How To Build A Software Business With AI This Weekend Zero Coding Skills Required
In this episode Omar Zenhom explains how AI has eliminated the traditional barriers to building a software business, allowing anyone with only writing skills to launch a SaaS product in a single weekend. He starts by framing the market shift: the old model required years of coding or hiring expensive developers, but today AI tools can replace both. He then walks through a step‑by‑step framework: (1) selecting a real‑pain problem using three diagnostic questions and a three‑tier pain matrix; (2) defining a razor‑thin MVP that solves one core need; (3) creating a Product Requirements Document (PRD) with the help of Claude, an AI assistant that acts as a technical co‑founder; (4) using a minimal toolchain—Supabase for the database, GitHub for code storage, Windsurf (an AI‑powered code editor) for generating code, Vercel for hosting, a domain registrar, Stripe for payments, and KITT for email marketing; (5) iterating via a prompt‑driven workflow where Claude crafts precise prompts, Windsurf executes them, and bugs are fixed by looping back to Claude; (6) a detailed Saturday‑Sunday schedule that maps each hour to problem definition, PRD creation, UI building, database wiring, local testing, deployment, and early user feedback; (7) a cost breakdown showing that a functional MVP can be built for roughly $40‑$80 per month. Throughout, Omar emphasizes mindset tips—focus on one task at a time, keep scope tight, use screenshots to communicate bugs, and commit to GitHub after each successful change. The episode ends with a call to action to share builds on social media and a teaser for deeper business‑growth content.
Chapters
A few months ago, I published an episode called Can You Build a Profitable SaaS Business? A Software Business in 7 Days Using Just AI. The response was extraordinary and the messages started to come in. People were asking me, Omar, how did you actually do this? What were the tools? What are the step-by-step moves I got to make to do the same thing? I want to build something, but I have no idea where to start. This episode is my answer. This is a practical step-by-step, nothing left out guide to building a software business with AI. Not in 3 months. This weekend and with zero coding skills. Okay? You don't need how to code. Just need to know how to write any language like English. I'm gonna show you the exact tools, the exact process, the exact thinking that turns an idea into a working piece of software that you can sell. I created a software in 48 hours just to prove this point again, and I'll show you how I did it. Now, before you tell yourself, oh, this is not for me. I'm not a technical person. I don't know how to code. I don't have a team. None of that matters. This is why this is so exciting. I'm going to show you and prove to you that you can do this easily. So this is a little bit of a longer episode because we want to make sure that we cover everything, the A to Z, to make sure that you can do this this weekend. So grab yourself a drink and let's get cracking. I want to make this super clear. We're in an opportunity right now that is unprecedented. For most of human history, building a software requires one of two things. Either you need to spend years learning how to code and learning infrastructure and learning how to put it all together and how to deploy and all that kind of stuff, or you gotta spend a lot of money to pay somebody who knows how to do this. Now, both of those options are no longer valid. You don't need any of that anymore. It's all over thanks to AI. Let me paint the picture. So if you've been following the show for some time, you know that in 2014 I started a software business called Webinar Ninja. Right. And in that time I had to hire a bunch of people to make this software a reality. I had to hire a designer, right? I had to design the look and the feel of the actual software. I have to add a front-end developer. Okay. That is something that you don't need to worry about anymore because it's irrelevant. I had to do a backend developer. Okay. I had to hire somebody who's a backend developer. I had to hire somebody to do infrastructure. Right? I had to hire somebody to also project manage the whole thing. Okay, project management. So I had to make 5 hires just to get this thing off the ground. So we're talking about literally something like $300,000 to $500,000 in annual spend on salaries, right? So we're looking at $300,000 to— again, $300,000 to 500,000. Now it's 2026, we have AI. AI is our best friend now, okay? We don't need any of those people right here anymore. You can do this all yourself, and you don't need to have any of these skills. And what I'm going to show you today with all the tools, you're going to spend less than $200 a month to get this all done, and it's faster. And back in 2014, it took me months to have a viable product that could sell to customers. Uh, I'm talking about 3 to 6 months. In 2026, this weekend, you can have a viable product that you can sell to customers. I'm not joking, and I'm gonna show you right now. Let's start with step 1. Step 1 is we need to decide what problem we are solving. Everything starts here. Before you open a single tool, before you write a single prompt, before you spend a single dollar, okay? You need to know the problem that you are solving. And I wanna be absolutely direct about something right now. The technology is the easy part. So many people are so bogged down by the technology, and I'm telling you, this is the simplest part of the equation. I've built many pieces of software the old-fashioned way as well as with AI recently. And I'm telling