AI Influencer Workflow Revolution with Kostas Ketselidis Uplodio Founder Exposed

The progress that is being made really excites me about what will happen actually two or five years regarding the workforce. I believe there will be a really large change. I I I thought it would be happening now actually. I think it got late let's say but I think in 2 or 5 years everything that we we know when it comes to let's say knowledge even universities and stuff will be 100% obsolete. Hi, my name is Demetri Bonichi and I'm a content creator, agency owner, and AI enthusiast. You're listening to the AI Agents podcast brought to you by Jot Form and featuring our very own CEO and founder, Idkin Tank. This is the show where artificial intelligence meets innovation, productivity, and the tools shaping the future of work. Enjoy the show. Hello and welcome back to another episode of the AI Agents Podcast. In this episode,

we have the co-founder of Uploio, Costas Gelis. How are you doing today, Kostas? >> Hey, I'm doing great. Thanks for having me. >> Yes, thank you for being on the show. It's exciting to to chat to um you today. A uh I was going to say a fellow Greek, but I'm only half, but still counts. Um appreciate you for for coming on the show. Uh I just wanted to kind of ask to get things started. What is your uh background and how did you get into AI and AI agents in general? >> Okay. Okay. So I'm pretty young. I'm 22 years old at the moment. So I don't have extension background. Uh but we started our startup uh with my best friends from I know them since 10 years old. Uh we didn't start as an AI startup of course in the beginning. Then we

pivoted to a more AI version because we saw it's it's really worth it actually uh except the AI AI boom and the capabilities that it gave us. Uh we thought that we have a way better value prop. So yeah, that's what we did. >> Very cool. Um and how did you uh you know kind of know that it was the need to take your product from being noni to AI? or was it just kind of the way that the the trend of the market was going or? >> Yeah, so it was mostly because we we started as a simple influencer marketing platform. Um we had users and the users complained that a process too manual. Uh so that is what we're actually solving. We have an AI agent that solves the whole that's the whole workflow of an influential marketer and uh yeah it was

too manual customers complained and the reason that's the reason we pivoted. >> Yeah totally fair. And tell us a little bit about what the product itself does now. Um it's kind of marketed on the homepage as your AI creator manager and an AI agent that finds, recruits, negotiates and manages hundreds of nano and micro influencers. What does that all mean? >> Yeah. Yeah. So to start with the problem, um many brands uh our ICPs are either B2B SAS, B2C SAS, FinTech, retail brands. Uh they want to scale influencer marketing, but they have significant bottlenecks when it comes to talking with creators, communicating, closing the deals. It's a it's a process that requires a lot of human uh input. Uh so they have a limit of course. And uh so our solution here is to help them scale to a lot of creators and influencers without

them hiring extra people and actually removing this bottleneck. Yeah. No, I I definitely think it is a bottleneck. I mean I as uh somebody who is a influencer um myself especially in like my personal YouTube channel outside of this podcast um I definitely know that I get a lot of interactions from different people and I know it's a pretty manual process to probably find people like me. Yeah. Yeah. Actually, an influencer manager, let's say if they want to activate 10 creators, they might need to to open 100 conversations. So, managing 100 conversations requires like two or three people. So, if we want, let's say, to run a campaign with 200 activated creators in multiple countries, you understand it's a mess. It's not actually possible. >> Yeah. And how does your product work to to kind of make this um simpler on a on a practical

level? Like what does it uh how does it do? What does it look like when people are interacting with it? Um, in order to get that easier experience, >> should should I go to technical or just >> Yeah, I think technical would be would be good. A lot of people on the show are um >> great. >> Basically, you know, I mean there's a lot of founders who would probably be interested in hearing how it works. >> Great. So, we actually um firstly we populate our database with creators. So we scrape, weate our database, we run our algorithms to um both analyze the transcriptions, what creators actually say in the videos and using other visual models to actually see how they look like, what they do. Uh more specifically, then we have our database. Then the user starts writing a prompt. Write a pro. Let's

