authentic self respectful and kind and just communicate the values of my my my experience my product and not try i think it's just that and just being a human and building that trust because the last thing you want to be is like those really hardcore car salesman that everyone like you know like so i think that's how i draw the balance i just come to terms that i'm very happy with who i am today I have a voice. I'm not afraid to share it now or speak up or walk away. Interesting. So how does that go into your role at Govio? Sometimes you might have to write harsh emails or say hard things. And you might not be kind about it because it might be really, really hard to do. Or maybe you're kind about it, but it's still hard. There's no easy way to. to tell a vendor or tell a provider or tell even an AI that what has been done is not good enough. Right. There's not, sometimes there's not an easy way of doing that. Right. So how do you, how do you, you know, how does that, how does that come out when you have to give negative feedback? That's awesome. I think it's like managing a team, giving constructive feedback. So many vendors coming to me where let's just like, let's not use a person like, let's like my AI. I'm working with it to build my product. and it just builds something and i've spent like days because it just keeps going and solving a problem that i'm like what is this like it's you should like why do you keep going so i'm like you know what maybe i need to take a step back and reflect maybe i haven't been communicating my expectations and my outcome clear enough and i think that's the same when you're managing a team or vendors like you have to be clear like hey in xyz day this is what i'm expecting specifically from you and then you get like ABC. Okay. So let's talk about how you talk to your AI because that opened up a different conversation in my head. So like, you know, you're building this thing with the help of AI and all those sorts of things. And now when you're doing it and the developer. i'm calling it the developer i'm raising my fingers with uh you know quotation marks when they when they make a mistake because they do you know they're not humans but they they hallucinate just like humans do i mean do you come down on it or do you be like this was good but or how do you talk to your ai developers that's interesting so i i don't beat around the bush i'm just very matter of fact i'm like here's this is not what i'm asking apology and like i i'm very i'm very polite to my ai because i just like quad released the statement like you cannot abuse your ai or whatever like that and i've noticed like if you give them a nicer prom they will be better in giving you the answer so i'll be like this is not what i'm asking for maybe i am not clear enough my bad and it was like here's what i am actually asking you to do here are the guardrails if you have any questions i because i built a suite of agents each agent has their own role like please consult with the qa and then you all are subject matter expertise i should be the last person like if you guys have like a mis disagreement within yourself give me a summary of what that looks like give me the good bad ugly whatever it is i will make the final call i'm like the human has to be in the blue and So you talk very kindly to it. You're not dropping the F-bombs and going nuclear. And I'm an F-bomb dropper. I know. That's why I was asking, you know, like, I mean, how are you not, how are you not doing that? Like, you know, it's, it just kind of comes out, right? Like, you know, all this thing is like, you know, this thing is just like trucked up beyond all recognition. I don't know what you're doing. Like, do you ever, you never, you never go into that with that? This is so interesting. No, I. no not even when my team is just so interesting but can i tell you

something else it's so funny so i'm using quad for something else i'm using gpt for something else and every and everything has their own benefit gpt knows my tone so when it's writing posts for me because i have an agent doing that i should you not it sounds exactly like me with the f-bombs that's good i mean like that means you know it's really well trained right so it's It's interesting. So you mentioned something earlier that you have a series of agents building the product for you from scratch. Now, most of our listeners have only heard about people doing that. They've actually never done it themselves. What's it like having a team of robots working for you? Do you feel like a Malaysian Tony Stark in a way? What is it like? Take us underneath the hood a little bit of what that's like. What's a team building event like? What are your Zoom calls like? Just take us through what it's like operating a company like that. it's amazing i'm such a nerd so i named everyone after marvel heroes and um x -men so i have dr strange i have iron man is actually my finance my cfo because he's so rich so you should know spider-man is my outbound my whole go-to-market this is so weird so i asked spider -man the bill a sub agent to help him with the end-to-end outbound go to market and he built a sub agent called mad and i was like holy he just built his best friend from the movie i was just mind blown but here's the best thing i put it in the way when they built them i'm like you guys are free agents where you all can tap into each other on your expertise have a discussion not everyone has to be involved And the prompts, you can see them having a discussion. But when, and then, so the first time I billed them, they were spending too much time having a discussion. And I'm like, holy, oh my gosh, this is like the analysis. Yeah, now you're watching, this is working in an organization and having so many freaking meetings that's useless. And I sat in parameters like, listen, if you have to