you, this part is the most challenging because the software, the actual code, the database, the deployment, this is something that AI can largely do for you. Okay? Uh, pretty much all of it's done by AI and AI cannot identify a real problem that real people have and are willing to pay to solve. That requires you. So I want to give you a framework so that you can nail this and make sure you're building the right product that doesn't give you headaches later on. That's easy to sell. Let's get into it. So there are 3 categories of products. Okay. There are nice to haves. Nice to have. Okay. And we have another category called, okay, a real pain. And then we have another category called desperate need. Okay? Now, the reason why I have these 3 categories is because this is really where people are willing to pay. So if we— if this line here is willingness to pay, willingness to pay goes from left to right, then we want to make sure that we're in this zone as much as well, as right as possible, as closest to desperate need. Why? Because if you're offering a solution to a problem that's kind of nice to have, it's not a must-have, it's a nice to have, then people are not so willing to pay. This is hard to sell, hard to sell. Okay. We don't want this. Okay. This is going to give us headaches. This is going to make us build products that people don't really want. What we want to do here is we want to play at least in this zone, the real pain where it's a sellable product. People, don't need too much convincing, okay? If they have a problem and they recognize that problem and it's a real pain, then they're going to go for it, okay? They're going to pay for it, okay? Not so hard to sell, right? I would say easier to sell. Now, if you have a product that people are desperate for, that they're like, "Wow, this is something I've been looking for all my life," This is going to be extremely easy to sell. Right? Super easy. So you want to make sure you sell in this area and this area, right? The real pain area and the desperate need area. Okay? If they are desperate, this is going to make your life so much easier. Now, obviously, if they're desperate, they're willing to pay more. So you can have higher prices with these products. Okay? Real pain. They're going to be somewhere in the medium-priced product, right? Nice to have. This is not even worth looking at. I would say this is a red zone. This is not a place we want to play because it's going to be very hard to please these customers because it's just a nice to have. So how do you choose your problem? Ask yourself 3 questions. These 3 questions are going to be a lifesaver. Here's question number 1. What is something I do repeatedly in my job, in my business, in my life that is tedious, that is manual, that is inefficient. If it's tedious for you, it's probably tedious for thousands of other people like you. Question number 2, what do I or people I know pay for that is either too expensive, too complicated, or not quite right for their specific needs to solve this problem? This means are people already spending money to try to solve this problem? And this is key. This means you don't need to be innovative. You don't need to try to solve a problem or try to do something that people are not already attempting to do. For example, people wanted to rent movies. Blockbuster was around forever, and then Netflix stole their lunch because they just delivered that solution via streaming, which is a different way to solve that problem, a more efficient way. You don't need to invent a category. You need to serve a category in a better way. Question number 3: What would make my professional or personal life significantly better if this existed? The best software businesses are built by people solving their own problems. I built my software business Webinar Ninja because I was running webinars. I hated the software options out there. I had to Frankenstein 10 tools to make it all happen. So I built something for myself and that really is me scratching my own itch because if you have a problem and you understand it very well, deeply, you know a good solution when you see it, you know what is needed to convince other people like you that this is actually a better solution. So I want you to do this exercise. Ask these 3 questions and write down 3 to 5 problems that you think would potentially be a good product idea. Then from there, choose the one you think is going to be the easiest to sell. Right? Not the one you're most passionate about. Not the one that you think is going to like change your reputation. No. Right now we're trying to sell a product. Which one would be the easiest to sell? The most convincing to sell as a solution to a problem. Choose that idea. And let's move forward. Next step, step 2, is to design your minimal viable solution. Now that you have your problem, you need to define the smallest possible version of the solution. This is called your minimal viable product, MVP. Let me show you what this looks like. If I'm creating an MVP, I'm really just solving one problem. This is just one problem that it solves. Okay. I'm gonna give you an example in a moment. This is where we begin because later on you're going to build out the software. You're going to have other nice-to-have features, right? Nice-to-have features, N-H-F, right? Then you're going to then have other features that are like extras, add-ons, all that kind of stuff, right? But right now, this weekend, we're going to just build a product that does one thing brilliantly. So for example, I created this app called Spoken, spokenhq.com. I literally created this thing in my spare time. So what does Spoken do? Well, Spoken allows