say I want influencers that um talk about uh fashion and are over 40 years old and look um like this spec old money lifestyle for example. Uh they can go that get specific and then they board the agent. So imagine a new person in your team. So you have to on board her call her Amy and you have to give her context of your campaign, your writing style, FAQs that uh that creators ask. So as Amy our agent has complete context there the user writes a prompt they get a list of creators Amy reaches out to those with personalized messaging handles the whole uh process and brings you creators in waiting for approval status so she did a whole workflow and brought you creators that are willing to work with you >> h very cool and what kind of I guess you know I'm taking

taking a look at your case studies tell me a little bit about like the types of companies that you've helped um if if it's possible um make their way through your product and have a really great experience. Like I see you have shippable on your website. Um tell us a little bit about what that experience was like. >> So yeah, so at the moment we have three ICPS that we are testing which one is better and they get the most value of one of them is SAS. Cable is one example. Uh so for example Cable came to us uh they wanted a hands-off solution kind of many growth teams that you know startups get funded they have to perform and so a lot of growth like two or three months they cannot hire people to do it so for example uh they hired Amy to

to run this uh so they're on the wrong wanting creators that talk about AI no good automations in this space a popular list reach out to those and got them I think in one and a half weeks like 25 creators uh with the user you know not do having to do the manual work the user just accepted those and they're they were activated. Yeah. So so it's a kind of hands of solution for them to actually do this process without having to build systems and managing a lot of people to actually activate 25. >> Yeah. Yeah. And I think a fair question is what does this kind of look like from a practical and pricing standpoint? Um or affordability. You don't have to get into pricing, but affordability standpoint versus um like an agency that helps find influencers because I'm sure you're kind of sitting

at a different spot in there. >> Yeah. Yeah. Yeah. So, so the the the reason to prefer us, let's say from an agency, >> first one, price, of course, we don't get a commission. Most agencies get around uh uh 20% commission of the actual uh you know of the actual budget that they have. Uh we can work with the subscription base that works better for most brands. So that's when it comes to the pricing standpoint. But the main thing I would say is that they prefer us from agencies due to the um h how fast the go to market will be. So like in two days they can launch a campaign and have let's say in a week creators with using an agency you have to on board them give them briefs do the back and forth they will come with an offer after 10

days. So it's a lot faster than agencies and for example if you are a let's say company in Europe and you want to expand to Latin America you have to find a Latin American agency or if you want to go to MENA or go to US. So that's a you have to manage agencies at some point. So with AM you just manage a dashboard of you know uh ready creators. >> Yeah. No that's totally fair. And just out of curiosity what made you specifically kind of get interested in the and your co-founders interested in the influencer type of market you know like I feel like maybe I I would like to hear every founder kind of has an intrigue into like why their specific AI thing is important. What what what kind of gets you excited about this type of work? >> Okay. I will

give you a weird answer. Um I I didn't work in influencer marketing. Uh so I worked as a marketeer, you know, freelance jobs and also work with worked in another company running paid ads. And uh I saw that this company could not hire influencers at all. So they had a huge budget but they can they could not hire influencers. And it seems strange to me because u you know UGC was exploding back there like 2022. Um then I wanted to start a UTC agency, Tik Tok agency. Uh one of the first in the local market. Then I thought okay I'm going to start an agency. Uh I want to do a SAS. It's more interesting for me. Uh then I started looking at the market and yeah then we progressed. You know the first months were really slow because we were pretty young. We didn't

know how to run a business. So we had this learning curve and after that we we went faster faster. >> H yeah. And you know you probably had your own experiences running that agency. So probably told you what how the difficulties of maybe what an agency owner goes through and um how you could probably even do um a software better um than maybe the agencies could do in the in the space. Uh, so what what kind of are some of the um main specifics about your product that have kind of improved recently that you're excited about? >> Okay. One update that uh we've developed yesterday, I'm not sure if it's even pushed on the production actually. It's a thing that um so we saw that it's kind of hard to adapt AI in some organizations AI agents specifically because you know employees have a specific

