Manuel David of Paradigm Outreach: The College Dropout Building an AI for the Entire Sales Cycle 18VC Podcast · Episode 13 · July 10, 2026 Guests: Manuel David Source: https://www.18-vc.com/podcast/manuel-david-paradigm Host: Manuel, thank you so much for joining us today for this episode of 18VC Recording. Do you want to maybe take a couple seconds to quickly introduce yourself to our audience? Manuel David: Yeah. Thank you so much for having me. Yeah. I'm the founder and CEO of Paradigm Outreach. I'd say Paradigm kind of started mostly as an outbound tool, like run your outbound on autopilot was kind of our motto at Paradigm. And we quickly decided that we want to switch to something where it's the whole growth cycle. So right now, Paradigm is a growth engine where you can do all of your lead prospecting and reaching out to said leads. So you can research them, you can find the leads, and you can send them. You can do signal-based campaigns or any type of campaigns. You can do that through email and LinkedIn. But you can also do posting. So you can post to different platforms, you can do different type of UGC content and image gen, video gen, posting. So we're trying to do the whole cycle, just the whole growth distribution engine. Host: Awesome. We'll spend more time digging into what Paradigm is. Manuel David: Of course. Host: But before that, we want to start learning a bit more about you, right? Manuel David: Yeah. Host: So what were you doing before Paradigm? Manuel David: Yeah. So what I was doing before Paradigm, I actually was a student. I went to Arizona State University and I was doing studying over there, computer information systems, and I actually had my own business. I started my own business because I felt like there was a need in my community, which was just my school, for kids who hadn't had internships yet. Host: Oh. Manuel David: And what I decided to do was make personal portfolio websites for them. Something that a few people knew in the tech industry, oh, you come with a portfolio, a little bit about your tech skills, and I kind of wanted to dig deeper into that and also democratize it so it wasn't just for tech and it was for everyone. So I started making personal portfolio websites for my friends- Host: Awesome Manuel David: just like websites. And then I started doing it for everyone around the school, and that's how I first started making money inside of building with AI. And then- Host: Wow Manuel David: from that, I just kept going. And Paradigm Outreach kind of came about from my brother starting it a little bit, hearing that he was building something, and I started to be quite interested in it because I was already building in AI. And when I first heard about Paradigm, all it did was you put in some research and it would personalize an email for you, and that's basically all it did. It didn't do any of the sending, didn't do any of the replying or anything like that. And I saw it and it was really, really cool. And so me and my brother started working on it together and we started building it up. And then I started using the tool first myself before I even really did- Host: Interesting Manuel David: any more of the stuff that we have now and making it super agentic. And all I did was get super personalized emails and then send them out to people. Host: Yeah. Manuel David: And I was getting a lot of replies. And from that, I think that's when I first started to believe in my product and believe this could definitely be a game changer and something that is very different from a lot of the other tools, because I'd used Instantly and other tools before. Host: Absolutely. Manuel David: And there was something different about this software that- Host: Yeah Manuel David: we were creating. And so once I started seeing how many replies we could get with it, I was like, "Wow, this is definitely something we could outsource. This is something- Host: Interesting Manuel David: that we could use in multitude of situations." Host: Interesting. And going back to you. Manuel David: Yeah. Host: You're a dropout founder, right? Manuel David: Yeah. Host: Currently. So when is the moment you realized that you should drop out instead of pursuing studying and building at the same time? Manuel David: Yeah. I would definitely say shortly after I realized how good it was working. My position with my brother was actually I was a bit of a growth/builder. So I wasn't just building, I was doing a lot of just using the platform and trying to get us clients and trying to get us leads. And once I started selling these $10,000 deals and getting on the phone with big people, I was like, "Wow, I'm doing this all from my dorm room. I'm doing this all from my apartment." Host: Yeah. Manuel David: I feel like there could be more from this. And I was going to school, and classes were boring, and my professors were great, but I just didn't feel fulfilled, right? Every single time I went to a class, I would just think about Paradigm Outreach. How can I optimize this campaign? Where should I be trying to improve this software, and where should I think of other areas where I shouldn't try and go into? Every single time I went to class, I feel like I was doing calculations on how to better it. Host: Oh, wow. Manuel David: And so from that, I decided, okay, maybe school isn't probably the best idea for me- Host: Yeah Manuel David: where I should be doing right now. Host: Yeah. Manuel David: AI moves fast, and I thought that- Host: It does Manuel David: if I was going to get on the train, if I was going to get on the cutting edge, then I had to move fast as well. And so that's when I decided to drop out. Not drop out forever, but necessarily... Well, that's what I thought, not drop out forever. But maybe take a break from school. Host: We'll see. Manuel David: Yeah. Host: We'll see. Manuel David: Maybe take a break from school and start building it out and kind of bet on myself and trust and believe in myself. Host: Yeah. So you feel like what you learn from the classroom, from the professor, from your classmates is not kind of relevant too much to what you're building in Paradigm, right? Manuel David: Yeah. I felt like specifically just being a computer information systems major, I didn't really feel I was learning any of the tools. I think that for right now, what school is right now is going to have to change fundamentally as AI goes. Host: So it's outdated in your eyes? Manuel David: Oh, yeah, for sure. I'd say being in San Francisco, people say it's kind of a bubble. And I think that San Francisco kind of living in a bit of a future situation- Host: Yeah Manuel David: compared to the rest of the world. Host: Yeah. Manuel David: And I see as you come here, the most successful people are the people that dropped out. And the most successful people are the people that took that chance- Host: Very true Manuel David: and bet on themselves. Host: Very true. Manuel David: Right? And nowadays you see people just talking about, "Hey, what have you built? Where did you get to? Who do you know?" And it's not like, "What school did you go to? Did you go to an Ivy League?" Which are things that I think are outdated questions, right? I think that there's going to be a pendulum swing from everyone going to school and trying to get to the highest school. Host: Yes. Manuel David: How can I go to the best school, to how can I be the most successful as I can be? And then they're going to weigh their options, whether go to school or don't go to school. Host: Yep. Manuel David: Which I think is then going to be like, okay, well, if I can do everything that school would teach me with AI, why do I need to go to school? Host: Why do I need to go to school? Manuel David: And then school is going to be like, okay, well, not as many people are going as before. We got to- Host: We need to figure out a way to get that, right Manuel David: Yeah, we need to figure out a way where we can still have students. Host: Totally. Manuel David: And so I think that's what's going to be the shift, and I think that it's going to be an amazing shift. Host: Yeah. Manuel David: And I think when that... happens, I'll be way super excited to get back into school. Host: To go back, yeah. Looking forward to that. Manuel David: But for right now, just kind of learning things I don't think pertain specifically to what my goals are in the future. I think right now I'd better bet on