argue for more than 10 minutes, cut it to the chase. my priority your priority govia's priority is trust customer first whatever so whatever decisions that you're making has to put those factors atop fiercely prioritize it significantly reduced to xyz minutes which is amazing So you reprimanded Spider-Man and Iron Man from too much water cooler discussion. And now they don't talk to each other anymore. They do. Emily's going to get mad at us if we talk too much. It has to be efficient. I mean, that's the whole point because they were taking a lot of tokens. Have you ever thought about what they're going to say about you on Glassdoor.com? Oh, my goodness. i don't know i think they're still very happy with me for now yeah maybe that's the next step but that is interesting though that you you actually saw them idling right or wasting time wasting tokens and wasting money resources and you had to tell them hey chill out right or don't literally stop chilling out and start working a bit more tighten things up and they listened and there was was there any drama afterwards was there any sort of like you know situation with them? No, they just stop. Okay, noted. We're good. We're going to do our own thing now. No drama, to be honest. Nick Fury is my chief of staff. So he knows somebody. But Nick Fury is my PA, my EA, my chief of staff managing them. And he was like, no, this is the final word from Amelie's guidance. And he actually slaps their hands on my behalf. And they don't listen. And I slap them again on the wrist. But it's a very respectful. It's a respect. Okay. So, you know, could you see like, this is going to get a little bit futuristic because now I'm really fascinated. Could you see a situation where there's like a Captain America Civil War sort of story happening here where there's like strife within the team of agents and some want to do something one way and one go the other way? Yes. i've already put in parameters because me but then here's again this is not new look at uh look at gpd codex has already experienced that yes i'm not afraid about the rift but civil war but i'm more afraid of them just going out of their boxes and start doing other things okay okay okay okay okay you see what i'm saying i'm not afraid of them fighting and arguing with each other i'm more afraid of them oh i built i did like a robot i developed my own thought What if I just do that without telling anyone? So it becomes like Ultron in a way. It's malicious and they do their own thing and it's actually counterintuitive. Correct. Let's go back to the mistake. When was there a time when you felt, you know what? I should have spoken up. I should have said something. And my people-pleasing way. got in the way of it was it when you were younger was it in corporate life was it in school was it all those things what made you feel that way so i came here in 2011 i think so 2011 it took me a couple of years that i didn't think i was good enough and English wasn't my first language so I've had people actually making fun of like oh English isn't your first language you're not good enough you're not this you're not that which really impacted but and the good thing is I think we have a very resilient culture like i think in our way like we were brought up very resilient so i i don't know how i went through that but in the data like i'm a very data driven person so i'm like yes but look at all the results i'm delivering so even though i was getting those kind of feedback that are not conducive i was delivering results and i was you know outcome that no one can deny or argue So I think that ended up building my confidence, my credibility, because you cannot argue with like, yes, I've done that. Right. And you feel like that is what gives you the confidence now is the fact that you've done things in the past. And when you have moments where you're maybe having imposter syndrome or shaky confidence, you go back to your credentials and remind yourself

of that. How do you remind yourself? I just be myself because I'm just being very truthful. I use all my experience and I will tell them, I'm like, listen, that's no right or wrong answers. That really isn't. What I can tell you, I'm a really good dot connector and I can just look at it. It's just based between a combination of experience and looking at data and just understanding what's happening around and how. here's the thing right just because you're super experienced and this is how i i got myself up it's like because i'm i keep feeling i'm not experienced enough i'm always learning and as a result of that i'm actually taking in more and more confident am i able to articulate things because ai wasn't too new but the fact that i'm able to build my whole suite of agents and whatever they're doing now with harnesses i feel like that's huge but i'm still humble enough to like oh i don't know that This is what I did. This is what I have, you know, this is what I will share, but I will still be humble enough to learn from others. Yeah. And what you said is actually very true. I can give you some validation because I just said not many people have built, they're not building products with multiple AI agents in the same way that their entire team is basically comprised of AI agents masquerading as the Marvel Cinematic Universe, like the way that you have. like take take me through the first time you you interacted with an llm and then how did you make the leap to decide you know what i'm gonna do this thing i'm only gonna work with robots what was the What was that? What were those experiences like? What are those stories behind those two moments? Yes. So one thing, I'm not only going to just work with robots. I love my humans. I love my robots. It's going to be a combination. Sure. No, I'm kidding. I'm kidding. I'm joking. I do love that. But the first time I started