15 Minute Founder E8: Starting a Company in the AI Era

In this episode of "15 Minute Founder," Alex Levin and Rebecca Greene explore the game-changing impact of generative AI on building and scaling startups. Reflecting on their journey founding Regal, they share how today’s AI tools could’ve transformed everything from product development to sales outreach—maybe even creating a company with one employee and $1B in revenue.

The discussion dives into practical ways AI can amplify productivity, from faster ideation to automating tasks founders usually take on. They also touch on how early-stage startups today are using AI to do more with less, reshaping the way businesses grow.

Main Topics Covered:

• Using generative AI to accelerate product ideation and research
• AI’s role in improving engineering efficiency
• How AI can automate outreach and customer-facing tasks
• Reducing hiring pressure with AI agents and tools
• Leveraging AI for more efficient business operations
• Real-world examples of how AI changes startup dynamics

Transcript

  📍  📍  📍  Hello, this is Alex Levin, the co founder and CEO of Regal, and I'm here with my co founder Rebecca. So today we wanted to chat a little bit about starting a new company in the generative AI era. So when we got going, really generative AI wasn't at a place where it materially impacted what we were thinking about, you know, as we started the company, but you know, our friends that are starting today, I'd say, The number one question they get from investors beyond, you know, are they using AI in their product for their customers is how are they using generative AI to change the way they're building their company?

And sometimes this is framed by, um, you know, a few people. And so Altman, the CEO of open AI often likes to frame it this way is, you know, when will there be a company with one employee and a billion dollars of revenue, right? How much can generative AI. Do the job of a company so that you don't need as many employees perhaps to generate as much revenue So, I don't know as you think about this Rebecca Where are the places maybe that you know, you think you could use generative AI today as you start a company 

I certainly think of the very early phases when we were doing more ideation around the company and more research and trying to validate like our hypotheses about the company. That's an area where we took a lot of our own time to do it, and I think we could have, you know, I would have never thought to outsource that to a consultant, but I would think about just giving ourselves much more leverage and productivity.

In that, um, so that might have been like researching competitors and what their products are able to do. We had to do hundreds of demos of competitors to convince ourselves that our vision for how we were doing things would be really different. And I think that could have gone much faster as an example.

Um, that's 1 area I think about. I also think, I mean, about even just building our products. I think we probably would have hired much, much slower on the engineering side than we did, um, initially for the business. 

So you think today basically the AI sort of generative AI tools allow Engineers to be much more efficient in what they're doing 

I think it allows them to be much more efficient. So I think you get way more output and leverage from a single engineer. That engineer still has to be a great founding engineer who can do many different things. That doesn't change that. But I think about at the beginning of our company, there was, you know,  Yeah.

Some just like POCs that engineers, like our best engineers had to do. And it's those kinds of things where, again, I never would have thought to contract it out on a marketplace, but where I think like you can, we can now find AI agents or build AI agents who could have like created a lot of those POCs for

It's basic apps. Yeah. I mean, definitely the other one for us is like we made a decision very early because we're in the context in a world that there would be human agents that would sit at our customers. There were plenty of people coming to us and saying, Hey, can you just do this for us? And in a world where like we didn't want to hire human agents, that was impossible.

But in a world where you have AI agent, it changes. So There have been a number of posts recently that, you know, I like that basically are saying software companies used to use software to do a job for their customer, and it was limited because there's only software, but now that there's generative AI that can replicate a lot of the tasks of humans, you can actually take off, you know, bite a bigger chunk of stuff off.

So instead of just like the software that helps with sending a text, you can actually. Right. The text or maybe the one that responds to the next text or plans. What's the next best action? And so that allows you to then take more of the job to be done from the customer and potentially charge the customer more for it as well.

So I think like it would have changed the way we looked at the product we built. And so I see that a lot in early stage startups now that our friends are building where you. Instead of biting off just the software piece, they're going after both the software and the labor cost or revenue, whichever way you want to look at it, which changes the dynamic of like what they're building.

