An 8-Year Old Can Build An App. What Does That Mean For The Future of Building?

Monisha Saldanha (00:00) Welcome to Colorado Tech People, the podcast where we talk with the founders and innovators building companies that shape how we live, work and connect. I am your host, Monisha Saldanha an executive with 15 years of experience in product management. Today's episode explores how AI is impacting the launching and scaling of startups. And I have with me here today, our guest, Ala Stolpnik.

Ala Stolpnik (00:28) stuff.

Monisha Saldanha (00:28) the CEO and

founder of Wisary We're going to talk about how software development has changed even from two years ago to today, how you can do more with less people and what the future brings for our children. So very excited to dive in. Ala, thank you so much for joining me today.

Ala Stolpnik (00:49) Thank you so much. Super fun to be here.

Monisha Saldanha (00:51) Great. So first question for you, what was the problem that Wisary was created to solve and what is the solution?

Ala Stolpnik (01:00) The journey started about three years ago and back then, ChatGPT just came out and I realized that the way we build software is going to change and started to think about how AI can help us with our existing processes, with the ways we use to build software. So basically what Wisary is doing, it's helping product managers think more thoroughly about what they want to build and why, and to make sure that

they create better requirements and it not just using AI to kind of accelerate everything they're doing. And as a result, when we know what exactly we want to build, engineers can go and build it. That was three years ago. A lot has changed in the AI world. And we will talk about that a bit today about how that old world that I started building for has been changing and what's the future for software development in general.

Monisha Saldanha (01:50) Yeah, well would be great to hear a little bit about that now. Can you tell us about the changes that you've seen?

Ala Stolpnik (01:55) So, so there, there are two things. There's one that what we're seeing in the industry, all the vibe coding, a kind of hype that now suddenly everybody becomes a builder. So suddenly you don't need to be a professional product manager or professional engineer to start something. You still may need experts to turn it into reality, but you can, you can start a company much, much, much easier. So that's just like from observing the industry and talking to

my customers. But at the same time, I was building a product and I was building an AI product. And just a few months ago, I fully transformed from human team to AI team. And just experiencing that, it got me to realize where the actual future of software development is, where the future of

a startup is what's what becomes easier. What does not become easier. So basically anybody can start a company now, but, you still need to know what you're doing. So you still need to know what do you want to build? What problem you want to solve, who you're solving it for. And then, ⁓ at what point do you bring engineering expertise?

At what point, Claude Code can't just do everything for you and you actually need somebody human behind the scenes. So as I was transforming my own development to AI first I also saw startups around me. Basically, starting company today is very, very different from what it used to be even a year ago. And even though I started building Wisary,

for professional product managers and larger organizations that are writing the requirement documents, they have their processes. Now I'm realizing that you don't need to be a product manager to be a builder anymore. You have a bunch of product builders out there and all they need is clear thinking and clear understanding of what they actually want to build. So that's where Wisary is heading is to become a tool for any builder.

And the funny story, just two days ago, my eight year old son, he decides he wants to build an app for his iPad to talk to his friends. And I'm like, well, let's decide what this app is going to do. So basically he was able with the help of Wisary to articulate what he wants to build. And the moment we iterated and he really

clarified what this app should be doing and the memes that should be sending to his friend and all the fun voices going to make. I could just take that, throw it into Claude Code and have it built. So if eight year old can define and build a prototype, an MVP, but build something real, now anybody can. And the only thing that some people

still need a little bit of help with is actually clarifying like what is this thing that they actually want to build.

Monisha Saldanha (04:40) Yeah, fantastic. can you clarify how Wisary helps? Because a lot of people are using Cloud Code directly to create products. But you mentioned with what you were doing with your son, he first used Wisary and then you took that output and you put it into Cloud Code. So can you talk a little bit about that?

Ala Stolpnik (04:58) Yeah. So when we all go to kind of all the AI tools like, ChaiGPT, Claude, Gemini, we need to know what we don't know. We need to know what questions to ask and how to ask them. My son was a perfect example for that. He has never defined the product. He has never built a product. So he didn't know what he didn't know. But the way Wisary works, it uses AI to ask the questions. It doesn't try to assume anything. It tries to

get that information from the person in front of it. So basically AI is not only used to write stuff and definitely not to blindly the accelerated stuff. It's used to ask the questions that force us to stop and think. For example, in that kind of a walkie talkie for Roblox app, my son had to think about what is his friend going to see when he calls him or

what he's going to see what his friends sends him a message. So these questions, Wisary is asking with the help of AI to help the person in front of it actually pause and think and make sure whenever AI Claude or Lovable or whatever, whenever the AI is going to build something that that something is going to be what you actually wanted to build and not what it's kind of just decided on that day.