working with an LLM is just the same as any others, right? Like I used it like, Hey. This is my raging email to someone at another department. Help me polish it and make it really respectful so HR doesn't come to their room. Who hasn't done that? Yeah, who hasn't done that, right? So it's a lot of those polishing, content creation. It's just very basic promo. And then you start a testing like, build me a campaign or whatever it is that I want it to just get built. and i realized like oh they can build a ton of things but they still need me to be in there because they don't have the context they don't have the knowledge that i have of that business and the best part was like with data you feed them raw data and then it can give you numbers that don't even match or not non-deterministic which is but but that's how i started using which is simple polishing email build Govio primarily without a product team, without developers, like human developers necessarily, and relying a lot heavily, not entirely heavily with the Marvel Cinematic Universe in your laptop? I'm not really going in blind, like technically, because I've built websites and stuff like that. That's what I started with. i first worked with quad and i just dumped my ideas and there's no agents there and i built it and with just the code and my idea and it came out and i was like whoa i don't know if this is me being dumb or super ignorant and holy crap i can do this let me just go do this maybe the founders the founders craziness that that gets some of the people super successful they just believe in that idea um but that was that was it when i first saw it working and i actually brought it in front of some of the operators like the users and they were like wow you made this and even show some of my product developers friends and they were like you you and so they were like flabbergasted but so i'm like yes it means i'm doing something yeah so let's talk about that right so you build this thing either you made this you how'd you do it right now when it comes time for How do you handle things like a code review, security, governance? Because now you're talking about board reports, right? This is not, you know, simple things. This is confidential information. And, you know, how do you make sure it stays in the right hands? How do you make sure Ultron isn't, yeah, taking that stuff and making some conclusions? I know, I love, yeah. But how do you handle code reviews, governance? Because, you know, you're building a product now. There is that agile methodology, right? How are you following it? So I started learning about a code methodology and I'm like, just make sure because I know Python is definitely more for scalable. So I started incorporating that, right? So again, it's not like I'm just going in blindsided. So I made sure everyone's consistent with that, but from a gut real standpoint, so this is how I built it. I'm not building where I'm. i'm like keeping all the data i'm legit like the data trust layer where i'm sitting on top of your crm so now you can connect me with your crm that is so safe and all i'm doing is i'm looking at the data from in there okay okay mine is really yeah mine is really just hey here's the bible whatever that playbook that that can exert the the rules and playbook and it sits on top of the crm and it says like 80% of your data is not following this rule. I got it. Okay. Okay. So you're basically, yeah, you're on top of another data set and that's why it's easy to do what you do. You have a very interesting story. I mean, how does somebody who's not a developer become one of the top? 10% of users are lovable, for example. It just seems like very, very, you know, like... I don't even know how I got that 10%. I'm not even using lovable. I don't know where that came from, but I was so shocked when I saw that. I'm not even using lovable anymore. Maybe one of your people are. Your people. Maybe my agents are. You're really enjoying this. Oh, come on. I have a whole org structure. And I have to write down who they are and what they do so I

don't forget. But anyway, I think it's because I keep things simpler. So even in a traditional organization, I work so closely with my product managers, the developers, and I love them. But you know what I noticed? They're so technical. They go down the rabbit hole. Yes. Overcomplicate things. and me i think this is where i'm lucky now where i've been salesperson i've been in product marketing you know i've tried to translate technical stuff to like how customers perceive it so i simplify it so that translates to my prompting to my agents i keep it so simple i don't write a whole paragraph that confuses my agent which can dilute everything in my core message so did you How did you perfect your prompt engineering? How do you improve your prompts? What are things that you do? Do you practice it? Do you have a danger room where you're doing stuff like that and you're making long prompts, short prompts, different kinds of words, different kinds of languages? Have you found things that work better than others? Have you given guardrails to how you manage your Cloud account or I'm assuming Cloud or ChatGPT? Let's do all of that. yeah absolutely i think it's all it's part of that and it's also like learning from influencers to power users plot i'm not really that there are people always better than me so i don't know if you know ruben has it or yeah big time yeah correct so i will look at him i will learn from him and then i know this like yeah i i am the kind of person that will provide a ton of context and you know what i realized like this with my team members as well when i was managing them i would provide so much context and it took them away from the core point so it's really managing humans with superpowers or interns there's no difference there i don't think