Xano vs Supabase - Open Office Hours - With Ray Deck (LIVE) - Episode 4

Hello, Ray! How's it going?

Well, you've seen another, another week, another podcast.

People really love our podcast and we're starting to get more and more positive comments and people stop calling us boring.

That's just because they don't know us yet.

Now, maybe people will start calling us controversial by next week because today we got a very controversial topic in the tech community and that is Xano versus Supabase.. And I think no matter on which side we are on, we'll try to be as centered as possible, but I think we'll always upset some people. But today we want to try to talk as much as we can about the facts and actually see what real Xano users are saying about it and how Xano is actually empowering people that have never touched code before, at least at that extent that you can do within Xano. To empower them to turn their experience, their business knowledge into products and software out there and automate processes. So how do you wanna start this exciting topic today?

Well, let me ask you this. How did you decide that this was the topic you wanted to be covering today?

Oh, I had like a complete beef going on in the Tuddle Discord about Xano versus Supabase, so.

Oh, now we need to hear about the beef. By the way, it's not Toddle anymore. This is a Nord Stream beef. So we're talking about Nord Stream, NordCraft, right? Nord Stream is something else. But yeah, Vikings and longboats and hats with horns. I want to hear what a beef is like in the NordCraft Discord.

It's a hot topic whenever you start mentioning Supabase and Zen on one sentence. I was making a comment. I was like, I'm looking for YouTube videos. I'm running out I haven't posted in a week. Do you have any ideas of videos I can do that do not include surveys? I shouldn't have sent that comment. I agree, that was a mistake. I did send it. 120 comments later, we found a way to calm down that conversation. So you've been looking at that kind of technology, you've been doing this for 25 years. So it was like, you know, I'm just goofing around since 5 years here. So how about we'll get the expert and we'll bring some— we'll shed some, you know, some light into the darkness and what's actually going on. Maybe it's something, you know, maybe it's not even Subase or Xano. Maybe it's Subase and Xano, or it's you do Subase, you do Xano, and I'll do this, you know. So I want to get with your knowledge in there to kind of like, you know, de-explode that situation.

Yeah, you know, there are the, the Xano, I think, positions itself. I think it's a marketing mistake they make. They position themselves, you know, in opposition to Supabase. The Supabase is a wonderful database backend, highly programmable. They've got great SDKs, but, and they have services that go beyond just the, the database part, but really at their center, there are, they're, they're merged into bits. Database, right? You, you store stuff in, um, uh, your, your, uh, your Supabase database. Uh, and they originally came out in opposition, I think, to Firebase, right? That's the reason why they call themselves Supabase, or that became the, the— they were at the time like an open source alternative using PostgreSQL, and now they are, you know, they got their multi-billion dollar, you know, valuation and, and all sorts of good stuff. Um, and then, you know, Xano comes along, and Xano, uh, wasn't really designed to solve the same set of problems. It's, you know, what you're really getting with Xano is like this, this virtual machine Right. It's a little bit more like renting a droplet from DigitalOcean, except it has stuff built on top of it and its own, you know, management sitting on top. And one of the things that includes is a database, a little PostgreSQL database. So using the same basic technology that Supabase does, but its thing is offering easier integration with their no-code, you know, function stack environment. So that function stack is, I think, where they really Shine. But because, you know, they were coming into the market sort of helping with people doing app building, they position themselves, right, as being, hey, a backend and a frontend. And, you know, the, the idea is that like in— you could have your WeWeb or whatever frontend, and then you'd be able to go to a Supabase backend, just database, or go to a Xano backend, which is, you know, also has its database. But I think looking at it that way kind of misses This is where each of them are really great because Supabase is a super high performance, super scalable PostgreSQL database that's fully featured and like does a really good job. And Xano has, has a database that's sort of sitting inside this machine that you just rented. But what it really has is the ability to manage all the logic that sits between the database and when it leaves that particular machine as opposed to running queries on it. And I think that, that, that thicker part of the backend, that, that, that brain that's in the, on the server side is where, is where they really shine. So that, and to your point from before, yeah, sure. People can and do use Xano for the logic and then use some Supabase for the backend storage or something else for the backend storage, or actually many other stories for doing that too. Because the place where Xano is most special is not in the part where it just competes with Supabase, which is data storage. It's really in that function stack.