you to embed in your newsletters an audio version of the newsletter. So if people don't have time to read the newsletter, they can listen to it while they're washing the dishes or when they're, you know, getting the kids ready or whatever it might be, or in their commute. So the point of this MVP is I just need this app to read out the newsletter. So basically the user would just copy and paste the newsletter into Spoken and it would create a narrated version with a default voice that allows them to read out this newsletter. That's the MVP, minimal viable. Now after I nailed the minimal viable version, I then was able to add some nice to haves. So one of the nice to haves is that it can clone your voice. So you have an AI version of your voice. So now when the newsletters are read, it sounds kind of like you, it's Pretty cool. And then beyond the nice to haves, we have the extras, right? So extras and add-ons. Well, then I added a KITT integration so that it integrates with your email marketing provider like KITT and it auto-populates the emails that you have in queue. So then you don't have to copy and paste. So as you can see, what's the MVP? Well, I just needed to read out the newsletters with AnyVoice just to prove the concept that people actually would prefer to listen to newsletters or allow other users or allow their readers to listen to newsletters rather than read it. In the tech world, this is called scope. We want to narrow our scope as much as possible this weekend because we only want to build for a very tight scope. One problem, one user, one clear outcome. And you want to cut everything else. A lot of you are going to struggle with this because you want to build more and do more features and make it better. But I'm telling you, you have to exercise restraint. Why? Because right now we're going to just test out if our product is indeed a solution to a problem. The core solution that we're providing works, right, is actually helping our customers. We'll have time later, lots of time to be able to add more features and make it better. So let's get into the tools. What are the tools you're going to need? There are only 5 tools you'll need to build out your app. The first tool is called Supabase. I know, cool name, right? Supabase. And basically this is your database. It stores all your data. That's all you need to know. The second building block here is GitHub. This is a tool called GitHub, right? What is GitHub? It stores your code. Your code is what actually makes your app happen. And you need to store this somewhere online. So it's accessible to everybody in the world when they're using your product. Now, who's going to write the code? Well, the next tool is called WinSurf. Okay, WindSurf. All right. And this is basically a code editor that uses AI to code for you. There's a lot of these out there and there's a lot of different tools, but this is the one I like. This is the one I use. So I'm just gonna give you what I use. It basically, it's your AI developer. This WindSurf tool replaces frontend developers, backend developers, all kinds of developers. The next tool you need is Vercel. Vercel is how you push your software and your code out in the internet, and it allows it to host that actual tool so that people can then access it anywhere when they go to your, uh, URL, you go to your website. By the way, you don't need to understand any of this right now, and you actually don't need to understand it ever. You just need to know that you need these tools, and I'll show you how to sign up for them. And then lastly, you're going to just have your domain, right? This is not really a tool, but basically it's just your, your website URL. Like for example, spokenhq.com. And this is where it's hosted. This is where the address of where people should go. And this domain is connected to all these tools so that you understand where do the user go? Where does the user go to sign into the app and see the website and sign up and all that kind of stuff. Pretty simple. Okay, now there's a special friend that we haven't talked about that is going to help us do everything. Okay? Basically it's the foundation to making sure that you are able to build as fast and efficient as possible, bug-free, or at least minimal bugs, I should say. And that is our friend Claude. Okay? So if you haven't heard of Claude, Claude is basically a better version of ChatGPT. It does a lot more. Uh, it understands how to code better. But remember, we're using WinSurf2Code. Claude is basically our project manager. They're the thinking partner that's gonna help you in each layer. Anytime you get stuck, you're gonna use Claude. And Claude, it's worth its weight in gold. So let's talk about price. So Claude, I would definitely pay for the pro version, which is $20 a month. This is gonna give you, uh, better models. It's also gonna give you more limits so that way you can just continue to work as fast as possible. Okay. Uh, all these tools, Supabase, you have a free account that you can sign up for is absolutely free. Uh, GitHub is a free, uh, code repository. It's absolutely free. Windsurf, not free. We're gonna talk about Windsurf in a second. Vercel has a free hobby account that you can start with. You can get your product up and going for free and your domain also not free. So let's put the ones that are free, which are Supabase, GitHub, Vercel. Okay. Your domain, you can get a domain as cheap as $9 a year, or if you have a fancy domain, that's really hard and a .com or hard to get, I should say, then you're going to pay thousands, maybe even hundreds of thousands. Some people even spend millions, but you don't need to do that. You could just go with