workflow they work with and it's strange to them to let's say have an an AI agent take complete control >> and talk on their behalf actually um so we made a feature that our clients approve its message of Amy so Our main value problem is to run on autopilot of course to bring you better results but we saw that it cannot happen. Uh our users were scared kind of letting an AI do that process. So we developed a feature yesterday that each user can check all the messages and this will give us an easier way to enter into larger enterprise that have specific processes and not let's say have a push back from users the actual users doing the work because most of the time decision makers want the uh the outcome and the value prop of Amy but you have push back from users

because it's kind of changes their processes. Yeah, I think that's a very important point. uh AI sitting in the current workflows of people is a lot better than you trying to add something new on top of like what they are they already have going um so to speak because like there's a lot of different and you know just out of curiosity um what are your thoughts on over time this tool other tools kind of being able to take on roles at companies right AI agents are it's like a phrase that obviously a lot of people are aware of that's why they watch the show. Um but the interest to me is where do you think kind of the uh trajectory of AI agents taking on actual roles uh is and what do you think when do you think that might happen especially with tools like

yours? All right. So, I had a conversation with my girlfriend last year regarding that uh when we were starting building you know having the having the idea of Amy uh I was telling her that in two years the whole job markets kind of fact I I don't know what will happen many many uh many players will lose their jobs so it will be a strange you know um situation here and I I don't know what will happen in this case uh but uh from the progression is really fast but I I'm not seeing that happening let's say in the next year or so actually people in many organizations uh I I don't know about developers I think many develop junior developers might you know have an issue with that I'm not sure about it I'm not uh I'm not the one to judge but marketeteers

I've seen that at the moment they're not getting replaced more than the that AI assistants that get them higher capabilities where yeah they're enhanced more productive this is the way we should promote our products in my opinion at the moment because either in other cases will not be accepted into organizations of course and um yeah um but I'm not sure in five years time five years time you know I have two theories either we'll have like u um many people will will be unemployed that's that's a bad thing so will be unemployed or AI will will be like kind of an assistant still because some regulations made that uh made them being assistants uh but this is for large organizations for startups I believe that startups will have they will be way better because they will have the opportunity to have like three person team

doing lots of revenue without >> so yeah it will definitely help startups but I don't know regarding large organizations >> I think at at this point we'll have a huge boom when consultant um you know the big four and consultancy companies start cutting up on the AI transformation and see that they can make a lot of money out of it so then they will push it to uh companies but I I don't think this will happen at the moment You know, I think that you made a good point though. It seems like it's probably more of an enhanced uh situation than a kind of replacement situation for a little while. And I actually probably um agree with you. It's hard to know in a couple years where it'll be at, but um for now, I do think it's going to be a lot more enhancement

than it's going to be um >> Yeah, of course. >> removing. >> Yeah. Otherwise, we'll have a huge push. The ones that are building AI will have a huge push back. And of course, you know, um >> Yeah. Who's going to who's going to build their own demise? >> Yeah. Yeah. Yeah. >> Exactly. So, just out of curiosity as well, speaking of that and the whole personality, why Amy? Why is it called Amy? >> Um, I think it's random. Ah, I I think we just had an idea before. So, when we went to raise our money um in RVC, we just uh we had a really cool challenge. Let's say when they were testing us, we had a really cool challenge to uh completely erase a pluro and make a new idea in one day. Make it start in one day. So by random we

call the another AI agent in recruitment Amy. So we just give the name to Yeah. our own agent now. >> Yeah. Do you feel like it makes it seem more personable like person like the when when uh talking interacting with customers? Because I I'm I've always wondered kind of there's the route of just having it be the brand name and then there's the route of also having the agents kind of have have a name. >> Yeah. Yeah. We want to make like say okay uh you on board Amy to your team then then I start calling her C when I reference to Amy. So it's seems kind of weird calling an AI agency giving a personality but yeah after a while you get used it. >> Yeah absolutely. And you know, just to kind of talk a little bit more about um the product itself