not a lot so give me give me an example of of too much information for uh for an agent for one of your for one of the avengers and one where there's the perfect amount like is there is there like an easy is there like a simple example that you can that you can share or yeah and we'll use spider-man again because he's like my outbound so i'll give him go to this persona and i will have a detailed persona written out and i'm like this is the goal of my campaign i needed to do from start to finish It has to have 11 minimum touch points. It's just too much, right? And then here's the content. The poor guy, I think he got so confused. He was just like building the campaigns and it's not even coherent. And I just, you know what I'm saying? Because he's like, what exactly do you need? I'm getting a ton of information. And then because eventually I got confused as well. Writing the prompt even. Wow. Okay. Right. So you let him down a path where he just couldn't follow because you didn't know where it was going yourself in hindsight. So how did you change it? What did you change it to? If it wasn't this whole big manifesto, what did it become afterwards? Wow, skills. So I have saved this. Oh my goodness. I give them memories. But I heard right now that my agents or agents, if you have agents, they can have dementia. interesting it's true so so anyway i give them skills i give them memory like hey in your memory this is who we're targeting but you have the ability to learn like this from not my prompting but from the internet and from any interaction you know with others as well so it's very interesting like i taught it to be self-learning self-improving autonomous agent And I see that it's amazing. That is so interesting that they're able to do that. And you're able, you're able to teach them to do that. I mean, that's like the, that's the bigger, that's almost the bigger feat. Let's talk about that too. I mean, you are objectively, I'd say you're further along the, the, the AI spectrum, right. Then most, most people are. Okay. What do you feel like, you know, when you, when you talk to. other professionals about their own or businesses about their own ai usage and their ai adaptation or their what's your general view of it are they are they doing it enough are they not doing it enough are they not taking advantage of it what's your what's your general view on on business climate and ai adaptation yeah okay this is a really good question i think the same thing they don't know what they want i think first of all they need to know what their angle is when they're using ai a lot of the businesses that i see they're using ai for the sake of saying i am an ai native company and i have an ai native team without truly understanding what that means there's no guardrails around the ai Which is scary. So you would think like having autonomous agents is scary. No, you have people going to AI, putting in data and then spitting out forecasts and following forecasts and firing people. Those are what I'm calling BS, right? You need to know the right way to use AI, how to securely make sure it's consistent across the departments in the organization. You just need to know what that workflow looks like. It's all about that. You have to have common sense. But how can people learn this? These things are changing so fast. There's so many things coming at people. How can they learn all that? it's a lot to ask right the data is changing they're trying to get the data in there and it doesn't work doesn't look the same we have to make decisions we have to use ai to do it there's all these sorts of forces pushing at them how do they how can they how can they keep on top of it that's a really good question so that's that's a top down but something like i can i'm starting to see this the newer organizations with folks that are more ai native truly ai native and they will encourage that the older organization that that's they that's nothing i can do there but i started an ai community to support folks and in my community i have two extremes one i'm like we're f because ai is taking over everything or i have this group of power users and we'll just be talking and

throwing ai slang harnesses or whatever It's just crazy. But the goal is to support. So I think like they are classes out there and they are mentors out there. They're the true subject matter experts out there like Ruben and so many others. I think they should learn from them instead of just. But there's nothing they can do because like you're right. Like the top down is just like you have all these expectations. And now like I have just gave you access to AI. You need to do more with less. Yeah, and that's really the trick, isn't it? People have to do more with less and the tools that they have to use to leverage to do more with less, they're brand new, they're changing. What those tools can do are changing all the time, right? So it's a lot and I commend you for being able to stay ahead of that, right? I mean, you actually have a team and it looks like a very well cohesive, you know, a very cohesive team. You guys can, you know, save the world from the blip. You guys can do so many things. I just put that in there. And, but most people don't have agents that can, that can, that have JDs and let alone can like, you know, you know, hallucinate even. They haven't got to that point yet where they can even have that sort of an issue because their relationship with AI is too introductory. It's maybe. There may be a step one where let me write the email to my colleague that's got too many F-bombs and I got to make it less nuclear. I got to make it more wind energy. They're still at that stage of AI. They're not at the stage of I'm working with agents most of the day. And do you know what the secret sauce is as you were talking? It is because I am so passionate