the most basic domain possible just to get your app going. My domain spokenhq.com, was a steal. I got it for $17 a year, so you could do the same. So let's just call this $20 a year. Okay. Now, Windsurf is a tool that I highly recommend you use, and it's only $100 a month for the pro version. Now these prices change, but I would go with the pro version because this is actually the thing that's doing all the work, that's doing all the coding, and it's worth paying for. So $100 a month for that. So if we total it all up, you talk about $120 a month for the, for the tools and $20 a year for the domain. Pretty, pretty cheap. Let's look over what these products look like. Supabase. This is the database. This is my database for, uh, Spoken, the project I created. And it's very, looks complicated, blah, blah, blah. Do not worry because Claude is gonna tell you step by step where to go, where to copy and paste. It's super simple. What you need to know is that you just need a database, sign up for Supabase account, it's absolutely free. The next level is GitHub. GitHub is actually something I would sign up first, because you can actually log into all these products using GitHub. And GitHub is absolutely free. You create an account. And it's pretty simple, okay, and this is where your code is going to live. And don't worry, you don't need to go to GitHub a million times, you can then integrate GitHub or sign into GitHub automatically through WinSurf. And that way your code is directly updated to GitHub automatically. Tech is amazing. Alright, the next level is WindSurf. Okay, the WindSurf is an actual app you download to your computer. Okay. Now, this is what it looks like. Okay, this is WindSurf. This is the project I'm working on right now. This is actually creating code for me with me just giving prompts just to give you a quick preview. But I have a pro account with WindSurf, as you can see, and it's pretty damn good. I have some extra usage balance. Sometimes I go over, but you're not going to need that because you're going to be creating a very simple product. And as you can see, like I said, it's connected to GitHub. So automatically when it creates code, it puts it up into GitHub. Vercel, Vercel is this tool, which allows me to deploy the code and allows you to host my app easily, absolutely free, sign up for a hobby account. It's free. And domain, I would recommend going with GoDaddy. Godaddy.com has been around forever. Uh, you can search for a domain and then go ahead and purchase it. What do you do with that domain? Do not worry, Claude's going to tell you exactly what to do step by step, nice and easy. And a quick correction, because I just looked it up, Windsurf actually lowered its prices. The Pro account is now $20 a month. Now, for $40 a month between Claude and Windsurf, you can do this right now. It's so, so simple. So just want to make sure I give you the up-to-date price as of today. So the total cost for an MVP is $40 between Claude Pro and Windsurf. Very, very inexpensive. When you're ready for production and you have real users, you're gonna pay about approximately $80 a month because you're gonna have to upgrade your database or maybe Vercel depending on how much traffic you're getting. So let's build this software. How do we do it? Well, it starts with something called a PRD, a product requirements document. Sounds very technical. Sounds very geeky, right? A PRD. Don't worry, you don't need to know how to write this thing. Why? Because our friend Claude is going to do it for us. Basically, this is a blueprint for your MVP. And the reason why we need a PRD is because when we prompt WinSurf, who writes the code, we want to prompt them efficiently. We want to make sure that we don't just go blah and it doesn't really understand our, you know, non-technical speak. So we're going to ask Claude to help us out. Now we're going to get into Claude in a second, but first I want to make sure this is crystal clear. Step 1 in building any software is creating a blueprint. All right. This is what we need to do. We need to first have a plan. Okay. Step 2 is building a foundation. So we understand that we have to kind of create the software with strong foundation. So every time we build upon that software, It's going to be solid. And then last step, step 3, is to actually build the actual house, all right? Or build the software. So it's just like building a house. You have a blueprint, you have foundation, you start building. Now the PRD is the blueprint, right? The product requirements document. If I don't create this, right, it's gonna make my life so much harder. This is where we start. If you skip this step right here, if you skip this step, You are asking for trouble because it's gonna require you to build everything twice. You're gonna, you know, put it up into the code editor, Windsurf, and it's gonna give you something that you don't like. So let's create this PRD. How do we do this? Well, it's very simple actually, and I'm gonna give you a script. The first thing we want to do is we want to go into Claude and we want to make sure that we create a project. As you can see, I have a lot of projects here. I have projects on, Creating scripts for the podcast. I want to refine my scripts. I have investments. I have travel guides. I have business coaching project. I have a poker coach. This is actual a poker coach. Claude is my poker coach. Spoken is the project I created, right? All these are different projects. So you're going to create a new project. When you create a new project, you're basically creating a place where you're going to work and it will remember all the stuff that you need to know. So we're going