and how it might compare to other tools. What's it like in regards to the I know outreach and that kind of stuff uh is a big part of it and managing the relationship. Getting into the more nitty-gritty details. What does it look like in regards to helping you enhance the content review process? Because I'm sure that there's a content review aspect to to this, right? um as influencers make content that people want to make sure that it looks good. How does that work um in your platform? >> All right. So, at the moment, we stop until the the influencers agree working with you and negotiate the price. So, we don't actually review the drafts. We've made the feature reviewing the draft, but we haven't we pushed it for a bit, then we uh we got it down because um it's a thing that influencer managers

want to do manually at the moment. They don't want an AI agent doing that for them actually reviewing the actual content of the influencers. So, we kept it down a bit. Uh then our next step regarding that is as we continue the process um after the activation let's say we will push it again uh more let's say optimized and also we get we'll get into the fintech compliance. So one of the three ICPS we have is Fintex. They have a lot of compliance uh teams working on checking the content. So it's a huge problem for them and we're going to focus on that firstly and then we'll go into general content. >> Okay, that makes sense. >> Is sorry the content review we do is before finding the influencers. So in Tik Tok there are multiple millions of influencers uh but most of them are

not good enough to part to do partnership with them. So we have vetting algorithms uh in order to find out if they're good enough to do a partnership. If they are, they join the platform. >> Okay, that makes sense. And you know, I guess maybe I'm trying to understand as well to the the other aspect of it like after somebody's already sure reviewed their product and reviewed or sorry, reviewed their plat their um profile and whatnot. I guess I'm getting at more of the post-production side of it, right? Someone records a a video. So you're saying that ba basically once you're cool with an influencer there's not like a process of reviewing the specific piece of content they would make. >> Yeah. At the moment Amy does not do that on autopilot. They can the the users can do it on upload. So the way

it works that Amy reaches out does the back and forth negotiates. Okay, I will pay Demetri $1,000 for this partnership. Okay, let's do let's make sure everything is ready. Then the influencer manager comes and activates that creator. And then we pass to the where we're saying Amy says to the creator upload your draft here. And then the influencer manager checks us inside upload. We haven't made a feature to for aim to check it automatically because this is something that our users want to still do because it's the creative part of the process. So they still want to have control over it. >> Absolutely. And totally cool to have control over, but there's still like a there is that process in there, right? Like like I'm saying there is a capability for someone to upload the uh the files and footage onto like upload and stuff.

>> Yeah. Yeah, there is there is there Gemini visual model to check the quality and stuff, but yeah, this is not really it. >> Yeah. Yeah. No, I think that's really cool because there's a lot of um I guess overall process that gets solved when you kind of make sure to put the review in there. So, I was just kind of curious how that works. >> Yeah. Yeah. But but um a product decision we have to make at the moment is uh how much involvement the user going to have because there are many teams that let's say want a hand of solution. So, they just want I want a 100 creators in a month just do it for me. They don't care about uh let's say a little negotiation tactics. They don't care about some details. So they just want the result. Um I think

it's a better way actually if you have the budget to operate that. And there are teams that want to check everything. They won't check the negotiation how the goes. They're more afraid to let an AI agent run on autopilot. So yeah, we have to find the balance on what Amy will handle what will leave up to the user. And do you think that um that's something you primarily try to take a look at like what the consumers think as well as obviously you're going to have your own opinions but is it probably from like feedback of consumers of for the most part that gives you some of those answers to those questions of whether you should or shouldn't kind of add a feature to Amy because people want to keep it manual or do you think maybe you'll have Amy possibly have some of these

features but you have the option to be like I don't want Amy to do that. >> Yeah. So yeah everything we build we get it from feedback from our users. We don't if we if users do not request request it um yeah what we will do is let them decide because at the moment we have these two type of teams and we we want to serve them both both of them. So yeah, um I've seen, you know, very advanced teams running putting a lot of content out, but they still want to u have control over the little things. So yeah, I don't I haven't made up mind what you do. So yeah, we have them for now. >> Yeah, just kind of curious kind of where where your head's at with it. It's always interesting to hear kind of the thoughts of uh talent or