Monisha Saldanha (06:11) So do you find that Wisary helps eliminate hallucinations from AI?

Ala Stolpnik (06:16) ⁓ Absolutely. What is hallucinations? Hallucinations, it's context that we didn't provide and AI filled for us. So what Wisary is doing is it's making sure that we provide all the context that we need. And if we didn't say something upfront, it's going to ask us about it. And the moment I provide the context, then there's obviously much less hallucinations because there's like much less space for AI to go and kind of go wild and do whatever.

Monisha Saldanha (06:41) And you mentioned earlier that you made a transition from a human team to an AI team. What was that transition and when did it take place and why?

Ala Stolpnik (06:51) So a few months ago, basically with the last release of Claude Code, the moment I tried it, I realized that

what I was doing before as the CTO of my company, I was supervising and I was giving direction to the engineers to go and build the thing. So I was giving them the requirements. I was giving them the technical direction, and then I was reviewing their work and helping them go in the right direction. So now I'm doing exactly the same thing with AI. So my job didn't change. I'm still doing exactly the same thing.

What did change is the speed of implementation. Also the quality of implementation because we're humans. We can't notice everything. Like obviously, even if I do review the code very, very thoroughly, I'll always miss something. And here having AI double checking me. So basically what I realized is by replacing people with AI, unfortunately, very, sad, but I was able to accelerate my work 20x,

which is pretty crazy for a small startup.

Monisha Saldanha (07:56) So you're no longer working directly with engineers, you're working directly with AI.

Ala Stolpnik (08:01) At the moment I

Something really, really interesting. I didn't only have engineers, also had designers, kind of other, other experts. And now I'm realizing that I need to pull them in for a different period of my product, but in very, very different role. So if you think about what was the role of experts before AI.

We're doing hands on work. Products people were writing requirements, engineers were writing code, finance people were doing financial projections. And then AI came. With AI, we stopped doing hands on work. We started doing more management work. So we give the goals, we set the direction, we review the outputs. So that's how working with AI was up until now.

But something that I'm realizing now is like really, really fascinating for me personally is I think the role of experts becomes training AI to do their job. So if now I'm bringing in a designer, two years ago, I would ask her to go and design whatever mock-ups or whatever screens for my app. A year ago, I would ask her to use AI and

guide AI to do the right thing. Today, I want her to review my product, give me feedback, and I'll be able to feed this feedback into AI skills so that the next time the feedback, today's feedback, she wouldn't need to be pulled in. AI can figure it out. She will be pulled in when the next level of complexity is going to be needed. So what's really interesting is suddenly experts become

the mentors and the coaches and not just not ICs and not the managers.

Monisha Saldanha (09:43) That's fascinating. And where do you see this heading? If you had to predict the future with AI, what do you think is going to happen a year from now?

Ala Stolpnik (09:51) Yeah, think the answer is the AI itself, models will keep getting better, hopefully also cheaper. So that trend will keep going. But I think what's really interesting in the startup industry and in the software industry is the adoption side. So small startups are adapting AI. The same way I transitioned to AI first, not only on development, also on operations and everything else.

I see a bunch of startups, early stage startups doing that and being extremely, extremely efficient. So they're moving really, really fast. And the bottleneck there is not implementation anymore, is not execution, is the right idea and the distribution. Actually finding the customers. That's one side of the spectrum. It is like super fun, evolving super fast.

The other side is this all huge SaaS companies. They're not adopting AI efficiently. And even if you have some individuals who are able to optimize their work, that's not significant for these huge companies. Even if you think about software development, a engineer in a large corporate maybe spends 10 % of their time on actual coding. So even if you accelerate the coding by

20X, 50X, this is not significant enough. If they spend all their day in meetings and syncs and collaboration, and you didn't accelerate that, that is not helping. And these are the parts that are hardest to solve and hardest to accelerate. So I think large companies, they will remain slow and eventually they may either find new business models or ⁓ fade out.

⁓ What's really interesting is what's going to happen to this early startups today as they grow. And I think the really, really fascinating thing is you don't need that many people to grow your business anymore. If one person can do a job of 20, let's say to have a stable team, you probably have two people in each role that compliment each other.

and use AI to do everything else. So suddenly you don't need to scale the companies to that many people anymore. You just can have much more companies doing much more different stuff now.