myself and learn through doing. Host: Yeah. Back to how to discover the pain point for Paradigm. You wrote on LinkedIn that many founders spend three hours every day writing cold emails, but the response rate is bizarre. It's 1%- Manuel David: Yeah Host: only at the end of the day. Is it observation from your own experiences doing sales and outreach, or is this the repeating pattern that you observed from conversation maybe from other founders around you? Manuel David: Yeah. We work with a lot of founders, and we work with a lot of companies that talk about how their outreach pains and their outreach struggles. And again, like I said, with me, I used other tools. I was- Host: Yeah Manuel David: going through the same struggles. And 1% is actually the industry average, right? So most people that are saying they're going to do your cold email for you are averaging about 1 to 2% reply rates. And that three hours, it's true, because if you're going through lead lists and you're going to figure out who specifically should I research, what vertical should I research, and you're doing that all from trying to search on Google or trying to talk to ChatGPT, it's going to take you a lot of time to get that out there. And then don't get me started on creating domains and getting your whole infrastructure set up before you even do your first send, right? And that's kind of what I wanted to solve with Paradigm, kind of just catapult you to the last step. We like already sending. We have everything else. You just got to kind of have a conversation with our AI, and you can't skip that. I think what's really valuable is getting your voice inside of your copy or getting your voice inside of what you're trying to send out to people. And so in our platform, we actually force the users to have conversations and- Host: Oh, cool Manuel David: lengthy conversations with our agent system, so it really understands what they're trying to say instead of just skipping to the end and saying, "Okay, let's send." Right? I think that your voice is very important and should be inside of the emails. And so, we're not solving for that part, but more of the other parts that can be automated, like just the finding and the sending specific part where you're not going into your inbox- Host: Yeah Manuel David: and clicking send. Host: Right. So in your eyes, is the problem that founders don't know how to write good emails, or it's just the entire outbound workflow is just broken? Manuel David: That's a great question, actually. I don't know if I've been asked that before, but I think what I see a lot is it's not necessarily that founders don't write good emails, because as humans, as me and you, you probably know what you want to say. Host: Yeah. Manuel David: We're really good at communicating. That's one of the foundational things about being human is we know how to communicate, right? And writing emails is not necessarily the hard part of outbound, right? You can send emails all day and not get replies, right? Host: Sure. Manuel David: I think it's the whole system that really has the problem. There's multiple levels. It's like, okay, if I'm writing this email or I'm creating this email, is it actually going to get read by a person? And if it gets read by a person, what emotion is it going to elicit in them? Is it going to be something that they want to reply to? Is it too long? What type of person am I sending it to, right? And so these are all these calculations that we kind of have to do as humans. And instead, I think that with an AI agent, you can kind of have someone you can talk back and forth with- that understands the person that you're going to be reaching out to and can be like, "Hey, they're a little bit busy in their inbox. How about you go to this vertical?" And just someone that can give you a helping hand. The AI agent doesn't know more than you, but it can help you in that thinking process where it's like, "Hey, I'm trying to do this. This is what I'm trying to say, and this is how I want to communicate to this person. How can I do that? Who should I be reaching out to?" And then that's the kind of question, because the AI specifically from my view, I don't think it should be doing the whole writing of the email and everything itself. I think it's the conversation where you feed the information, and you feed the context, and it understands your messaging, it understands your voice, and then tries to construe that point to someone else. Host: Absolutely. I think one of the reasons that we ourselves also experience of a low response rate is actually, for example, we also got some inbound messages from others. Some of them, for example, they are founders or investors, and these kind of people we are kind of more interested in talking to. And they are some of just our fellow students. And they're just curious about what we are doing, and I'm not saying that it's not good, but if we have a very tight schedule, maybe we'll prioritize these guys who are founders or investors- Manuel David: Of course Host: who can be our guests more directly, right? Manuel David: Yes. Host: And I think the reason why we are responding to some of them and not as immediate as to the others is because the first type of people, they provide more value for us, actually. And I think the reason about a high response rate actually is, do you find the right people where you do provide a true value that they also value, right? So can you tell us a little bit more on how do you discover the right people? I think it's not only about sending as many email as possible but find the correct person to send the email to. Manuel David: Yeah, for sure. I think you hit the nail on the head right there. I think it's a counterbalance between volume- and also who you're reaching out to, right? Host: Yeah. Manuel David: And so once you figure out exactly where you're supposed to be on that scale, I think that's whenever you can hit something that is actually a true kind of, I guess sympathy or symphony of- different types of tools and how you're orchestrating. Excuse me. I think, to just answer your question bluntly- It's difficult. Right? And you're not going to know your first outbound. Your first outbound campaign, a lot of people like to say, is not going to be perfect, right? I think you don't even know yourself what you're trying to say until you try it and get- Host: Yeah Manuel David: some nos, and then you're like, "Okay, these people said no. I kind of understand." "Maybe this is not something that's going to work for them." So I think that really it's about experimenting a lot. I don't think it's about sending thousands of emails to people who don't want it, but I do think it's very much about experimenting- and trying your hand at a different thing. Right? So let's say I'm sending something that is founder-focused, and I'm sending it to somebody who's a founder, Series C, and already has all of their stuff booked, right? I don't understand necessarily yet when I'm reaching out to them, or when I'm sending out to them, that this is going to be their reality, that they're not really trying to move on to another tool, just as a solo founder myself. But what I'll see from the feedback that I get from the campaigns is that, hey, this is something that people aren't looking at. This is something that people aren't liking. Why? Right? And so we try to have our agent do that before you even do the sending. So that's why we really, really invest in those conversations where it's like, hey, I want to reach out... If you tell our agent, "Hey, I want to reach out to this marketing agency, blah, blah, blah," it's going to be like, "Okay, cool. Done, but why?" first, right? And it's going to ask you some questions of like, okay, well, maybe you shouldn't send to them. Maybe you're looking more at D2C. Host: Oh, wow. Manuel David: And so it kind of allows you to have a conversation with our agent back and forth and tell you why the specific vertical or the ICP that you think you might want to target might not be the best one. And I think that's something that's super valuable- Host: Yeah Manuel David: that our platform does that not a lot of other people do. Host: So you can actually understand our intent- Manuel David: Yes Host: behind your action. And then not only that, I think something new that you-- that something very interesting I heard is that your agent seems to be able to actually