not talent sorry founders and and kind of where um they want to go with things uh moving forward with this idea of agents and how they kind of impact uh the market moving forward whether it be job or or otherwise. Um, what are if what if there are any out there do you think are cool agent tools or processes that you've seen people work with in the influencer market as well? Is there anything else that you've kind of seen that you found to be interesting? Because obviously you're managing like relationships of influencers, but is there anything else that you've kind of seen through working with influencers that uh you found have been pretty interesting? not really I would say um regarding influencers. So there's kind of a standard process. There are not many different ways to go with influencer marketing I would say. So the

most interesting things are our I think two competitors that we have at the moment doing kind of thing. So that's the most interesting. Um but yeah I'm using a lot of AI agents. So yeah, I could check. >> What would you say then? Just I mean fair question to follow. What do you think it just something that differentiates you from your competitors? >> Um I don't know yet to be honest because um so we have I we have two large competitors and they are in the st same stage with us. So we are us three were experimenting with things. So they just in the same time as us. >> Uh so I don't know yet. I I don't. So they're not established companies. the established influencer marketing platform has stayed to the current model because yeah, I don't know why they stay to the current

model. Um, but yeah, the new months are pretty early as well. >> Fair enough. What then would you say is something that you'd like to see your product be able to do um in the next 6 to 12 months or what what would be like the ultimate vision for what Upload and Amy can do in the in the future? So in the next month let's say let's not say 12 months let's say three to six months a shorter term I would like to have the complete process but because each company has a different process maybe we're thinking of making making like um users to customize how their own workflow should be kind of like n10 with nodes but we haven't made it yet we're think about it uh so yeah in order to be able to serve multiple type of customers because we've seen that

each customer for some reason has a different process. So that's bad to productize something. So we have to find a solution there. Then the vision for after a year uh is to expand to other uh marketing operations as well. Let's say running paid ads um other operations that are really crucial to a team. And then let's say you know we want to make a workforce in marketing. let's say have a copywriter, a paid ad specialist. So this workforce that we get to your organization, we can cover them with multiple solutions. >> Yeah, that makes sense. So and you mean not finding people who can do that, you mean having an agent that can literally do the the thing >> correct? Like doing the copyrightiting, doing the advice. >> Yeah. >> Yeah. So if it is, let's say a huge organization, established one, uh they might

not want this kind of solution. But if we're talking let's say about a startup that go was founded like uh recently and they cannot go out there and find an agent they cannot go hire a lot of people doing that. So really rapid growth. This is a great solution to get into market fast. Let's say if you have a complete marketing team of doing uh growth activities. >> Makes sense. Um okay. So where in the world of AI agents have you been kind of most surprised this year as to how um capabilities have changed? Obviously there's been been a bunch of different things in this past year. There's always so many new interesting things, whether it be MCPs, um, the new SDK toolkits that like OpenAI and Anthropic have kind of put out. What's been top of mind for you as it comes to keeping

a high or keeping a good knowledge of everything that's going on? What's been like something that's interesting to you that's been released tech-wise recently? Um I I don't think I have a specific answer to that because there are many in many so as a founder I I have to get involved in many kind of operations. So it might be sales marketing might be general let's say um founding team operations. So I I have to be top of mind in all these related fields. Let's say um the thing that intrigues me the most when it comes to AI is that I use them mostly it's sales related activities. So I try to automate these things. Uh so the way I use AI in my day-to-day life it's either I'm 100% addicted to cloud. So, I've seen myself, let's say, be super addicted in in a bad

way sometimes because, let's say, I stop being creative. I stop making my own decisions because I ask what what to write here, what to say. I'm not sure if it's a bad thing actually, but for some days, I've been really addicted to it. >> Uh, but uh otherwise than that, I use it a lot for sales related activities. I I like this part. I use clay, I use valley, another great tool. Uh so yeah, but but I'm keeping my mind in many fields. I'm not sure. >> What about the I guess you know you just mentioned sales and how the stuff there helps uh with agentic. What what necessarily is that um what specifically are you using? Cuz I'm this is probably just a part of the conversation to to get a get a feel as a founder, you know, other founders can be listening