We do also see a photo of the function stack right now on the screen. And there's actually a photo of the first API that I built for Despia. It actually really manually—

it manually went—

you can take a look at that still on my IP. It went on your URL and it scraped your HTML file and got your CSS file and got your JavaScript file and put it back together without you having to host it. Because the first idea of Despia was, how about we hosted for you what's on your site so we can bill you by users? Then I was like, oh, it costs money for us to bill you by users, so we gave that idea up. Enforcing that costs more money than actually, it wasn't worth it enforcing it at the end of the day. But that was the first endpoint that I ever built for Despia.

It's cool.

It's like a sacred almost thing. It's the first thing I've built in Xano that really made me sweat. You see 4 Lambda functions right after each other.

Yeah, which is less scary than it used to be, thankfully, in that system.

This is the old version, by the way. This was scary. I think I have somewhere like a 30-second delay just to do some load balancing.

Yeah. I think this complexity you're handling in there, that is totally where it is strong. You know, like you do the same thing like Make or Zapier, you're spending a lot more time sort of bending over backwards deal with like the connection between these services. Whereas what Xano is doing is what you're doing, like extract data, make these API calls, transform that data, ship it somewhere else.

I think those things are two separate things. I think right now when you take a look at what people are building, most people are building some sort of AI application. Whenever someone builds a SaaS, you usually have some engine going on and that is your secret sauce of your business. That's where your IP is. And I think Xano is a great way where you can have a place where you don't need to necessarily be that technical, but you need to understand your expertise of that engine and how you can turn that engine into a product, into an API that can be served to your users. That's where it's really great at. And we're going to even have some special guests today. We'll bring in them a little bit later on to see how they're using that and how they actually get empowered by Xano. But the tremendous thing that I think is when you, like for example, like Silman, I hope one day we can have him in here as well. He has this amazing AI engine to turn, you know, text into full commercials with voiceover, with everything. And he's using Xano as his, you know, central point of logic for those pipelines. And I think you should see Xano more as a, pipeline rather than a database because it's not that good of a database. It's good for most use cases, um, but if you are building the biggest to-do application in the world, you probably should build this in Supabase and not in Xano. But if you're building an AI that comes up with things for you to do and has a proprietary engine doing that and needs to do the data entries automatically you probably want to go with Xano because you can build this proprietary engine very simple in such a function stack that's just easy to configure. It just looks scary because in the condensed view, but it's way simpler if you go to the expanded view.

I really like the word engine that you just used there because engine is defined by its moving parts. Yes. The parts that are in here are things that do, that have action, that get you from point A to point B. You're using the pipeline, but really we're talking about making motion. Whereas database's job really is to hang onto things and hold still and deliver things back. But like it's largely about being static. And the, I think when you are in that more action orientation or when there's more to be done, Xand does really well. That's why the to-do list is really good example of, I mean, it's a silly example, right? But it's an example like the kind of thing that Supabase is particularly good at because in the case of a to-do list, you're going to add to it, you're going to remove from it, you know, you mark it, you update it, mark it as done. It's like very, we call it CRUD, you know, create, read, update, delete. In terms of your relationship with the data. It doesn't require a whole lot of thinking in between. All the thinking is being done by the human. When you need thinking to be done by the machine, that's where you need to be doing more logic. And the question is, how do you express that logic? And I think it's just worth saying for a minute that like neither of us work for Xano. This podcast isn't being sponsored by Xano. The reason we're talking about things the way that we are is that, you know, we're both running these communities and like help people who are trying to build, you know, meaningful applications, workflows, systems, right, that are, you know, creating real business value. And the thing we found is that there's an element you need to be able to store stuff, but an even bigger element that you need to be able to have the machine do stuff. Because if you just have the user do stuff, well then all you've done is like shuffle things around in terms of where the user does work. But if you— but the core of automation is that it has to have an engine, right, that this thing is moving autonomously. And the way that it moves autonomously is going to be through this kind of logic. And there are a limited number of ways that you can do that. To do that. I mean, the— if I would say that prior to 2021, the leading way of doing that kind of logic was probably to write a Lambda, right, like in AWS, and do serverless function, which was, you know, has its own kind of challenges to it. Before Lambdas was actually even more challenging, you had to go deploy a virtual machine and go get that done. But the— but now there are more options, and Xano's not the only one, but it's a pretty good one. And the— and, you know, there's interesting question like what's going to be the right option 90 days from now, but certainly Talking about Xano is a way for us to talk about this piece that matters, which is how do you get the machine to do work for you?