counter back of what the users want to do, right? That's not something that we commonly see in a lot of AI applications nowadays. Manuel David: Yeah. Like we like to say, it's kind of opinionated software, which means that it has a backbone, and you're not just going to push into a corner and say, "Send a billion emails," right? "Here's my money." Right? Because that's not what I want my product to do, and I don't want that to be the output of my platform. I don't want it to be a whole bunch of people- Host: It's like spam engine, right? Yeah Manuel David: Yes, exactly. A whole bunch of people who are sending spam getting on my platform and telling my agent to send the spam, right? It's actually going to understand the root cause and the why behind someone's doing something. Host: Wow. Manuel David: And then it's going to try and figure out the specific vertical that's going to work. It still might not get the right one, but I think two heads are better than one, right? And so you're going to start in a way better direction than if you were just like, "Hey, I think I already know. Here's who I want to send out to. This is what I want to do. Just send out my emails." So it's more of a partner, and like I say, it's more of like an agentic SDR system than just something that sends emails for you. Host: So the clearer the users understand their own intents and the purpose of the campaign, the clearer that you would deliver the results. Manuel David: For sure. Host: But what if the users just have just a very rough or generic idea and doesn't really have a clear direction of what they're trying to do? Manuel David: Yeah. I love that. We have something that's called an offer guru, which means that you come in, you bring your website, you bring in your idea, and it asks you a whole bunch of questions, right? And these are thought-provoking questions. So if you don't really understand what you're doing and who you should be targeting, through the context that it has of you in the conversation, you'll kind of get pushed to one side. Host: Wow. Manuel David: We actually do that before anyone can even enter the platform. So if you were to enter the platform right now and you were to kind of have a loosey goosey just like a, "I want to reach out to everyone," it's going to be like, "No. You can't reach out to everyone." And it will not let you go into the platform until it's decided that you and him have come to an understanding where something's good enough, and it makes sense enough to reach out to people, and you have a good ICP. Then you can go into our platform, and then you can see the sending capabilities, and you can see the posting, and you can connect all your tools. But before that, you're not actually allowed to get into our platform. Host: Wow. Manuel David: We lose almost 70% of users that way. Host: Wow. Manuel David: But honestly, I like it that way. Host: Yeah. Manuel David: Because it's opinionated, and I don't want people in my platform who aren't serious about their outbound. Because they're going to send 100 messages, they're not going to get a whole bunch of replies- Host: For sure Manuel David: and then they're going to blame the tool. Host: Yeah. Manuel David: Instead, I would like the tool to kind of gate, a quality gate, the people that are coming into the platform. And then the people who are quality are going to enjoy the platform, because then they're going to be like, "Okay, I had this 20 message conversation with this agent and understands what I'm saying, understands what I should be going after, and it's actually given me some advice." Host: Yeah. Manuel David: "Now I feel like I'm capable to go and do my outbound." Host: And it will significantly bump up your response rate. Manuel David: Exactly. Host: I loved it. Yeah. And you mentioned about initial marketing attempt was trial and error. Maybe you target- Manuel David: Yeah Host: a group first, and maybe it doesn't work well- Manuel David: Yeah Host: and you try another one. Right? And how does your system also incorporate this feature of kind of learning from either a success or a failure- Manuel David: Yeah Host: and then make it to a more precise target later? Manuel David: Yeah. Host: How does that work? Manuel David: Yeah, so we have a multitude of ways that we do that. One of my favorite things that we do is multi-agent orchestration. Right? Which just means that we're getting some agents that are smaller- to then distill the knowledge into the- bigger agents. So if you send 1,000 emails, it's really kind of impossible to send 1,000 emails to a huge model without spending a whole bunch of money, right? And so what we do is that every single email that we're sending, a model's understanding why it sent that, and it's saving that into the database. Like, I sent this because, I sent this because. And then the model that's kind of bigger, the overarching model that is- trying to get better at sending emails and telling people what to do is looking at that and seeing, okay, this model sent that because of this. And 100 different models sent that because of this, and that specific thing is not getting a lot of replies. And so- let's kind of change how we're sending emails later on and- why we're sending and who we're sending to. So that's kind of how we have that feedback loop. Host: Definitely. And we want to move on. Your product, we understanding that it is a platform that helps either founders or their salesperson- Manuel David: Yeah Host: to outreach in a more efficient way, right? Manuel David: Yeah. Host: And one key question that jump out is either are you selling the result or are you selling a tool? Basically, the result is the number of success leads you get. Manuel David: Yeah. Host: And the service is basically you either do a subscription or whatever, like a month subscription, how many emails you sent out. And that's a service, right? And how do you see on your product on this point? Manuel David: Yeah. That's a really good question. We actually do a two-pronged strategy. Host: Okay. Manuel David: So if let's say you're a team that's pretty big. You could be like let's say you're series A and you're just looking for more meetings booked on your calendar, and you don't necessarily want to learn another tool. We will actually either go in and teach your SDRs how to use our tool, or we'll just run it for you, and then we'll give you the result. Host: Cool. Manuel David: We're doing that for a few different companies right now, or we'll use our agent and we'll do everything, and we'll run it for you, and then give the other person the result. But let's say you're a solo founder or you're an SDR, or you're someone who just kind of wants to play around and do outbound tools- then that's when you'll go to our platform. And then you'll use it, and then you'll see how it can help you, and then you'll learn from it, and consistently get results from it. Host: That's interesting. And another follow-on question on that is- Manuel David: Okay Host: because I think we are trying to figure out who are you selling it to. Manuel David: Yeah. Host: For example, if you are selling to the founder, to the boss, then probably it's more result-driven selling strategy, because they only care about how much cost I can save, for example- Manuel David: Yeah Host: for my sales team- Manuel David: Yeah Host: and how much revenues I can get from that. And on the other side, if you are selling to the salesperson in the company, then they probably don't want themselves to be replaced, and they maybe want a tool rather than a thing that can replace them completely. Manuel David: Autonomously. Yeah. Host: Yeah. So who are you actually selling to, and- Manuel David: Yeah Host: does that different in a way of your selling strategy or how you, the feature, how you define your product, I guess? Manuel David: Yeah. Host: Yeah. Manuel David: I definitely think that depends. And I think that kind of cycle is what we've been trying to figure out, and we have a really, really interesting way that we've solved it, right? And so for Paradigm, we like to have the whole end-to-end system right here. So you can start as early with us or as late with us as possible. Right? A lot of tools, either they do one part. Host: They do one part. Manuel David: Yeah. So they do the sending, they do the lead prospecting- or they do the posting on Instagram or whatever. We have everything. So