Monisha Saldanha (12:04) What an exciting future. You mentioned earlier that there are some things that are easier with AI and some things that are more difficult. Can you elaborate upon that?

Ala Stolpnik (12:14) Yes, I think anything that is repeatable, anything that you can pattern match and you can automate is much, easier. Writing code, writing tests, reviewing code, running tests, documentation, double checking that code matches the requirements. All these things you can do much, much more efficiently.

and much faster with AI. What I think becomes harder is to remember to pause and think. Just because it's so easy to go to AI with everything and just let it do stuff. I think it requires discipline to actually slow down for the tasks that are really critical. And that's exactly the idea behind Wisary is to force us to pause and think. Are you sure you want to build this product?

Do you really need that edge case or can you maybe push it out of scope or maybe forgot something else? So kind of to force us to stop and think and not just take whatever AI generated is for granted. I think that becomes harder. And even for people who are aware of it and like trying to be responsible, it's just too tempting to outsource our thinking to AI. And we need to find a way not to.

Monisha Saldanha (13:25) Let's talk a little bit about building a company in Colorado. How would you characterize the environment for startups in Colorado? What makes it easy? What makes it difficult? And how is it building a startup outside of traditional tech hubs like Silicon Valley?

Ala Stolpnik (13:42) Yeah, think it's super interesting. So I moved here from San Francisco. So kind of a very, very different environment. So you have there, you have everything, have like a variety of companies, you have talent, you have funding, you have much more opportunities. But the thing that surprised me the most when I moved to Colorado is when I asked somebody, what do you do?

They don't start telling me about the cool startup they work for. They tell me that they ski and run and bike. That's the main difference that people here, don't only live for their jobs. They care about the community. They care about the people around them. They care about their free time. And, I found that kind of that's sense of startup.

community, it's obviously much smaller, but it feels much more intimate and much more fun. And three years later, I feel like that I know more people here and I'm much closer to much more people here in the startup community than I was in San Francisco after eight years.

Monisha Saldanha (14:39) How do you find the environment from a support of startups perspective, infrastructure, mentoring, accelerators, all of that? How do you see that in Colorado?

Ala Stolpnik (14:51) I think there's a lot of investment in that, in startups, in women starting ⁓ companies. So I was, for example, part of an Exponential Impact accelerator in Colorado Springs that was amazing. And I think, again, it's all about community. It's all about building the right connections and making the right introductions. So I think there's definitely a lot of

awareness and investment of startup ecosystem. It's just really fun to build a startup here just because there is this support and community and bunch of other people trying to do the same thing.

Monisha Saldanha (15:26) That's wonderful to hear. And what made you decide to leave Silicon Valley and settle in Colorado?

Ala Stolpnik (15:32) So there like two decisions. One was to leave Silicon Valley, which I made long time ago, but I didn't know where to go. And then I visited Colorado and I discovered the mountains. The moment I discovered Boulder, like I realized, okay, now I know where to move. It was a really easy decision at the time, even though I had to quit my job and

move with the family, but yeah, I think it was like the best decision.

Monisha Saldanha (15:55) You're living in Boulder now and you did the Exponential Impact accelerator in Colorado Springs. Did you relocate to Colorado Springs to do the accelerator?

Ala Stolpnik (16:05) The drive is not that bad, kind of once a week, once every two weeks, we had in-person meetings. Part of it was virtual. So it wasn't an issue at all. So yeah, but got to get to know the Colorado Springs community as well, which is, it's really interesting how diverse Colorado is in some ways.

Boulder is different from Denver, is different from Colorado Springs, and each one of these places has its own character and vibe. So even that experience was really, really interesting.

Monisha Saldanha (16:40) And what type of support did you get from the accelerator?

Ala Stolpnik (16:43) The XI specifically, it was all about networking and connections. I made some really good friends. I got exposed to other kinds of businesses. That's another difference between Silicon Valley and I guess the rest of the world. You have different companies here. You have people doing different things. Whether this is

hardware, space, government work, something that I totally forgot that existed when I lived in small bubble in San Francisco.

Monisha Saldanha (17:08) Great. And you mentioned you changed your ideal customer profile from expert product managers to is it everyone now because anyone can build or who would you say is your ideal customer persona now?