How do you build engines that are proprietary to your business? Because that's actually where the IP is and where you're going to be different than anybody else vibe coding their application. Because if you're going to go to ChatGPT and you're going to build an edge function, it's going to call ChatGPT and it's going to just do whatever AI knows how to do with it. How is that different from someone I can build myself? If you come up with a proprietary idea and you use Xano to express this idea, you're not a programmer, but you're going to express that idea that comes from your domain expertise, build this engine. I think I just made that word up, and I think it's a great word for what Xano is doing. You built this engine, this business engine, doing this technology for your business, you know, working with leads, working with AI, working with whatever. You can commercialize that and you could even then use Xano and add some authentication to your endpoint to actually sell that API to other people, which is also possible in Xano.

Which is also true. We have a comment in from Bia Landfreak who has, I think, is making a really useful point about Supabase and coming on the other side of the argument. I think that this is worth thinking about. He's using Supabase for, we talked about price and SQL logic because it does the SQL part, well, right? And so the question is like, where does that brain happen? And he's talking about using the brain with N8n, which is an automation platform, right? And handling authentication through configuration, which is what's called row-level security. That's what RLS stands for in the comment. And the, which is sort of a different mental model for dealing with authentication. And it's a perfectly reasonable thing. Some people find that very comfortable. And the, but it definitely shows that like, you know, in all these cases we're thinking about how do we handle the automation part, right? And do you handle it through, you know, so Xano has one approach to handling it that I find pretty compelling. N8N does have some really good stuff and is now much more AI native. So like particularly doing LangChain-ish stuff, it's, it's very, can be really cool. And there are, you know, other options like I've seen people also using like Make.com in the backend for being—

Andrea suggested that today. That you can use your base and make that would be a good combination.

Yeah. We shouldn't give the impression there's only one way to do something. But having decided what the problem is, which is there's storage, there's logic, and then there's the front end, the user experience. It's that middle part that I think doesn't get quite enough attention. And in this case, the real choice isn't Supabase versus Xano, it's going to be N8N versus Xano, right? And then like, what are sort of the reasonable choices and what are the trade-offs as you're going among those?

Yes.

Also, there is a difference, I think, when you start your business, you're going to work two-dimensional. When you step into Eris, you're going to start being three-dimensional. This is the difference between using Zapier to get data from something and doing a loop in Xano. From my understanding, the last time I used N8N, that was a year ago, but from my understanding, N8N is very good at being two-dimensional. You can go on all kinds of different ways on that two-dimensional space. But when you need to go with loops, when you need to go with for loops, while loops, foreach loops, this is, I think, from my experience at least from using Make and Zapier, this is mostly where those systems fall apart. And where you start feeling what else is out there. And that's how I found Xano. Yeah, and I found you because I was like on the Xano forum, like, this is like— I, I tried it myself and I was like, how do I do this thing? And like, if you start Xano, the first week is full of rage against whoever built this, and then it gets better.

And when you find Ray Uh, the, uh, yes. And, and having met the man who, who built it, um, the, uh, I, I certainly appreciate that feeling. No, Sean's a wonderful person. The, but, but I, I do think there's like your, your point about dimensions I thought was a really good one. And this idea of looping, and there's also a question like where the incentives are. So neat one because N8N has like a self-hosting option in addition to having like just sort of cloud-based hosting. And then you look at like Make and you look at Zapier, both of which are basically selling, you know, things by the action, right? They're on a meter. And the— in both those cases, one of the reasons why looping is so challenging in, in these circumstances is because they are designed to minimize the amount of compute you're really spending, right? So they want everything to run as fast as possible because they want to get more of those actions and spend fewer dollars on the compute. And that's— and which is, you know, reasonable, you know, in terms of their incentives. But it becomes why the loops are hard, because if they loop, then they need to spend more on compute. In fact, they created something that Maybe it's not going to end for a long, long time. How do I manage that? How do I make that work? One of the neat things about Xano is that it has that— I was talking about being a virtual machine, you have your own virtual machine. That is simultaneously the thing that actually constrains you when you're trying to scale because you need to go buy the next machine up. It doesn't scale smoothly, it goes stepwise. But the other thing is in the context of having that machine, they're like, hey, crunch all you want. Do as much as you want in here. It's your machine. Doesn't cost us anymore for you to be using that machine heavily versus lightly because it's the same.