about what I do and it matters to me the quality and the output of what my AI agents spit out. I review. them the reason why there's so many hallucinations and just garbage output out there it's because the people and the users just put things into ai comes out and i've gotten this from a former manager that that puts into ai built a 10 pager campaign and everything is just so repetitive and he didn't water the check So that's what it is, right? You see what I'm saying? So AI is so powerful, but the humans were not accountable enough or passionate enough. There's no synergy there between the AI and human. Yeah. It's like you're in a kitchen. And you have like a sous chef, right, who's underneath you. But it's your restaurant. It's called Emily's Restaurant, right? So, of course, every dish that goes out, you want to see before it goes out to a table. But otherwise, if you don't care, then you're just churn and burn. You're just throwing things out there. You don't care what it looks like. That's it. And it's not just about AI. So, you see, even with humans, it's the same thing. If you have an intern, you still have to check and give guidance. And I think that is an ultimate... you know, tell, show and tell off, like, is this person a true leader? Because that's how you know a good person, a good

team member that will evolve and eventually become a team leader. That's so interesting. And just, do you find that some of your agents have a higher need of approval than others? Has that the genetics from you been subconsciously baked into any of the Avengers at all or no? Hulk is my security. So, and I built it to be more needy for a very good reason because it's security. So if there's anything that he's flagging, like this needs a lawyer review, this needs blah, blah, blah, blah. This is against the EU data, you know, violation or whatever. Correct. So I will look at it. So I intentionally built that in. How do you intentionally build it in? Do you feed it Disney movies or something to watch? How do you build in sensitivity? How do you build a sensitivity? What's your secret for that? It's really just you are the expert of customer privacy violations. So you look at all the laws out there and just be very sensitive about the words, the usage of the words and everything. So it's just very skeptical. Oh, it's a skeptic. Okay. And like, let's talk about training them. Okay. Because you do have to train them, right? Like, do you, I mean. do they have behavior i mean do they do does their behavior evolve i mean as they get more information have you seen that from them yet or or no i mean or how do you like how do you keep a sensitive hurt how do you keep like the hulk from becoming too insensitive like i mean what is your is is that a thing even and and do you try to like uh would you feed it like the transcript from the lion king just to get it back to normal what would you what would you do to like i really treat them like humans like that's what i'm saying yeah if i'm managing my team and i'm training someone who wants to be a people manager i'll be like so for example spider-man has his first team member ned i'm like spider-man you need to be empathetic And you are also dealing with customers and people outside of Govio. You need to be more understanding, empathetic, and just more personal, conversational. So Spider-Man is just more human in nature. Right. Iron Man is a very matter of fact. Like, this is not in line with our budget, blah, blah, blah. So it just tells me as is. I'm like, because I'm the person. You're an accountant. Yeah. i'm like brief and be you know like just be gone kind of you know like i need a matter like facts um so i think it depends on their roles i would still joke with them and have fun but because you know but this is so interesting but nick fury is is like nick fury my chief of staff he gets things done i'm like this is my I'm like, you give me summary, bullet points, everything. But when you're treating the agents, just make sure you're talking to them with respect and they know what the priorities are. And okay. So this is all incredibly, really fascinating for me. Maybe because of the superhero names. I don't know. How do you communicate with them? Are you communicating natively through Claude? Are they in Slack channels or? stuff like that how are you communicating with them are you guys emailing what's the situation i'm just claude just through claude all the time so it's like a lot of tokens a lot of tabs open just talking are you typing or do you voice with them i'm like typing so it's quieter so it's not mommy's yelling at the mommy's yelling at the computers in the room again it's not like that oh my god hey spider-man do this for me i'm like i just cannot i'm a nerd but i think that would be very cringy we got a financial league iron man get it i know i just don't i love them but imagine myself have to say that aloud and how long have you been working with them like how long has it been so here's the thing right the last time i checked i think this is a month ago i've spent 6 000 hours with them so i'm like working like 12 is like the minimum. Sometimes I work 15, 18 hours a day just talking to them. Unbelievable. 