to this. What what's helping you on a day-to-day basis to to be more efficient and effective? So on the sales side, what what what would that look like? I guess. >> Great. So three things. Uh I use Clay, you know, it's a >> clay. Yep. Clay.com. Yep. >> I use Clay for enrichment and I get the the list from from Sales Navigator. I push them to Clay and then I run all my enrichments and then I write personalized emails in volume clay. Uh then I've been trying for a month a new tool called Valet. Uh that is for LinkedIn outreach. I like it a lot. >> How is it spelled? just just right here. >> Linked. Okay, cool. Tell me about that. >> Yeah. Yeah. It's a great one uh that um you put you upload a list and they do it's an AI agent

kind of like Amy kind of the same process. So they write you messages, you approve them, sends them to prospects and does the and then helps you in the pipeline. And then I would say when I when I'm in the phase of um managing a prospect sending a no foreign stuff I use cloud a lot again. >> Interesting. And what are your thoughts on kind of have you played around with the new uh 3.5 I'm guessing or sorry 4.5 on it? >> Yeah. It's pretty good I would say. Um what I've done is I I have made a project include that I include all my all my conversation with clients. Every mail that I have I push it there. Uh so yeah it has complete context and it gives me really quick and fast answers in my context because otherwise if I open a new

chat I have to explain stuff. Yeah this has been really great. >> Very cool. Um, and what do you think is kind of as an early stage founder some of the best ways that you've managed to find uh the different customer pockets, right? I'm sure you know you're doing all this outreach. Like outside of that, like how how did you manage to figure out who you're supposed to go after ICP wise, how to reach out to them? How how did you kind of go about figuring that out? >> So, I haven't figured out our ICP yet. So we are testing three. We are testing fintex, retail, DTC and SAS uh SAS companies. So yeah I don't have an idea what is my SCP at the moment. We are reaching out to all of those we are having clients from these three verticals. So yeah I

think when we more customers from each of them uh then is when we going to decide that okay now we have that ICP. >> Yeah. Yeah. So then how are you going about like going through the process of even you know when you're doing the the clay like enrichment clay outreach and stuff what what goes through your mind in order to kind of make the testing happen so to speak. I'm just trying to get like how how do you even go about testing right? Yeah. Yeah. Like what's your process? >> The question that I pose to myself and also the different enrichments that I do is is this company would benefit from large volume creator marketing. So that gives me the answer. So let's say if we get a fashion brand, yeah, they would benefit. If we get the neo bank, of course they would

benefit again. Um, so that is the main question I I answer when I I I I ask when I see a specific lead. Uh, and that is I would suggest to other founders doing that process. Uh, and yeah, I think LLMs are pretty good in figuring that part out at the moment if you give them enough info. So we do a lot of enrichments from many angles to find to give get a lot of input to some prompt in order to decide if it is a good lead or not. And then we score these leads and then we start the outreach >> and with the products that you've kind of used for outreach and stuff. Obviously >> at the moment my sales text is sales navigator. Then I go to clay. Then I it's kind of complicated. Then I go to airscale to get emails

and phone numbers. Then I push them to instantly and I'm using value on the side. So So yeah, we have a complicated um sales stack. Uh I need to to make it simpler, I guess. >> No, it's a lot of people's sales stack. They have a million different like they they clay's in there, right? Everyone's got clay in instantly and then you're trying to figure out kind of what are the other things that go. So I was just curious. Yeah, Clay's actually it's an it's a really innovative company product wise because they they were one of the first doing the doing the blank canvas type of in the sales process. So you so that is what we try to do with taming in the next months actually because every salesperson knows that uh things that they know the process best and it's software. So, Clay