to create a new project. Okay, we're going to click New Project. And we're going to name this project. So we're going to call this, you know, Project X, okay, whatever your project is called, you're going to call it whatever your app is called, right? And then you're going to say what you want this project to do for you. You can write something like as simple as you're gonna be my technical co-founder, I'm creating an app using windsurf. And I want you to be able to create clean prompts so that I have a bug-free environment. And I can create this product quickly. We're gonna start with a PRD and then move on to the prompts. So you can just write that, for example, and then create project. Okay, so we're gonna create the project. Now we have a project, we can now do the first prompt. Now what do we write in this prompt, this first prompt? Now it's very simple. We can just simply use this prompt that I'm gonna put for you here. If you're watching on YouTube, we'll put it in the description. And basically this prompt is just telling, uh, Claude, I wanna build an MVP for a software product this weekend. You're just telling it exactly what you want, right? It's like a genie. Here is the problem I'm solving, and you can insert the problem that you're solving, 2 to 3 sentences. Here's the target user, right? You know who the target user is. Here's a core feature of the MVP. What's the one feature that matters? Okay. And then please help me create this product requirements document, the PRD. Okay. I need clear descriptions of what the app does. I need the core user flow. This is basically you telling a developer, This is what I want this to look like. This is what I wanted to do, right? The data model. Here's the tech stack. I have a preferred tech stack and includes the products I talked about. So you could just rip off mine, that's fine. And then the specific screens or pages of the MVP that needs to be there, like a login screen, uh, you know, all that kind of stuff. It's going to understand what to do. You don't need to even actually tell it what screens to do. And number 6, what is explicitly out of scope of the MVP? This is like out of bounds. We don't want to go beyond this We don't need all that stuff. Okay? You want to keep the scope tight, right? And this is all you have to say and say, in one weekend, I'm a non-technical founder using Windsurf AI so that it knows how to prompt for Windsurf. And that's it. Pretty, pretty simple. So let me give you an example what this produces so you see for yourself. So I'm, I'm literally going to just dictate using an app called Whisperflow. So what problem am I solving? I don't love all the project management products out there. Uh, I want to create a project management software that is specifically for podcasts or podcast producers, uh, that makes it, uh, easy to produce a podcast step by step. Okay. Just fill that in for me. Uh, my target user, small businesses that run podcasts, solo founders, or even small podcast teams. The core feature, The core feature is a detailed Kanban board, uh, like Trello that allows people to work through the process of, uh, creating a podcast, uh, making sure that, uh, it's easy to follow and people are notified when each step, uh, is moved around along with the cards. Okay. I'm making things up here. I don't even know if this is something that people want. Uh, I haven't validated yet, but right now I just want to give you an example. Let me hit, uh, send. And as you can see, Claude has generated a detailed PRD, which is great. If you look at here, it's actually saying, hey, I made some choices for you. You could push back on them and I can update the PRD for you. Uh, and it just gave you a real MVP. And then on the right here, you can literally copy and paste this. This is all you have to do. It even named it, right? Named the product Podflow. Right. And it has all the information that you require in a PRD, like what the app does, the user flow, the data model, uh, all the technical stuff that you need to have. It understands what you need to have and it produces it for you. So you can literally just hit copy. Okay. And then go to Windsurf. And then in my Windsurf, like this is one of my projects that I already have going, but in my Windsurf, I can just hit paste, and just send it to Windsurf, and Windsurf will start building. It's crazy. So you could literally just go right to Windsurf here. I created a new, uh, you know, this is my old project here, so I'm gonna create a new project here, call it Project X. Oh, actually, I'll call it Podflow. It's called Podflow now. Okay, and I'm gonna select the folder on my computer. I'm just gonna go to Documents New folder, call it Podflow, create, because it's going to create the actual app on your computer first, and then you're going to deploy it on the internet using Vercel and all that kind of stuff. So here we go, Podflow, and that's it. So now I have my WinCE project. Here we go, and I can literally just paste my PRD right here.. And I'm going to just go ahead and then hit send. And when I send it away, it's going to go and do it. Now I have the field full PRD. It shows you step by step what it's doing, right? It says, hey, I got what you sent me. And here are all the steps. And this is what I'm doing right now. Step 1, step 2, step 3. It's literally building it for you. Super simple. So when Surf is building, you watch what it does. You just accept when it asks you to accept. It's very, very easy. Now, things may break. You might get bugs. That's okay. Guess what happens? You go back to Claude and you say, hey Claude, I got this bug. It's telling me there's an issue. You screenshot it, you send it to