They're ready, you know, if you work with load balancing and Xano and if you find creative ways to do things. Yeah, I found you can save lots of money if you're very creative in how you structure things in Xano, if you structure them correctly. The most common performance killer in Xano is user error, actually. Of course, always. But I've seen things that have a loop in a loop in a loop in a loop in a loop in a loop in a loop. I was like, you can just use the filter function instead and save your loopity loopity loop.

Yeah. That idea of just being able to crunch all you want, I think, is the whole thing of being serverless versus servered or versus being what's called instance-based. And the, and that question of like how you want to use your system is one going to drive what your bills are going to be, but also kind of explains why these things price the way they do. Because since Xano is giving like, hey, you've got your own machine, they need to charge you for a machine, right? And that sort of is driving their economics. Whereas Make, for example, could be like as cheap as like $9 a month or something when it starts. And you can do a whole lot with it without having to pay much because they are limiting the amount of compute, which is keeping down their expenses relative to you. So like when you start, you cost them nothing, you do a little bit, they cost you a very small amount. Whereas with Xano, you can do not use it at all, but still cost them the same amount because they have to put that machine in place for you.

We actually got one comment. I want to look at that and then we're going to get some guests inside of here and see what their experience with Xano has been. I think that's a good one. I've been hearing this all the time. What about edge functions? Can I just use an edge function? Why do I need Xano?

Sure. Edge functions, or more generally, you know, serverless functions, are what I would have used all the time back in 2019. Yeah. And they, they, they, they do the job. They work fine. Supabase offers their own serverless. They call them edge functions. They, uh, you have, um, you know, AWS offers them, Google offers them, DigitalOcean offers them, you know, Azure. I mean, everybody has their own version of these serverless functions, which is basically a They give you a URL, you can call it, it will run a bit of code, it will deliver value back to you. Those things can work really quite well. They have their own challenges in terms of organization and what have you, but probably the biggest thing about them is they're all written in code. They are also written in code and solving problems that are a little bit more challenging for AI to understand. Right now AI does a really good job understanding how to do UI stuff. But this business logic stuff, especially when you care about it with a great deal of precision, becomes a little bit more challenging. Maybe the AI is going to be getting better at that and that will cause more competitive advantage towards doing it more in the serverless function or edge function approach. But right now I'm finding that using— I find I will use something that looks like what you have, which is I'll use the no-code for the structuring of the overall flow of it, Right? Yeah. And then I'm using lambdas, right? I mean, I'm dropping into like just, just enough code to deal with these more complex— I'm dealing with an SDK or doing some math or whatever that works better in code than it does outside. And because the thing is structured this way, the ownability of it goes way up because I'm not spending my code lines on table stakes stuff and trying to get people to understand everything to understand anything. Now you can understand a lot by looking at the structure and like, wait, hey, here's a little hard bit. Great. I'm just going to skip over that and go to the next line. Totally fine because you can still get the overall understanding of how logic works without having to be committing to learning how to read Python or TypeScript in order to do it.

I really like this approach that Xano is taking. It takes ownability to the next level. You only have to focus on the hard parts and you do the other parts the smart way. And you save lots of time. And that's actually what it is. The idea of Xano is you're going to spend $50 more in the software to save $2,000 less on developers. That, that's the idea. That's why I like Xano. It, it adds up at the end of the day.

I, I, I— that is, I think, key to any tool choice or automation choice is like, you know, you're spending a bit more in order to not pay for the most expensive thing possible, which is human labor.

Yeah.

So I think now is a good time, we're halfway in this, yeah, against this in here, that have been using Xano, have been doing tremendous stuff in it Let's get that all started out. We got Tara inside of here and we got Jack, and they're in the same room even though he got a nice background. I'm going to put him on mute that you can hear him through Tara's microphone. But they've been using Xano and they've been building some incredible stuff with it. They don't have the background of being the developer. They worked with developers before and I think This is the target audience for Xano, and this is really where Xano empowers people to build creative and tremendous stuff that they otherwise, they would have, you know, they would have to go to developer who doesn't understand them to then spend lots of money and time to build something that they don't like to then just start over again. And it's actually the point where they are right now, and Xano has been empowering them to do some exciting stuff. So without further ado, Tara, Jack, Thank you so much for being here, and we're excited to hear, you know, kind of like a little bit of your journey. How has it been working with Zano? And, you know, just tell us everything. Oh, I think we need to get Jack on here. Yes, we can hear you now.