And then I think also a lot about what we're starting to see now where like we have built, uh, within the marketing org, basically like the regal LLM, like you can effectively give it stuff and it'll like write it in our voice. I mean, that would have saved so much time. Like if we could have had the, like Alex and Rebecca LLM and like started writing content much faster, you know, for LinkedIn, for blog posts, for different pieces, for the website, and just said, like, spin this up and try it.

Not that it's the final version, but it would have allowed us to like move more quickly and not waste time on some of those areas. Um, I mean, I also think a little bit about,  uh, the, the sort of customer facing jobs. Like so much of the customer facing job is like aggregating everything that's going on, understanding it, and then like creating a deck and like presenting it back to the customer to then decide on the next step.  haven't, to be fair, I haven't seen it done yet, but I don't think we're that far from a world in which you can basically say right now there's AI that says like, how many calls did you do last month? Or like, help me make a slide with these three points. It'll do that individually, but I don't think we're far from a world in which you can say,  You know, basically like ingest all my emails and just all the like voice, you know, the transcripts of conversations we've had on zoom and based off of that, like, what are the three goals for the company?

Give me a report showing the outcome and, you know, put up a page for us to have like a, a brainstorm about what the next three things are we're going to do together. And instead of like having to have a human being do all that work, like I'll just have it do it for me. And then I can like present that and then immediately move much faster and have a much sort of smaller team that's customer facing. 

It's funny, customer success and even implementation more so would have been the one area that I also hesitate to say that I would have done anything different in, in terms of, you know,  Us doing it for a very long time at first to implement customers, understand their pain points, their needs, what's going well, what's not, because it's also the bridge between the customer and the product.

so it really, not only is it about like convincing the customer with these three ideas and understanding their business, but it's also that feedback loop to like, are we even building the right product?  

definitely want to be involved. I'm more think it's more. It's not that the A. I would do the conversation. It's that it's a prep for every meeting. So like if you have one customer like great, I can go and do all the prep for it. But now you have 10 and you go shoot. Do I have time to do prep for 10 meetings to have a great meeting to get that feedback?

Well, no. So I need to hire a person. So if instead the I can do all the prep for the meeting and I can now do 10 meetings without another person. I think that's going to be material. Like I say, I haven't I haven't Seen it all the way there, but I've seen some startups working on this idea to see how you can aggregate all the data and start getting that in one place and starting to make those, those, uh, decks and sort of things.

I  you know, when, when  I make, when I would make the financial statements, I think it's also materially different because, you know, it's what we've talked about, there are so many opportunities to do this with fewer people, you don't need to actually.  Spend as much money. So I think the way you sort of build a startup at the beginning can be much more efficient, either running off of your own sort of money being like ramen profitable or whatever for longer, or if you're going to raise money, uh, you know, raise less money and take less dilution, or, you know, if you're going to raise the same amount of money, have that last you for longer, that's a material change to the logic.

Like, I think the mistake that I'm seeing is people are raising the same amount of money, if not more, and spending it. The same amount of time or less, which is bad. Like, you know, if you thought 3 million before in the old world should last you 18 months, now 3 million should last you, you know, 24 months or something.

Otherwise, if you're just spending the money at the same speed, you're not like taking advantage of like the leverage that you're getting from the cost savings. 

Yeah. I also think the advantage of feeling like you don't need to hire as quickly, I still think hiring your first. Sales person, your first CS person, like those things still, your first engineer matter. And I.  Think you should be as thoughtful as possible there, but it's, it's about being able to give them as much leverage as possible too before you feel compelled like, oh, I've gotta hire five engineers now, five, six, because that's where things start to like very quickly get out of hand. 

when you feel like you're the very first hire, like you've trained them, they've seen how you've done things, you've learned, it's a, it's a back and forth, a give and take. And then when you turn it on, you raise that round and you're all of a sudden, like every team now is 10 people. That's where

Yeah. much harder. Yeah. The other one I think about, and this was not my line. Uh, I was listening to somebody on the radio and said, well, you know, why is AI trying to. Uh do the things I love like paint and draw and whatever why couldn't it do my laundry and my dishes? And so like I think a little bit of that also in the startup world Which is so many people are going after these like very Uh high end use cases and I understand why but to some extent part of what was hard at the beginning of a startup Was just like after my day job, I remember for like hours at night I would do all the things like that are the you know Yeah.