depending on what you want from us is how you can use our software. You don't have to use it in one specific way. So let's say you're an SDR and you really, really enjoy looking at copy, changing your copy, and trying different methods. Then you can talk to Perry, and you can talk to our agent, and you can try and do those experiments, and- you can do all of that as you want it. If you're a founder and you don't really care that much about your copy as much- and you kind of just want something that's running autonomously and- and getting things on your calendar, then you can be a founder and you can be like, "Hey, look, this is kind of what my company is." Have that conversation so it understands, and then it's like, "Just run autonomously for me- and then do all those experiments and everything yourself, and just get meetings booked on my calendar. Here's my calendar." It'll do that as well. Host: Yeah. Manuel David: So I think it really depends on the user- how they want to use our tools, how they're going to use it, and they're able to do all of these things, right? So we actually don't compete with tools like Instantly or Apollo. We have multiple integrations with them. Because, well, our job is to help the user on what they feel like is not working for them, rather than just try and solve the whole piece of the pie, and like... There's a lot of people who are smart at Apollo and Instantly. So there's a lot of stuff that you can take from them. So if you want to use both platforms, please do. So it's not necessarily like we're saying, "All right. We have the whole end-to-end system. You have to use all of us, or you can't use all of it." It's take what specific part you need- use it as best as possible, and then go along with your day. Host: Yeah. And another deep dive on that direction is- Manuel David: Yeah, of course Host: can you share a little bit more about what the percentage of people, like who you are actually selling to, what percentage of them are actually the founder, the boss, and what percentage are, for example, the sales team or the sales representatives? Manuel David: Yeah. I'd say, the people who use the platform, it's overarchingly, normally there's probably people who are sales-focused. Right? Normally, I'm on conversation with SDRs and people who already know other tools, like Apollo Instantly, and are kind of familiar with outbound tools. And they might be using them, might not, or they're using the full Paradigm platform. And then the people that I was talking about with, who kind of really just want- meetings booked on their calendar, those, again, the people that I'll talk to, and then I'll run their outbound or someone on my team will run their outbound with our agent, and then we'll go ahead and deliver them the results. So I'd say mostly the people who are on the platform are people who have already kind of familiarized themselves with more sales tools. And then the people who work with us- and kind of get the... I'd say that's more of a higher ticket- are the founders who just want the results. Host: Okay. And on top of that, can you give us a real example of how the outbound workflow looks like for- Manuel David: Yeah Host: this type of population, using incumbent solutions- Manuel David: Yeah Host: versus how that would be different had they used Paradigm Outreach? Manuel David: Yeah. I'd say something that's super, super, super common is people, they go to Apollo, and then they go to Apollo, they have a whole bunch of different filters. They try and figure out who they should be targeting and who they should be specifically talking to. And Apollo doesn't necessarily have, in my eyes, a great agent where you can just be like, "Hey, this is who I want to target." But their filters are great, and so you'll get a whole bunch of filters from Apollo. You can't get signals from Apollo, which is something that's quite difficult. And so you'll grab all those leads, and they're kind of dated leads where you can get research on them, and then you'll bring them to Instantly. You'll go to Instantly, and Instantly will try and do the templated copy. You'll try- Host: To figure out who Manuel David: and figure out what copy works for you. Host: Right. Manuel David: It doesn't have great research. It's mostly just like their first name and- Host: Sending out emails. Manuel David: Yeah. So they kind of try and configure your copy, and then you'll go ahead and try and send through Instantly, and then let's say you'll put them in HubSpot, your CRM, to see, and you'll see, like, okay, has this person replied? This person's replied. Have we booked a meeting for them? Right? So it's a multitude of different tools, and then checking every single day, all three of those tools. And specifically in Paradigm, you can do every single one of those things. If you just wake up one morning and you talk to Perry, you're like, "Hey, I want to reach out to these specific people." Okay. It'll say, "Okay, we're reaching out to these people. Do you want to do a signal-based campaign?" "Yes, I want to do a signal-based campaign." Let's say I want to reach out for outbound. Okay. All right. Only target the people who are hiring SDRs currently. Okay. I'm only targeting those people right now. How many do you want? And it's basically, okay, I want 1,000. Okay. Reaching out to 1,000. And what it will do is it'll literally go and will surface all those signals, all the people who are hiring for SDRs currently, and then it'll reach out to them in those emails. And then when it responds, depending on what you've chosen, it'll either ping you with a notification that, hey, this person responded, or it'll actually respond to them for you and be like, hey, et cetera, answer their question, and then it will have a conversation with them, book them on the meeting, and then after the meeting will ask you if you've won them or not. And then if you've won them, we'll say yes, and it can then attribute it to a win, add it to the revenue. And then as you're going through the conversation, it will take them through your custom CRM, and it will even follow up and remind you like, "Hey, I know you met this person at the event the other day. Have you followed up with them? You've added them to the CRM. Have you followed up with them?" And it will send that to your Slack because you can have Perry inside of your Slack. Host: For sure. Manuel David: And so then your whole team can see when people reply, and then your whole team can see your wins and kind of what's going on with your outbound system, what's going on with your growth cycle, and all of that can systematically been done through Slack. If you want to just open the platform once to set everything up and then communicate with Perry in Slack and never- Host: Yep Manuel David: look at it again, you can do that. Right? And so everything is kind of there for the user to do how they please and what's most comfortable for them. Host: So from a raw lead to a booked meeting- which part inside this workflow is based on deterministic values? For example, meaning that basically, it's going to generate the same results no matter how many times you try it. Manuel David: Yeah. Host: And which part is based on pure AI reasoning and which part must it still involve a human in the loop? Manuel David: Yeah. I'd say deterministic, to kind of answer it first, I'd say the deterministic stuff is actually the infrastructure behind it. So like the spam prevention, how good is my inbox, how good has my sending been in the past? Am I going to end up in spam? Am I going to end up in promotion? Am I going to end up in primary? Am I going to end up in important? Host: Important. Manuel David: Sorry. We have algorithms actually that run every single email that you send to make sure that you're not landing in spam, and we'll actually do a seeded inbox test with inboxes that we own to see where that email's going to end up before you send. And so that it will be like, "Hey, this is going to end up in spam. We're going to regenerate it." So, those are the deterministic factors that we're- Host: Infrastructure Manuel David: the infrastructure, I would say. And then I'd say where it comes to AI reasoning is really the copy. Between the conversation and the AI and the research, there's a lot of AI reasoning there, which is the tricky part. How can I get my AI reasoning not to be stupid? How can I get my emails- Host: Right Manuel David: not to sound super AI? And honestly, a lot of people are like, "Man, my emails sound super AI." And I think honestly, I'm coming to a point where- I believe that is not necessarily the problem to fix, right? A lot of people who aren't very familiar with email think that it's like copy is the most important thing, like the words that you're saying to the user is the most important thing. But like you were saying earlier, it's actually who you're targeting. If someone sends me an email and it sounds a bit AI, I might not love the email. I might not love to receive the email. Host: Yeah. Manuel David: But I might read it if it's offering something that I really want to- Host: If it's an investor Manuel David: if I really want to have. Host: Trying to get an internship. Manuel David: Right. If it's an investor trying to give me a million dollars, I'm going to read the email and I'm going to respond. Host: Yes. For sure. Manuel David: There's a lot of emails that I've responded to recently that sound kind of- Host: Very robotic Manuel David: robotic. And businesses in general are robotic. And I'm not saying that our copy sounds AI, but I'm saying that that's not necessarily the thing that you should be trying to solve. It's more of like who you're reaching out to and making sure that they actually want what you're offering. Because if they do, then your copy, it matters a lot less than you'd think. Host: Sure. Manuel David: Because if you're giving someone what they feel like is a helping hand- Host: Yeah Manuel David: then they're going to be like, "Okay, yes. Regardless of what this looks like- Host: Yeah Manuel David: I want to get to it." Host: Yeah. Manuel David: And so I think it's really, really about who you're reaching out to. Host: Yeah. Manuel David: And so the AI reasoning part is understanding who you should be reaching out to from the research, and then creating the copy from it and sending it out to them. Then I'd say the AI, the human that wants to be in the loop- Host: Yes. Yeah Manuel David: honestly, I don't necessarily use our autoresponder too much just because I like really responding to people and then carrying on those conversations. And specifically, I believe we're never going to do anything where we're going to have an AI that takes the meetings for you. I think that's where the person should be in the loop. A person should definitely be talking inside of the meetings, understanding that person's business, and then further on carrying out whatever task they said they were going to do. I don't think that the AI should do the whole process- Host: Absolutely Manuel David: but I do think it should automate- Host: Yeah Manuel David: the things that are hard and make that easier. Host: So speaking of the personalization email, I think there's also a fine line that should be drawn on the degree of personalization of the email outline. Manuel David: Yeah. Host: If you're being too generic, people may refer to it as a template. Manuel David: Yeah. Host: But if you're being too personalized to the point- Manuel David: Yeah Host: that people may think you're doing creepy surveillance thing. Manuel David: Yeah. Host: So how do you determine the appropriate extent of the personalization so it's useful, but it's not like a creepy surveillance thing? Manuel David: Yeah. I think that it definitely depends on your vertical. Like if you're reaching out to somebody, let's say that you're an agency, you're a model agency, and you're reaching out to a model. And you're like, "Hey, I know this about you, and I know this about you, and I know all of this about you." The model's like, "I don't necessarily know that this is the agency that I want to be doing with." But if the agency reaches out to you and you're like, "Hey, I saw you did this event in the past. You did quite well. We'd love to speak to you more about how we could get you on as part of our team," or something like that, that seems like, oh, these people know about me. They've done research on me. They want to have a conversation. Host: I see. Manuel David: So I think it specifically depends on who you're targeting. And the flip case, if I get an email that it's like, "Hey, saw you did this event. I want to talk to you," I'm like, okay, yeah, maybe. But if I'm like, "Hey, saw you did this event. I know what your software does. This is specifically why I want to talk to you because your software does this, and you're in this vertical," then for me, I would love a more- Host: Absolutely Manuel David: personalized email because I would understand that, okay, that means that someone either did their time or whatever- Host: Yeah Manuel David: the system that they're using- Host: Yep Manuel David: did their time to reach out to me. And so that- Host: Yeah Manuel David: I think it really depends on what type of vertical you're reaching out to. Host: That's so well said. So blueprint is something the user needs to set up themselves, right? Manuel David: Yeah. Host: Just this clarification question. What is blueprint, and what is included in the blueprint? Manuel David: Yeah. And so, what we like to say is there are no templates in the platform. And so blueprint is basically a conversation where the AI is having with the user, and then it saves all of that knowledge into what it says is a blueprint, and then you can use that blueprint to kind of like you're saying, make it more deterministic, the type of output that you want. And so if I find a blueprint, which is basically, let's say, just a conversation- Host: It's like a context? Manuel David: Yeah, context. Host: Okay. Manuel David: Stored context. If I find a blueprint that works for me a lot every single time I'm doing campaigns or reaching out to people, then I can consistently use that blueprint without having bad fidelity and it changing a whole bunch. So I can consistently reuse it and reuse it. And if I want to try something else, then I'll create another blueprint and I'm like, "Okay, it's time to try this. Let me create another blueprint, and I can try this other experiment." So those are kind of what blueprints are. It's not necessarily like a user's out there typing- but it's more of the conversation that it had with the AI so it can understand what specifically they're offering to the people that they're reaching out to. And so that when it reaches out to certain people, that it can go ahead and say, "Okay, this is the conversation that I'm pulling from. This is how I'm supposed to be kind of doing the outreach." Host: Got it. Manuel David: "Let me go ahead and do it." Host: Got it. Totally. Yeah, and I think one very important part, as you said, is the trust actually during the sales process. As you mentioned, agents can maybe do the outreach, sending the emails for you, booking the meeting for you, but the talking to the person, building the trust between business partners, I think still should be done by a human. Manuel David: For sure. Host: And how does the human participate in the whole process of your product? Does your product stop by basically after booking the meeting, it stops there, or does it engage further in later process? Manuel David: Yeah. Like you're saying, we say get more booked meetings. That's our motto. Get more booked meetings. Paradigm, get more booked meetings. And so as far as what we do, yeah, once we get that meeting booked on your calendar, that's kind of where we stop and we're like, okay, we're passing it to you. We're hands-off from here. But like I also said, we also do posting, so posting on Instagram, posting on LinkedIn. If you want to post about some events that you went to recently, and use that to get more clients, that's a way that you can do something that I guess is a little bit further from booking a meeting. But yeah, the whole goal of the platform is to get people on those conversations with other people, because we think everything up until that is just kind of a manual slog that you don't really want to do, like writing emails- Host: Yeah Manuel David: and doing research. You don't really want to do that. Host: For sure. Manuel David: And it's really the having conversations like we're having right now, those are the parts that people want to do, and those are the parts that we think is important. So that's what we give to the human, and we let the AI handle everything else that can be automated. Host: Yeah. As you mentioned, because