You can hear me now? Okay. Hey guys. Hi Ray. Hi Yasin. And Toria is actually sitting right here.

Hello guys. You guys can hear me?

Yeah, you might be able to hear it from here. So if there is like a voice overlap, just Perfect. Tara, do you want to start with your own, or should I just go through, get mine out of the way, and then you can, you can continue? Because she's the main developer, uh, yeah, so you probably know a little bit more, uh, that she takes care of that logic. She— we want to own the intellectual property. Um, do you want to explain that later, Tara?

What do you want to do?

Yeah, you can quickly start and then I can take over.

So, um, I would like to start this by first, uh, you gotta have some sort of domain expertise in what you're doing, in my opinion. And we are coming from non-technical backgrounds. The biggest issue that we face is actually not the concept of trying to learn stuff, but being able to— being dependent on other people. And this was one of the main reasons why we kept failing, even though I don't want to blame others. It's my mistake that I had to wait on someone or rely on someone to get me to the places that I wanted to go as a non-technical person. And that, that cost me, guys, like my life savings, right? Uh, if you were supposed to go through that whole journey of having kids, owning a house, and, you know, uh, buying one more, whatever, uh, and then die happily after— that didn't happen because we were— even though we were using offshore services, we were giving money to people expecting that the product was going to be done, and then it was just taking— you guys know the, the process. Essentially what happened was we kept failing and then one day it was around the COVID time. We said, you know what, we gotta learn some stuff, man. We can't just rely on people. We are in lockdown. Probably the world will come back. So while we have this time, let's start learning. So I started with the classic HTML journey, whatever. And then I was able to find someone in the places that I was lacking easier then. But again, I was relying on the person to build the product. Essentially what happened was fast forward, I'm like, okay, I'm not gonna have anyone. It has nothing to do with equity split because I would, I was very generous to offer equal, if not more sometimes, but we need to have some sort of ownership. The second I made that decision, which is incredibly hard for me at that time, I realized that there are tremendous amount of opportunities out there for people like us. You have Ray's community, you have Yacine's community, and thousands of others or more people that are experienced or willing to teach you every single thing. And essentially, if you understand your business, if you understand the kind of value that you, you're providing people, and I'm not talking about the hype stuff like have a nicer background app or something like that, but actual value that you provide people, I think Every tool is there for you to build. Now, moving towards the business logic of things, Xeno was incredibly helpful versus Supabase because I don't have to learn anything. I can just literally use my human language to choose particular blocks and then start building it. Now, for technical people, mainly what I'm hearing is, yeah, but it's a lot more expensive. You can get it done faster. Well, the goal for me is not to save money as a, as a limited resourceful person that I'm a solo or like a dual entrepreneurs. The main goal is to generate more revenue, not to just preserve cash that, you know, like limit my spending. I want to get stuff done as fast as possible, being less dependent on people with as less maintenance as possible. That's the goal for most of the entrepreneurs. But the second when you say, all right, it's much cheaper to go through Supabase because you can actually write the same thing and it will be fractional of the cost. Well, if your goal is to constantly save money, how could you think about the growth? So the abundance is there, but you just have to get into the right mindset. So that's my experience with building stuff, but some might agree, some might not. And I guess it depends on where you're in your journey. If you are a technical person that is working for a massive company, of course, your priorities are going to be different. But if you're just launching an MVP and you want to get it out there as soon as possible and you don't have any resources and you're not technical, unlike you guys, what's the option? What do you need to do over here? My goal, get it out there as fast as possible, as less maintenance as possible. And, and start your journey. Because the, the real problem starts when you are actually selling, not building. That's what I realized. You build stuff, it's amazing, right? Like, my wife makes amazing meals, but you guys have no idea about it because she's not selling them. Like, nobody knows about it. So, you know, it's good for me, but, you know, I don't pay the, the— I don't pay her for that, right? So that's my experience, guys. I don't know if it was helpful for people. Hopefully it is.

I think it is helpful for a lot of people because I think there are way more people like you watching this that are asking themselves those questions. Which way should I go? Should I sacrifice some of my time because people on Reddit say you should use Supabase and no-code can't scale? Or should I use tools like Xano to get it out there faster and see if there's a market for it? And can I even scale with that? I think those are the questions we see all the time. I think me and Ray, we can show you all the specs and all the technical stuff why it's actually a good way of doing it. I think as long as we tell that, they will always be like, I'm not sure, they run their communities, maybe they just say that to get us in there.

Yeah.