Claude. Claude says, sure, give Windsurf this prompt. You give it the prompt, it fixes the bug. It's that simple. And by the way, this is completely normal. Anytime you build any kind of software with AI or without AI, You're going to have bugs and bugs basically just mean things just didn't work out. There's something broken in the code. The next step is we're going to connect our database to Supabase. Guess what? This is super simple and you don't really need to understand why you need this. It basically just stores any data that the user needs to use your product. I'm not going to tell you how to do this because there are many steps and you're going to forget and you don't need to know from me. All you need to do is go back to your project manager. Go to Claude. Hey Claude, I need to connect my app to Supabase. How do I do this? And it will give you a prompt that you can put into Windsurf. It'll give you the steps to do inside of the Supabase website. You're gonna use Supabase a lot because every time you create a feature, you're gonna need a data table so that the data is stored somewhere. Now there's a lot of technical mumbo jumbo. How do you create a data table? Do not worry about it. Why? Cuz Windsurf will create the data table for you. It'll create the code and say, hey, here's a data table, copy it. Go into Supabase, go to the SQL editor, and it will show you step by step how to get to that. And you're going to paste it and you're going to create that data table. I'm telling you, you just follow the instructions that AI gives you and you're going to have a lot of fun because you're going to be learning a lot along the way. You're going to be actually getting your hands dirty and realizing, hey, I'm, I'm a developer now. This is crazy. Now there's two things that I want to wrap your head around. Okay. Do not get overwhelmed. This is very simple. When you're building out your app, you first build it on your computer. Okay? It's not out on the internet yet. You're building on your computer and then you're going to deploy it to the world. Once it's ready, once you've tested it, once you know it works, then you can put it on a URL on a domain and people can access it. So how do you actually see what your app looks like as Windsor is building it? Well, you're going to actually, when we remember what you saw, I was creating a folder in my computer to create the app. Well, there's a way to access all that code and then show you what it looks like. And it's called localhost. So if I go to my screen right now, you'll see that this is my app, but I'm not on spokenhq.com, on localhost:3000. This is where Windsurf decided to build out this app. And I'm basically seeing this app on my actual local machine. And this is where I can look at things and say, oh, I don't like this. I want to change that. And I can screenshot it. And then create a prompt. More about that in a second. Okay, so this is just— uses your local machine and it's just like any other website. So if I go to dashboard, for example, it's gonna, you know, uh, prompt me to log into the app. And that's the login screen. And then you'll have a login. That's why we needed the database, right? So the database is storing in Supabase. So I just wanted to show you that this is kind of where you see the code being created. Now, if you wanna change something that you see, you don't go to Windsurf. You always go back to your buddy Claude. For example, if I wanted to center welcome back, sign in, uh, I can just go ahead to Zyte here and I can— this is my screen grabbing tool— and I can just grab this here and I can go ahead and then guess what? Go to Claude, go to my project, go here, and then paste my screenshot and explain what the problem is. Say, hey, I don't like the fact that this thing is justified left. I want it justified centered. I want to— this is the login page. Can you create a prompt so that, uh, Windsurf makes this change? And then it's going to produce a prompt for you, and then you go into Windsurf and you paste it. By always going to Claude first, you're going to minimize problems, bugs, issues, or Windsurf just not understanding exactly what you want. Claude is really gonna make it super simple. Now, if you're gonna attach a screenshot in Claude, make sure you attach it as well as a reference in Windsurf. So that way you have a much cleaner way to develop your product. Every time you wanna make an edit, do this process. Screenshot, go into Claude, ask it to prompt the fix, and then create the prompt. Pop that prompt into Windsurf. Before we move on, I just wanna say that you have understood 80% of all development. Okay? This is all it is. And most of your time is going to be spent doing this, going to Claude, asking it to create a prompt, going to WindSurf, pasting that prompt, it producing it. You go see on your local what it looks like. Okay, I like it. I don't like it. Move on. So here are 5 tips because you're going to spend a lot of time in that whole process to make your process a little easier. Number 1, one task at a time. Never give WindSurf 10 instructions in one prompt. So this is why I like to go to Claude first. To make it sure that it understands to keep it simple for Windsurf, right? You want to do one task, one change at a time. You want to wait, you want to test and then move on. Why? Because you want to build upon broken code or build upon broken features. Tip number 2, you're going to have some things that don't work when you test it after Windsurf has created it. These are bugs. Like I mentioned, it's not broken. It's just not working as expected. There's something wrong. So every time you go into Claude to create a prompt to fix a