we see there are kind of not strictly similar products, but they actually targeting a kind of broader context. For example, the Boardy.ai. Manuel David: Boardy, yeah. Host: Yeah. I think what they do is connect you with the right person, right? Manuel David: Yeah. Host: I think, but they're not necessarily only targeting sales or outbound- Manuel David: Yeah Host: inbound perspective. They also handle other stuff. Manuel David: Yeah. Host: Maybe just you want to find a friend to chat, they will introduce you- Manuel David: Yeah Host: to them. But do you see them as another form that probably your business is within their scope, and would they be one day kind of be a competitor- Manuel David: Yeah Host: to your product? Yeah. Manuel David: No. Actually, I love Boardy. I think it's a really cool software. I use Boardy myself, just because I think it's a really cool system. I think that the difference between me and Boardy is Boardy is mostly about networking- about connecting people, whether it's business-wise or, like you said, you're just trying to do a friend. And you're connecting people, not specifically doing outreach and trying to get them on your calendar, right? I think Boardy is a really cool system that is kind of adjacent to mine, but very much different. We do posting. We're kind of like a catchall, right? Come into our platform, post through our platform, send emails through our platform, send LinkedIn through our platform, and get more clients, get more booked meetings on your calendar. Specifically what Boardy's trying to do is just like, "Okay, how can I get this person introed to another person?" Host: Yeah. Manuel David: So, I'd say that's where we kind of differentiate from Boardy, because it's cool they give you intros- but a lot of times my texts with Boardy are kind of like, "Hey, these people are in your same realm. Would you want to talk to them, maybe get more information from them?" None of my clients have come from Boardy. Right? Just because Boardy is kind of just like, "Hey, do you want to talk to this person?" Ours is specifically business-focused. How can we get clients- Host: Yeah Manuel David: people who are actually going to buy whatever you're selling- onto your meetings. Host: Definitely. And do you consider a voice agent being a next step of your product? For example- Manuel David: Yeah Host: the voice agent that can directly do the customer call for you? Manuel David: Yeah. Host: Do you see that down the line? Manuel David: I love that. I actually very much enjoy voice agents, and one of my favorite things about building out Perry was the conversation that you can have. It's an actual voice conversation. As far as, we've thought about cold calling and conversations, but where we think that technology is right now and where we feel as a company humans need to be in the loop, we don't necessarily see ourselves going and diving into anything that where the AI is having conversations and speaking to the lead. We just don't find that valuable right now. But I think that as technology advances, that's something that we can definitely take a look into, and we'll decide if that's something that we want to dive into or not. Host: Yeah. Definitely. And another question is- Manuel David: Yeah Host: we are curious about how to actually find the potential customers that you should send the emails. Do you just look on the public internet information? Do you kind of go through LinkedIn, or how do you do that? Manuel David: Yeah. Our system does a lot of things, right? We have a multitude of different AI APIs that we use- and data source APIs. And so we actually triage maybe hundreds of APIs. Host: So a lot. Manuel David: And then from that information and from that research, we have a synthesis process that takes the most important things- finds the best signals, and then passes that to the email writer. And then that email writer then creates the email and sends it, and then it also has a lead, of course, and sends it to them. So we actually kind of do what we like to call a funnel, where it's like we got all of this information, all this messy data, and then we continue to aggregate it until we turn it into one person. This is the best person for this specific query, and now we're going to send it out to the email, and then we do that thousands of times. Interesting. Host: So I think there's, just to follow up on that, so you're building like a personal engine for outbound. But then you also have to do outbound for the product itself, right? So how has the experience of doing the outbound for your product shaped the way you build your product? Manuel David: Yeah. I'd say that when we, the beginning of the year, we were very heavy into dogfooding our own product because we were launching and wanted to make sure everything worked, so we were doing a lot of outbound for ourselves. And we even got on the phone with people with the likes of Sweatcoin and- Host: Oh, wow Manuel David: from doing a lot of outbound and posting. So that's where we got a lot of our first clients, and it was working well. Like I said, the product works. It was working well. But there was just kind of like a way that I felt from the people that I was talking to that they weren't in San Francisco, and they weren't necessarily the specific tech people that I want to be talking to, and I didn't feel like they were going to be pushing my product to where it needed to be. And I didn't feel like I was going to need to innovate with them. Whenever I have a new client, I like building something new. I like hearing about different use cases and new use cases. And so I kind of turned down my outbound and started doing more inbound stuff, so doing events, posting on LinkedIn. And from doing a lot of that posting on LinkedIn and posting on socials, I started to realize that that's something that a lot of people use as a channel to get on meetings, and that's why we decided to turn it into a distribution engine instead of just outbound. Host: Right. Manuel David: Because specifically posting, it gets you even more people. Host: Absolutely. Manuel David: And so then we added it to our system. Host: Yeah. Manuel David: And we're like, "Okay, yeah, this is perfect." Host: Yeah. Manuel David: And so now we can post, and I post, and people will DM me like, "Hey, do you think I can get a free trial?" And I'm like, "Yeah, go ahead." And so- Host: So are you posting using Paradigm yourself? Manuel David: Yeah. Host: Cool. I should try it too. Manuel David: Yeah. Host: Yeah. So, again, not trying to roast you, but- Manuel David: No, of course Host: do you think your product will still have the moats that it has right now after the next major, maybe foundational model update? Manuel David: Of course. Please, bring them all. Bring the best model out because all it does is make my product better, right? We try to use the best models as we can so that we can keep the human aspect of our copy in whatever we're doing, right? And since they're like, if Opus 4.9 came out or Fable 5 finally comes back- Host: Oh, wow Manuel David: that won't do anything to necessarily my moat or anything to my business model, because with Fable 5, yeah, you can send 10 to 15 emails, but try sending thousands and thousands of emails through a system that has a lot of infrastructure and not landing in spam. And I think that there's a lot of knowledge, and I think there's a lot of engineering that happens inside of my system that I don't think is necessarily easily replicated with just a new model coming out. Host: Yeah. Manuel David: Right? I think new models can help people write better emails. But I don't think necessarily new models themselves, and Anthropic's focused on how can I like-... destroy the sales cycle for me. How can I destroy all of the sales SaaS out there? Host: Correct. Manuel David: Right? Um, but yeah, with new models coming out, I think they do more help than they would do harm for me, specifically just on the basis of they would allow me to figure out different ways where I can write copy that sounds more human, more non-generic and do that instead of, I don't know, Fable five figuring out how to send-... like 10,000 emails Monday through Friday. I don't know. Host: So where exactly does the moat of Paradigm Outreach lie... You talk about infrastructure. I assume that's one of the biggest one. Manuel David: Yeah. Host: Anything else like data