Ala Stolpnik (17:24) So it's a, it is still including professional product managers in larger organizations, but now it's also available for any builder. As I said, from eight year old that wants to build his first app to even product people who are starting their startups and want to scope out their MVP and all the way to especially non-technical founders that have

amazing ideas, but they don't know where to start. What I've seen happening before a lot is business people with great ideas would go to engineers, ask them to build MVP. Half a year later, they get something half broken and they don't know what to do about it. And it's not what they even imagined. So, so what I'm helping them with is to

understand very, very well what is this thing they imagine. And the moment they can understand it, they can communicate that to the other side. And the other side could be AI for prototyping, or it could be like actually engineers that are going to build it.

Monisha Saldanha (18:23) And in building Wisary, have you been surprised by anything in terms of how people are using the product?

Ala Stolpnik (18:31) I think building an AI product kind of in this time is super, super interesting because what I was witnessing is including my own experience is how people perceive AI and how that perception changes along the way. So when we just started and Chat GPT just came out, like the first reaction I heard from people like, ⁓ it's writes with a style that is recognizable as AI.

And I, I'm proud of my job. I don't want people to know that I'm using AI. So that was probably two, three years ago. Now, if you're not using AI, people look at you what's wrong with you. So I think, to see how people perception about AI is changing also forced me to change how the product looks and behaves. So we change, we basically.

read it kind of the entire user experience, twice, kind of based on kind of how people expect it to interact with AI. Starting from one a one attempt prompt and getting something to much more interactive process. And now we're talking about MCPs and we're talking about connection between different tools.

Seeing how people expect to interact with AI and building the product that supports that was super, super interesting.

Monisha Saldanha (19:48) Can you talk to us a bit about your tech stack? So what are the tools that you're using to build your product and how are they interacting with each other?

Ala Stolpnik (19:57) The infrastructure of the tech stack is basically the same as it used to be before. Standard full stack, a web app. The tooling is a, Wisary on the product side to help me define the requirements for the next feature. And then Claude Code on the development side. So that's anything that engineer is doing basically is happening in Claude from

writing the technical plans to implementation, to review, to debugging, testing, whether it's unit tests or regression tests. So all that is happening in Claude. And then the output of that is basically test plan, the documents, what has been built. And that's kind of a great connection between the engineers and documentation of what they built and the product requirements that we started with. And now you can actually see

where they disagree.

Monisha Saldanha (20:46) Is there anything that you've learned along this journey that you would love to tell your former self who was just starting off building Wisary? Is there anything that you know now that you wish you'd known when you started?

Ala Stolpnik (21:00) Yeah, I when I just started, it was like the early days of AI and I think like similar to many others, like I was panicking. Like things are changing like every day. Like there's a new model, there's new capability. And initially I was trying to stay on top of everything and be the cutting edge of everything until I was able to build kind of a mental model of, okay, like this is where the models are. Yes, they're getting better.

I don't have to always use the most recent one. I can stick to certain technology for a while and then switch when I need to. So I think just, yeah, stop panicking if you're feeling that like I'm behind all the time. Because it's also kind of when I talk to a bunch of people out there, everybody feels that they are behind with AI and it doesn't matter how far along they are. Like everybody feels they're behind.

And I think it just doesn't make sense to chase every new change. The moment you have the mental model of, what's possible today? What will likely be possible tomorrow? It's much easier to do what's right at this moment.

Monisha Saldanha (22:06) And did you make any pivots beyond, you know, widening the scope of your ICP? Are there any other pivots that you made in building Wisary?

Ala Stolpnik (22:06) Thank you.

Yes, it's funny because when I just started, I was actually building a tool for junior managers like myself. And initially I was thinking like, okay, like I was engineering manager, what was hard for me? Was a task breakdown and technical plans. I actually started to build a tool for that until I realized that nothing matters if the requirements are not good. And kind of that's, that was the first premise to realize like who I'm solving the problem for

and who, will be the user of the tool. And then also as our understanding of AI evolved, the tool itself evolved. So initially we all discovered that ChatGPT can write stuff. So initially the product was writing stuff. And then we realized that it hallucinates and if it doesn't have the right context, it just creates garbage and it's not really helping. And then it's like, okay, let's, let's bring rag. Let's bring the context.

And then I realized that in software development world, we never document our knowledge and all the knowledge is just in somebody's head. So no matter how well the technology is, if it's not available in digital format, AI is not going to do anything with it. So that's where it pivoted to the the direction of using AI to ask the questions and actually get this information from our heads.