Think about like, you know, HR and onboarding. Think about all of the legal stuff. Think about all of the finance and accounting stuff. Think about all of the, like, I don't know, extra work that goes on the, you know, I don't rent and like office set up, like the stuff that happens. So like, to some extent, if I could have AI do my laundry and my dishes, so I could still focus on the customer facing stuff, that would have been a massive.

Transition. The other place I think about that sort of general concept is, you know, it is a massive unlock at every company. I've been at when I've had, uh, an executive assistant. That's really been very deep in my workflow. So not somebody who just sort of separate, but literally like in my email, you know, on calls, helping sort of move things forward faster, understanding the context of what's happening.

And I think still that can't be done by a I yet, but I don't think we're far from a world in which Transcribed Instead of just the executive having that, which is like the premium sort of like human being, every person in the company has something that's less but better than nothing, right? And has like a person that basically listens to all their calls that you know, they can take voice notes to and say, Hey, please, you know, write this email to three people and set up these next three meetings that automatically acts on And so the leverage created from that would be enormous across the company because everybody in the same way I have my executive assistant, they would have this leverage as well. Yeah. 

Yeah. Just making every single person much more efficient at their job. And the pieces that candidly, like are hard to do, no matter how smart, how much you focus on the business content and all of that, it's just the pieces that have to get done.

Oh, yeah, I like can't tell you before I had any how much of my time was spent scheduling and like basic stuff that again wasn't the highest and best. That's why I said, like, the, you know, that's not the only one, but the laundry and dishes analogy, like, you know, so how do you make sure that as a founder, you're set up in a way that.

You know, all that stuff can be done as quickly as possible. So you can focus as much of your time on what's really happening. I mean, the other one I think about is like a lot of founders, you know, haven't done this before. Like we were lucky we had, but they're faced with something and they'll spend two weeks stressing out about, well, how do I make my first hire or how do I open an office for the first time?

Or how do I fire somebody the first time? And if, you know,  They have advisors that can help them, or if they had AI that could help 'em understand what to do in those situations, just massively speed up those so that they're actually focusing on the things that are gonna move the business forward and not just these very operational pieces. 

yeah. One area you interestingly have not mentioned, like getting much more leveraged out of AI. In the early days, if we had done it, um, is in sales, would you have done anything differently? We

to your point, I like the customer interactions. I think that is the critical piece of understanding, like how people are reacting. So I would've still done all of that.  There's definitely.

did a ton of the outreach ourselves.

Yeah, so there's definitely much more automation in the outreach that I think I could have done better with AI.

So like there were, I've told you this story before, like I should have hired the first BDR earlier with the way I used to frame it. Now I think about as I should have hired the AI BDR earlier. So somebody to help me with, uh, Identifying prospects, sending the first email, you know, maybe doing a little bit of like an outbound phone call, you know, not crazy things, but would have given me a little bit more leverage, especially in weeks where I couldn't do it because I was doing other things so that that would have been more consistent.

So for sure, like, As founders today, like I expect people to have those sort of systems set up and now we've set it up because, you know, now the AI is better, but really had none of that when we were going outbound at the beginning.

Yeah, 

So yeah, you know, if you're starting a company today, you know, and your investor asks you about AI, it's not only about your product.

Like it's also about how you make yourself more efficient. Um, and I hope you're using this to again, you know, have your funding last for longer. Um, as always, you know, feel free to reach out to us at hello at regal. io. Thank you, Rebecca.

thanks, Alex.