bug, be as descriptive as possible. Why doesn't it work? One of the features I like to do is like to right-click on anything and do inspect element. Let me show what that looks like. Let's say for example, something's not working here, like this button's not working. I could just right-click on this, uh, screen, on this browser screen, and do inspect. Okay, when I do inspect, it's going to just kind of move things over here. You're going to see all this code. Do not worry, just go to console up here and you're going to see some errors. That you can just literally copy and paste or just screenshot these errors. And from there, put it into Claude so it understands what's happening on the backend. Tip 3, when you put the prompt into Windsurf that Claude gave you and it maybe have a question or maybe it doesn't understand it, don't try to answer it right into Windsurf. Go back to Claude, tell Claude what Windsurf said, and then you go ahead and it'll tell you how to clarify or how to help you to get through that hurdle. Or Windsurf gives you some instructions like how to set up Vercel and you don't understand these instructions, go to Claude and say, I don't understand these instructions. Can you show me how to do this like I'm a 5-year-old? And it will give you step-by-step instructions. Tip number 4, Windsurf likes to overachieve. Okay? It's gonna sometimes go outside the scope. So what you like to do is that you always tell Claude, hey Claude, every time you gimme a prompt, remind Windsurf not to go outta scope, not to do anything outside the PRD. Until I tell you otherwise, until you finish the minimal viable product, until you finish your product, uh, its first version. And it will say this in the prompt so that when you pop it into Windsurf, it doesn't, you know, kind of go on its own. Tip number 5, you want to commit to GitHub consistently. So as soon as something is working on your local and you've seen it and you like the way it looks and like the way it feels, make sure you just tell Windsurf, hey, deploy this, or, you know, I want you to commit it. To GitHub. And it will integrate with GitHub for you. And then therefore, it will automatically put the code in GitHub so that you have clean code up there that is working properly. So just so you understand the levels, there are 3 levels to developing code or deploying code. Level 1 is Claude is creating the prompts for you, okay, and then giving it to WinSurf, you're going to give it to WinSurf, and it's writing the code. Level 2, this gets created on your local machine, right? You're going to get to test it like I showed you on localhost. Once it's done like that, you're going to pop it into GitHub. You're going to— okay, it works. It goes to GitHub. And then level 3 is the code exists on the internet. And this is where Vercel comes in. And we're only going to get to this point or this stage once we're ready to show this to the world, cuz once it's on the internet, it's out there. People can access it. Google will crawl it and it will index it. And that leads to the next step. How do we deploy with Vercel? I'm gonna show you visually first so you can conceive what's happening here. So what does Vercel do? Well, Vercel is allowing you to do two things. It's combining your code and your database and allowing it to be accessed on a website on the internet. So for example, when people go to spokenhq.com, they can access the app and make it work. Okay. How does it pull it? Well, it pulls your code from GitHub, right? And it pulls your database from, or communicates with the database from Supabase, right? From Supa. And it allows your product to actually exist in the world. Right. It's interesting how software is created. Basically you build it on your computer and then from there you need to be able to make it available for the world. And that's what Vercel does. And Vercel is a great way for you to do this for free because they have a hobby plan. The good news is that Vercel, GitHub, all these are connected. They are integrated into Windsurf. So you can literally just tell Windsurf, I want to deploy to my live environment to, so I could see it on the internet. And if you're not sure how to word that again, go back to Claude and it will do that for you. We'll connect with Vercel and I'll make sure that you can go ahead and find your actual, uh, app living on your domain, your website. Now there are some configurations you're going to have to do with your domain so that your domain is pointing to the right place. It's very simple. You could just ask Claude and Claude will tell you what to do inside of GoDaddy and all the settings. And you do it once and it's over. Hey, if this episode is getting you thinking, is getting you fired up, you can't wait to start building. Awesome. Because I have another episode coming up that you're going to absolutely love. And I want you to hit subscribe. It's all about how do you fire good people that are bad employees? This is never easy. So I want to give you some clear instructions on how to do it right. So you feel good about what you've done, but also do what's right for your company and the person you're letting go. Hit subscribe so you don't miss it. The next step is we need to charge money. Okay, you're creating an app, and you're gonna have to have a way to allow people to check out and buy your product. And this is where Stripe comes in. Absolutely love Stripe, a big fan of Stripe, been using them for over a decade. And they are built for this type of thing. Now, you don't need to understand how to configure Stripe, but you do need to sign up for a Stripe account. It's absolutely free, it costs nothing. And you only charge when you have