gives you the option to not change anything but rather than than scale your existing process. Uh, so actually that's what we're going to try and do within as well. >> Yeah. No, that makes sense. And I guess, you know, obviously that's sort of the the sales side of things, but just on a personal level, I know you said Claude and maybe that's the answer, but what um is your personal favorite tool to use right now? Um or like and try to think of ones that maybe are outside of the normal LLMs that you've kind of found in the AI realm. And then what is your favorite feature of your own product? yeah I would say that yeah from my personal operations um I would say the answer is claude actually um okay there >> so what I do as a founder I do my personal

operations handling clients you know all the things and then >> sure >> doing sales >> so and in marketing related stuff I don't use any tool at the moment to do it uh because I want to learn it by myself in the beginning and then starting to automate it. Uh so yeah, at the moment I would say it's 100% cloud doing that then other um yeah then valley and clay on the sales side. I these are the tools I'm using and I would say I'm not early early adopter in AI tools. most of the pros that I I see um so I believe that before starting to automate something you have to figure out what process works for you and then starting automating it. So yeah that is what has worked for me at the moment. >> Very cool. Yeah know and and then what

about the um Amy side of things? Amy. So yeah, my favorite part that it's actually the most valuable prop to our users is the the whole closing a deal that Amy can do with influencers. So there are influencer tools and there are tools that search for influencers. So these things exist, but we haven't seen a tool doing the back and forth and actually closing a deal with influencers like a human would do. negotiating the price. We've seen, let's say, Amy negotiating a price from uh 40 4500 to $800 from a creators. We've seen great things that Amy doing negotiations. So yeah that is something that uh really excited me seeing my product do our product do and also this what it will double on because it will be a really uh this will be the way to actually get into organizations them making sure that

can negotiate on their behalf um even better than they can do. And what have you kind of uh I mean last question before I let you get the floor for the for the rest of the show for a plug, but what have you kind of seen or have you used your own product to kind of get influencers uh to share your product? I'm curious if you've gone that meta. >> With it at any point. >> So yeah, um I would say if we're about to run influencer marketing, we would do that for LinkedIn. uh most probably because our ICP we want to reach decision makers mostly so we should go on LinkedIn um what we've done but we started doing it then we didn't continue we'll do it again is let's say run an affiliate program for applio uh because these influencers talk with many

talent management agents they work with a lot of brands so we'll give them a referral code and uh they will promote upload to their brands so it's kind of a reverse way actually. Uh yeah, but we haven't actually ran it at scale to to >> Okay, got it. Just curious. And and kind of last thing to close it out, what is um something you'd like to plug kind of uh obviously the product or anything specifically that you'd like to say before we end it out? Your final thoughts? >> Can Can I repeat? Sorry. >> Um I would love to hear just as a final thoughts, you know, anything else you'd like to say about your product and and whatever you'd like to plug um to to close the show out. Okay. Um, so a general a really general thing that you know the prog progress

that is being made really excites me about what will happen actually two or five years regarding the workforce. I believe there will be a really large change. I I I thought it would be happening now actually. Um I think it got late let's say but I think in two or five years everything that we we know when it comes to let's say knowledge even universities and stuff will be 100% obsolete. Uh so yeah the best thing we can do is actually be on top of things and you know learning every day about uh new things and try to implement ways to be more productive in our work. Um so yeah that is actual thing we're making a tool to solve uh this problem with influencer marketing and uh making brands scale it a lot. Um yeah I I suppose this uh the whole industry will

get way ahead and we haven't seen it peak at the moment. We have many years to of progress. >> Yeah. Very cool. Well I appreciate the the insights on everything and and kind of your thoughts on uh the industry at large and what tools are working for you. I guess just last thing I'll say to everybody is please make sure to go check out upload.com and meet Amy. Um, >> yeah, we appreciate you for being on the show today with us and uh, >> great conversations and yeah, >> thank you man. >> Thank you everyone. Please who is watching, go check out upload.com and make sure to leave a like on the YouTube video review the Apple podcast and Spotify podcast that includes you Costas. um so we can get uh as many eyeballs on Kosus' uh product as possible. Once again, that's upload.com. Thanks

for watching and we'll see you in the next one. Bye-bye.