6,000 hours, right? That is incredible. Where can people go to learn more on how to build like a, like a team of agents like this? Like, is there like a playbook? Is there like a course or is there like some sort of resources that they can, that you can recommend to people that go here, learn this, do this and read this and you can kind of replicate what Govio does? I think Ruben has it. I don't know, but I feel like this podcast, if it goes out, then they'll be like, oh, we can talk to Emily or we can talk to someone else. Because that's not really my forte. I'm not necessarily a pro in building agents, but I know what I want to build for my agents. That's what I'm saying. I think there are people out there that teaches you how to build your own agents and how to harness them. I don't know if you know that Java is like a new AI tool where it helps provide more deterministic numbers and whatnot. That's a whole chat group. They will provide you help. Jeff, okay, but what you're doing, it's not so much building an agent, building skills. Yeah, those are common practice, I would say, for most people. I shouldn't say most, but people in our industry sort of thing. Building a product the way that you have with a team that does different things for you. If you actually have, on the Govia website, you could actually put these people as different. roles iron man is your cfos spider hulk is your director of devops of security you could actually do that and they have personalities and you could probably screenshot even like their conversations and how they talk and how you guys are joking around like this is a living breathing hybrid company of humans and agents right nick fury chief of staff i mean you could give them avatars you probably have and you know like and you probably have these sorts of uh you can give the origin story enemies in their lives that they can that maybe make them who they are enemies and friends who they they make who are vacation policies you might even give them w-2s at the end of the day who knows right so it's

kind of like yeah you never know you never know it's incredible it's very different than what most people are doing with ai emily let's be honest and and i think the only difference is because i treat them like humans right because there are a lot of companies out there that are building agents to replace humans ai sdr victor.com and a lot of others like i'm not you know what i'm saying and the way they build them is very ai like robot right yeah whereas you're doing this because you're bootstrapping a company and this is the best way to bootstrap a company in 2026. and it's working especially to treat them like humans yes so so let's talk about govi will be like you said it's working like tell us about where you are now what the next what's the next what's the future look like in the short term and the long term it's not easy bootstrapping founder in 2026 especially with ai i'm not gonna lie every day even when i'm building my agents and i'm seeing them interact with each other and i'm like having existential crisis yeah i'm like why am i needed now but you know in six months i have gotten three customers one is not paying because that's when my first customer i'm like i think that's where my asian culture came out i was like oh i don't think i deserve to be paid so i got free for that and then someone told me like no you need to be paid so i got my second paid customer not a lot but still paid and then the third one was also paid customer wow um incredible It is. And now I have a pipeline. I have 12 founding advisory members that have volunteered their time out of their busy days to spend time with me, test my product, give me feedback, and then recommend me to other people because they trust me. And I'm going because of them. I would get messages from them saying, this is amazing. Keep going. We're rooting for you. it's it's amazing to have all these experts you know supporting me to be my cheerleader i am slowly like as i'm talking to them and then as i'm building continue to build this and test this out like i'm seeing this as like bigger than just gut rails and everything like i eventually envisioned this to be Time-saving, you know, and hold people accountable without the politics and drama. You know how in organizations now, there's so many people just finger pointing each other and just having discussion with no evidence, or I'm not clear about the evidence, we just have to go back fine, we should waste time consuming. This tool is going to call out exactly who dropped the ball without sounding accusatory. Because of all the timestamps and everything, it's very evidence -based. It tells you how much pipeline bookings is this impacting. That's my vision there. I have vision, vision too, as my agent. You have vision? That is very visionary and just thinking outside of the world. How much time did you spend naming these different agents? I don't spend a lot because I'm such a nerd. I remember their names better. I cannot remember people's names. That's the only reason I cannot give human names because I will forget it. I'm so bad at names. But I love my agents. So I picked all the agents that I love. This is fantastic. This is amazing. Emily, where can people learn more about you and Govio and connect with you? Oh my gosh. Come to my LinkedIn. I have started being so active on X. I'm like, oh my God, you will die laughing, Sigil. So I'm on X. Most of my conversations is with Grokbond and Bordy. This is so bad. This is terrible. It's good, but it's terrible. Look, Emily, thank you so much for your insight into running a SaaS company on the edge of everything, really. Most people are not starting SaaS companies, software companies in 2026 for obvious reasons. I mean, here you are doing it and doing it well and doing it in a very, very unique and modern way. I hope you don't ever need to wait for approval or validation of anybody. You know, you got it from me. So thanks for everything. Well done. It's been a pleasure to have you on. Thank you so much. Thanks for having me. I appreciate you. And thanks for saying that. I'm right back at you. Yeah. And if you guys learn anything, go check out Govio. It's a company on the rise. Emily is a founder on the rise for the way she's doing it. You can learn a lot from how she's operating, what she's doing, what she's building. Check it out. I recommend it. This has been another episode of Rebels 500, everybody. Thanks for listening. Take care.