a transaction and their Transaction fee is very reasonable. I think it starts at 2.9%, 30 cents per transaction. And Stripe allows you to charge people so they can use your credit card, and then that way they grab the money for you and then they deposit it into your bank account. So how do you integrate Stripe so that you have an easy way to check out and have checkout pages and all kind of stuff built in? Very simple. We go back to Claude and we say, hey Claude, I want to integrate my app with Stripe. Claude will then give you instructions. Hey, go into your Stripe account and give me this information. Why? Because it wants to create a prompt for you so that you can give it to Windsurf so you can add it to the code so that Stripe is integrated with your product. The second thing you need is going to need a way to migrate the email addresses you're getting from people, because you're going to maybe want to send them emails, uh, reminders, offers, upgrades, whatever it might be. Okay, so I use KITT as my email marketing software. So you need to integrate with KITT. Again, how do we do this? Well, we make sure we have a KITT account and then we go back to Claude and say, hey Claude, I wanna integrate with KITT. Show me how, give me the steps, give me the prompt for windsurf. And that's all you need to know from a technical point of view to build your MVP this weekend. And that's your mission. Build this thing so you can get in front of real people as fast as possible and let the market tell you what to do next, what feature you should build next, how you should price it, all kinds of stuff like that. So I want to give you a fail-safe plan when it comes to building your MVP this week. I'm going to give you a game plan, what to do each day. And let's start with Saturday morning, right? Saturday morning, you're going to identify the problem you're solving. You're going to create the PRD with Claude so that you have the foundation. In the afternoon, you're going to start building out the different pages and screens of your app. With the features that are required so that you can be able to use this thing on your local machine. So you're going to prompt Claude to create the prompts and then paste it inside of Windsurf. In the evening, you're going to connect to Supabase, which is easy. You're going to sign up and then you're going to tell Claude, hey, I need to connect to Supabase using Windsurf, and it will give you the right prompt. At night, you're going to test everything end to end on your local machine. You're using your localhost to figure out what's right and what's wrong and what's, what you need to fix tomorrow. And that's what you're gonna do Sunday morning. You're gonna fix bugs. You're gonna polish. You're gonna see what's not working and make sure everything is working. Midday on Sunday, you're gonna deploy to Vercel. You're gonna connect your domain and you're gonna make sure that your app actually lives in the real world. Again, not sure how to do that. Go to Claude. It's gonna give you the steps and it's gonna give you the right prompts inside of WinSurf so you can paste it in WinSurf and it connects directly with Windsurf. In the afternoon, you're going to share your app that you've built that has real functionality with 5 to 10 real people you think would be interested in just trying it out for free. In the evening, you can create a simple survey using Google Forms and collect feedback so that you can know what to do next beyond the weekend. By the way, you don't actually need to charge money at this point. You don't actually have to have the Stripe integration. Until after you kind of understood, okay, this is what I'm going to charge. This is how much it's going to, uh, it's going to be. You could just tell Claude right now, this is a free app. Let people to sign in with their, or sign up with their name and email address and they'll create a password and all that kind of stuff. When I did that video a few months ago about how to build a SaaS in 7 days, the response told me something, told me that there's an appetite for more. People want to build things, but they're not really sure how to. I hope I showed you how simple it is and how you could just, just get started. I haven't given you everything when it comes to building out a software company, but this is getting you the momentum you need to start learning on your own. And of course, come back to the podcast to find out how to make sure your business is profitable, how to make your first hires, how to make the most of the product that you have for your customers. But you have everything you need. If you have $40, you can build this thing this weekend. The window of opportunity for non-technical founders to compete in the software space has never been wider. The tools have never been more accessible. The cost has never been lower. If this episode gave you the blueprint you needed, go back and watch the episode that started it all. Can you build a profitable SaaS in 7 days with just AI? It's more of the big picture and what's possible, and, uh, I think you would really love it. That episode shows you real-world results of another app I built called Nicki. That was an incredible journey. This episode you just watched gave you the steps, but the one I did before shows you what's possible when you follow through. Go ahead and watch it and I'll check you in the next one. Thanks so much for checking out the 100 MBA Show. Hey, get going, start building. And if you want a little bit of accountability, me on social media. I'm pretty active on Instagram @omarzenhom. When you are sharing your product with the world, I would love to see what you build. Take care.