distribution or something else? Manuel David: Yeah. Um, I'd say a multitude of those. So, um, infrastructure is a really difficult one. I was actually talking to someone a little bit ago, and they said that they're trying to build a sequence- Host: Oh, God Manuel David: and I said, "Good luck." Right? Um, specifically there's a lot of technology that goes behind like creating inboxes and creating domains, and there's a reason why there's not a lot of like- Host: Sure Manuel David: websites out there that do it, right? There's a lot of websites where you can send and just like, yeah, you can go and get your emails, and you can send, but there's not a lot of websites out there that do the domain and the inboxes for you because it's very, very difficult, and it's very expensive. We actually started as a email infrastructure company, and from that we then grew Paradigm. And so, um, because we have that knowledge and because we understand how to build good systems and send out emails and, um, create inboxes that don't land in spam, I think it's very, very difficult for other people to do it. And so for the data as well, we have a lot of data on, um, the people who are using our platform, what's gotten replies. And like I said- Host: Ooh Manuel David: um, in AI, every single time it sends an email, it says why it sent the email, and so we can save that. And then as we aggregate more and more data, or, uh, it's kind of becomes a self-serving loop. Host: Better at like interpreting the signals, right? Manuel David: Yes, exactly. It becomes better and better at, um, understanding what words specifically is going to get people to respond. Host: Interesting. Interesting. Um, how did you first validate, is it actually like a product market fit- Manuel David: Yeah Host: for this product? Manuel David: Um, I'd say that, uh, after, specifically for us, our company, like when I was in college, it was working well for us. Uh, we had a few different clients because we had an AI automation company. That was what I was using Paradigm for. It was an AI automation company, it was an AI agency, and we had a lot of people who were trying to automate like their lead sourcing and like their lead prospecting and that kind of cycle. Um, so we started letting them use it, and we started giving it to them. And so whenever they were like, "Oh, please, we need more features. We love this so much," I think that's when I first was like, "Okay, yeah, this could definitely be something. Maybe I should take a break from school- Host: Yeah Manuel David: and, uh, definitely go big." So. Host: Yeah. And how many users, um, or even like paid customers, um- Manuel David: Yeah Host: do you have on the platform right now? Manuel David: Yeah. Host: Feel free to just give like a rough number. Manuel David: Yeah. So we have, uh, around 100 users in total, um, across like free trial, um, paid users and, um, uh, like the enterprise clients, so. Host: Interesting. And are you seeing... I'm just kind of curious to figure out, like do you see any particular type of population that it's like easiest to sell a product to? Manuel David: Yeah. Host: And is there any population that just churn more easily than others? Manuel David: For sure. I'd say probably the easiest population to sell our product to is people who've already used sales tools before because they can really see like, "Oh, wow, this is really impressive. Like, I can do a lot here, and, um, it's very easy because I don't have to do a multitude of different tools." Um, I'd say the hardest is probably the founders who want to use the tools themselves because- Host: Oh Manuel David: they just want to go fast, and they just want to do everything super quick. And our system kind of wants to slow people down and understand the process more. And so I'd say that's probably, uh, the biggest people who churn are the people who just want to go super, super fast, and they just like want to send an email out 10 seconds- Host: Makes sense Manuel David: in. Host: Oh my days. Manuel David: Um, uh, and- Host: It'll get you blocked, right? Manuel David: Yeah. Right. And so, um, those type of people is very difficult to kind of like retain them just because they want to do speed. Um, but the people who understand that, okay, like as I build with this platform, I can do a lot of volume- which a lot of people like. Um, you can do like 10,000 emails if you want, or I can do a lot of volume here, and there's a lot of stuff I can do here that I can't do on other platforms. I'd say those are the easiest people to kind of talk to, like the salespeople. So yeah. Host: Yeah. And I think, uh, let's imagine a scenario where like everybody's already using Paradigm there. Manuel David: Paradigm. Yeah. Host: And, uh, let's say there's 100 people in the room. Each of them are using it, and they are sending 100 email to each of them each day. Then each of us will receive like high, uh, like 10K emails every day, right? Manuel David: Yeah. Host: Then, uh, would it be kind of... I already is feeling kind of like information overload right now, even without that many of emails coming every day. Manuel David: Yes. Host: But do you see that if that's gonna happen, like, uh, these kind of AI outreaching tools being more popular, then people may need some AI to read the inbounds for them- Manuel David: Yeah Host: to filter down to a more specific thing that a human can focus on. Manuel David: Yeah. Host: Yeah. Manuel David: Um, personally, I completely think that that's where our world's heading- Host: Okay Manuel David: because, like I said, like there's amount of research that I can do on a lead to see if they fit. Um, but if someone has a business that's close to me and that lead fits specifically for them too, there's nothing I can do to stop that person from reaching out to them. I think, uh, there's gonna be a world where there's gonna be a filtration of like basically all of the information that goes into us, and Google's already doing it. They're training Gemini so that it can kind of like filter out the emails and give you the ones that actually are important and the ones that make sense for you, um, knowing like what you do and what emails you open. Um, so I do think that there's gonna be a future where there's gonna be a huge filtration of just emails and content in general- and allows the user to kind of just have more peace of mind and only see the email that they feel like is, uh, kind of helps for them. So yeah. Host: Yeah. Manuel David: But in that world of everyone using Paradigm, I think- I think that'd be an amazing world. So yeah. Host: Yeah. Manuel David: Let's get there soon. Host: Yeah. And another follow on that is actually, for example, if the two of us are doing business right now- if you are, for example, selling Paradigm- and I am maybe a potential customer of you- and let's say I'm also using Paradigm- but you're just selling some other products other than Paradigm, right? And would that be, for example, my Paradigm agent receive your Paradigm agent's message, and would it be kind of... How do I say that? It's like you're playing a game where you're both playing the player's role and also the judge's role. Manuel David: Yeah. I love that kind of a catch too. I love that. Yeah. Our system can do a lot. So go ahead if two people are on Paradigm, one person from Paradigm reaches out to it- it will definitely recognize, "Hey." Host: Yeah. Manuel David: But yeah, it can do the responding for you. I kind of urge people to, for inbound stuff, I normally tell my agent to just quiet down on that stuff because it will text me of like, "Hey, someone I'm letting you know on LinkedIn or email- and they're reaching out for this." I normally tell it like not to respond to those things, because I want to look at them and see the validity for myself. But if you do say, "Hey, Perry, I want you to respond to my inbound stuff and let me know which one makes sense," it will go ahead and do it. So yeah, I definitely think that that's something that could go ahead and happen. Host: Like two agents talking to each other. Manuel David: Yeah. Two agents talking to each other. Host: You have both the buyer and seller market. Manuel David: Yeah. And I think that's kind of where the world's come to. You never really know if you're talking to an agent or a human, and it's getting less and less important- the distinction. Host: Yeah, definitely. And yeah, we're