Since then, that's the core thing and core value of Wisary is asking these questions and pulling the information out of human heads so that AI can do something meaningful with it. And then the most recent pivot was ⁓ our first interface was as a Confluence app. We're integrating as add-on into Confluence because

initially as I said, I was building solutions for professional product managers in larger organizations who are often using Confluence. But as I started to see all these new builders coming along that don't want to hear anything about Atlassian or Confluence or old antique SaaS tools, but they do need something.

So we pivoted to a standalone platform. now anybody can just sign up with their email and start using immediately. So yeah, so there's like a interesting evolution and it's, only has been like three years, but so much has happened in like the AI world, whether this is technology or the landscape of how companies operate and who is doing different stuff.

So it's a fun journey.

Monisha Saldanha (24:33) Yeah, really exciting

journey. Are there any mistakes that you made that you'd like to share?

Ala Stolpnik (24:39) So I think one of the mistakes that probably many early stage founders are making is trying to raise money when they're not ready for it or when the market is not right for it. I think I wasted a lot of energy on trying to raise money until I realized that the industry is not there. The requirements, especially now in kind of

the age of air that everybody can build, the requirements for traction are much, higher. And on the other hand, we don't need that much cash anymore. So when I realized that I can keep building my company very lean, very fast without external capital, it's freed up so much of my time and so much of my energy to focus on actual thing of building the company.

as opposed to trying to get funding. So I think if looking back, I wish I would have realized that earlier and I didn't spend that much energy on trying to raise money when it wasn't right time.

Monisha Saldanha (25:36) That's a really great lesson. Don't try to raise money until you're actually ready for it and need it. You may not need it anymore in this new world. Thinking ahead to the future, what is your vision for Wisary? What impact will it have on people's daily lives?

Ala Stolpnik (25:50) I think anybody who wants to build a product will be able to build a product. And ⁓ it starts with just identifying the idea ⁓ and the prototype and even like deciding that I want to test out this idea. So, my hope is every product builder out there, Wisary will be there to do tool for that phase. Now let's say they

vibe coded 10 different ideas and they decided this is the one that is going to work. Now they need to go and translate that into actual products. They will need real engineers. They will need real designers. They will need real experts. So here, Wisary will help them define what is the gap between what they have today and what they actually need to ship it to real users and real customers. Once they did that, and hopefully it succeeded,

now they either bring in a professional product manager or by this time they became a professional product manager. And here they start to iterate. They start to more functionality and more capabilities to their existing product. And here again, Wisary's with them to help them evolve their product, moving forward. So, the idea is to allow anybody to start a company and to grow with them as they mature their company.

Monisha Saldanha (27:04) Is it your intention to stay in Colorado and keep building Wisary here?

Ala Stolpnik (27:05) Amazing.

Yeah, absolutely. Snowboarding on the weekends or climbing on the weekends depends on the weather and yeah, and building Wisary the rest of the time.

Monisha Saldanha (27:11) Nothing.

Well, fantastic. And last question for you. What is one book you would recommend every builder read and why?

Ala Stolpnik (27:26) The Mums test, that was one of the first books I got recommendation for when I was just starting about how to ask the questions with minimum bias and how to try to get the truth from the answers. In a way, that's also what Wisary is trying to do is ask the questions and try to get to the root of

the problem or the solution. But yeah, I think the mom's test, was something that forced me to rethink how I interview the customers, how I validate my idea and decide what's going to work and what's not going to

Monisha Saldanha (28:03) And how often are you talking to customers?

Ala Stolpnik (28:06) Um, all the time. I think what's interesting is the scope of who the customers or potential customers are keeps growing. So the more and more people are building a random person I run into and just randomly chat, chat with maybe one day a builder and a customer. So for some customers, I talk intentionally with some customers

they become customers as their journey progresses.

Monisha Saldanha (28:33) Yeah, so the mom's test

is really helpful for you then. Great. Well, Ala, thank you so much for your time. This has been such a wonderful conversation. I've learned a lot. And I think our listeners have probably learned a lot about the changes that AI is bringing in terms of building and launching startups, particularly in Colorado.

Ala Stolpnik (28:53) Thank you so much.

Monisha Saldanha (28:54) And I'd like to thank our listeners for listening to this episode of Colorado Tech People. If you enjoyed our conversation, consider sharing this episode with someone who loves building products. Be sure to subscribe so you don't miss future conversations with founders and leaders shaping Colorado's tech ecosystem. Until next time, keep building, keep connecting, and keep creating experiences that bring people together.