already touching a little bit on the future you're looking at, but on a little bit more recent, like a shorter period, like in maybe 6 months or 12 months time, how do you see Paradigm going towards? Manuel David: Yeah. I think in 6 to 12 months time, I kind of see Paradigm as being in a place where you can come and you can kind of just offload your whole work cycle to the agent, whether it's posting, whether it's doing emails, or whether it's doing even more, right? I have a lot of experience building agentic systems and workflows, and I think that Paradigm, as being my brainchild, I think that there's a lot more that we can do than just sending emails and just posting. And so I think that as we get more and more clients and as we get more and more different use cases for how people get their clientele, I would see Paradigm running its own multiple social media accounts- while also doing multiple campaigns and multiple different things, maybe even posting like YouTube videos and- and a lot of different client acquisition channels. So that's kind of how I see our future doing, just going into multiple different funnels and figuring out- Host: Right Manuel David: how to optimize their strategy, and only doing the one that's best for the company. And- Host: To be more comprehensive, but also selective. Manuel David: Yes, exactly. Host: I do remember at the beginning you mentioned that you're also exploring a possibility of image generation or video generation. Manuel David: Yes. Host: Is that where it will fit in? Like for example, generating YouTube Shorts or Instagram Reels. Manuel David: Yeah. Host: Yeah. Manuel David: Yeah. So yeah, we're doing that right now and we're trying to find some partners and some people that are interested in posting- using our platform so that we can kind of see how it works. And we're not very experienced yet with an image gen, video gen- so we're trying to see what people want to use it for. But I know UGC is something that a lot of people do- and so we're currently creating UGC content- and we're trying to look very closely on what the output are- Host: Yes Manuel David: from our users- Host: Yes Manuel David: from that are creating the image gen and the video gen so that we can see- Host: Yes Manuel David: is this something that we can kind of master as well? Host: Yeah. So this question may be out of nowhere, but if you were to describe yourself in a word- that's irrelevant to what you're building, what would it be and why? Manuel David: Faithful. Host: Faithful. Manuel David: Yeah. Host: To what? To whom? Manuel David: I'd say faithful to God, because I'm a big believer in Jesus Christ, and I think that I took a big step and a big leap- leaving college, to just come to San Francisco. Host: Yeah. Manuel David: I'd never been to San Francisco before. Host: Yeah. Manuel David: And it took a lot of prayer and it took a lot of faith. Host: Yeah. Manuel David: And I'd say that I've been very faithful and very excited to see what God has planned for me and has planned for my life. And I'd say that doesn't necessarily have anything to do with what I'm building, but- it's a big part of who I am and why I am here- Host: Right Manuel David: and what I'm doing- Host: Right Manuel David: things that I'm doing. Host: That's a huge trade-off that you made, either financially- Manuel David: Yeah Host: or socially or personally to pursue building- Manuel David: Yeah Host: in San Francisco, right? How do you decide whether it's worth it? Or maybe you don't need to decide that. Manuel David: Yeah. Host: What's your take? Manuel David: There was a lot of talking, of course, with my parents as well, and my brother, and praying a lot. There was a lot of prayer. I don't necessarily know if there was one deciding factor that I was like, "Yes, I for sure need to come to San Francisco." But I felt like there was, like I said, God calling on my life to be like- "This is what you should do. Take this chance." And then looking back, I'm like, "Wow, I'm so glad I took this chance." And there's a lot of things that happened in my life after I came to San Francisco that wouldn't have happened, and I was like, "Wow, I would've missed the most amazing things." Host: Like this podcast. Manuel David: Yeah. Yeah, exactly, like this podcast. I would've missed the most amazing things and meeting the most amazing people, and it's like, I really truly feel like this is where God sent me to be because I know how I was in college and I know how I felt, and I know what my direction was, and I felt like I'm living more in my potential coming here than I was staying in college. So yeah. Host: Yeah. And I think this is the last two questions. Manuel David: Of course. Host: This one will be, as you are playing the role of a founder, a CEO, what are some of the largest mistakes you made? Manuel David: Yeah. Host: Or that leads to maybe a not so pleasant result, right? Manuel David: Yeah. Host: And how do you resolve that if there's- Manuel David: Yeah Host: bad consequences? Manuel David: I'd say I've made so many mistakes. We ship fast at Paradigm, and we iterate. It's break fast and fix fast, learn fast. So I think that there's a lot of different mistakes that I've made, whether it's just completely breaking the system, whether it's creating something that spent $200- Host: Oh Manuel David: and the emails were garbage. I've made a lot of mistakes, and I think that what I've just learned from most of these mistakes that I've made is that it's okay and as long as you learn from it. If you make a mistake and you don't learn from it, those are the most costly ones. So if I were to give you one specific mistake, would probably be just not believing in my system when I should've and not telling people about it when I felt like I should've. Because I've talked to a lot of people, and they've been really excited about Paradigm. And I've kind of felt like, "Oh, maybe it's not ready yet. I don't know if it's going to break. I don't know if it's going to work for this person." Host: Is that the imposter syndrome thing? Manuel David: Yeah, kind of imposter syndrome and kind of second-guessing yourself. And so I'd say those are probably the biggest mistakes, not believing in myself when I feel like I should've. That have definitely cost me the most. So, yeah. Host: Yeah. And for our audiences out there, they may be student founders and young founders. They still feel uncertain if they should take the leap of faith- Manuel David: Yeah Host: into building or doing their own startups. What would you say to them? Manuel David: Yeah, don't drop out. Host: Woo. Manuel David: Yeah. I don't necessarily recommend dropping out, but I do recommend taking a chance on yourself. And I think that the only thing that you can do in this world that God has given us is take a chance on yourself. So I think that if you're out there and you're building something or you feel like you're excited about something that you're building, tell people. Tell your mom, tell your dad, tell your family, tell your friends. Maybe tell your professor because the biggest thing that you can regret is just not telling people, not getting it out there. Because there'll be a lot of noes, and there might be a lot of yeses. But you can't regret not telling anyone. So yeah, I'd say that's my biggest advice for everyone out there is just tell people about your product, tell people about what you're building, and then take a chance on yourself because if you won't take a chance on yourself, who will? So, yeah. Host: Definitely. I think that's exactly what we want to do. As the podcast- Manuel David: Yeah Host: we want to share the story of founders to others and let more people know you, know your product, and maybe also help investors to discover these high-potential founders. Manuel David: Yeah. Host: And yeah, thank you so much for joining us today. Manuel David: Yeah. Host: And we'll glad to see what you're making progress in the future- Thank you so much, Manuel and wish you all the best. Yeah. Manuel David: No, thank you so much for having me. It's been lovely being on the podcast. Host: Yep. So folks, stay tuned. This is the wrap for this episode, but not the wrap for the entire 18VC podcast. Manuel David: Woo. Host: So we'll have more episodes coming up. Stay tuned, follow us, and subscribe to our channels across YouTube, us on Spotify, and Apple Podcasts, and LinkedIn as well.