I can definitely understand.

Also, it doesn't help that Ray has the Xano logo in the background and a blue t-shirt.

Yeah, it looks like he's being sponsored, isn't it? But he's not.

He's not.

He's not.

Yeah.

And he does have great videos and comments that I found on internet about Firebase authentication, which I think many people will have to go through that pain. Xano authentications versus Firebase authentication, stuff like that.

So yeah, he made me an advocate for Xano. And you're also part of his community. How do you like it so far at Ray's community?

Well, I'm surprised, like, uh, you've been doing it for a while, right? I didn't know, right? I had no idea. But the, the wallets, I mean, I could just type exactly what I want and then I could find any kind of like, uh, stuff there, like with Tara, because you're building, uh, other than that, you're very responsive. So thank you for that. I don't know how you manage it, if it's AI or you, uh, but like you post something and then you get something like, this is a bot, you know, uh, maybe he has some NA then running at that level that I don't Oh yeah.

If you are also interested in taking a look at Ray's community, scan the QR code on your screen and join statechange.ai. If you're stuck at anything Xano, N8N, Supabase, he is the one to help you get unstuck. But I would also be super excited and super appreciative to hear what your experience with it has been, Tara, because you've been doing actually the heavy lifting working on it. We've been doing this in our office hours as well. So I would love to, you know, kind of like get your point of view on Xano and, you know, kind of like how it empowers you to do exciting stuff.

Yeah, for sure. Hello everybody. Uh, first of all, thank you very much, Justin, for inviting me. Uh, this is gonna be, I think, my first public speech, uh, about this, um, this, uh, project Xano and all the technical part because I can say 3 months ago, I didn't know anything about Desano, Webflow, Tuddle, FlutterFlow, nothing. I'm a real estate agent. And we, me and Jack, we were just entrepreneurs. We wanted, we have organizing skills. We are using the CRM systems and we just want to come up with some ideas for the real estate agents and start our application building journey, like with this and with the different developers. So we did maybe 3 or 4 different projects in the different categories and work with different companies, partners, developers, coders, no coders. And then every single time, because I was doing the entire wireframe, there is always a communication problem. I never ever like reflect on what we need to do with this app. And all the developers, I'm always saying, okay, they don't understand me. They don't understand the logic. They cannot build what I want. They cannot figure out, like, you build something and you miss the back button, you miss the cancel button. So they don't think. That was my, that was my problem back then. And we always hear that one of the founders needs to be technical in order for you to become a successful product owner. I never ever understood what does that mean until this time. Because 3 months ago, I start my journey with Yasin, with the help of Yasin, we work with someone else. And then we realize I need to take care of everything. And I start learning from scratch, the JSON files, the Lambda functions, database, what does that mean? How we can function it with no-code tools like Tuddle, Nortegraph, now FlutterFlow. So we experienced so many different things and we start something and building the product for contractors, for homeowners, for construction, construction, for construction industries. In that product, we have very complicated functions. And it's really tough for a no-coder person, especially like someone for me, I, I cannot find how can I manage it. I know in human language how can I describe it, but I need to implement it to my product. So with Xano, what I really like about it, first when I started to work with Xano, I had 10 databases, like everything so simple, one-to-one, one-to-one. And now today I'm rebuilding my database. Now I have 3 databases. 3 main databases and it has everything inside it. I have the Lambda functions. Now I'm like, I'm masking some data inside this. I know I'm doing some security stuff. I'm creating my own Lambda functions and everything's over there. I really like it because what I'm thinking, I put it in that function stack and I put something here. Everything's operating inside that function stack. And I can send it to my frontend, which is very helpful for me as a no-coder person. I completely understand every single thing. I can search and find some videos about it. It's very clear language for me, and English is not my main mother language, and it's completely easy to understand. I don't need additional terminology. Of course, with the help of the AI. But the biggest things, you cannot rely on the AI only. You need to think about it, which then gives you that enough platform to think and write it down under the functions to have whatever you want from it. It's amazing tool for me and I can say, Yasin knows and some of the friends from our community knows that as a woman, I build a product within 2.5 months. By myself. And 3 months ago, we worked for one hard-coded— we worked for this specific product, one hard-coded person and one designer, plus Jack and me. Now I can see 90% of the things at the backend I did by myself with the help of Xano, of course, Yacine, and the communities. And now hopefully with Ray.

Congratulations! That's exciting.

Here.

Well done.

Thank you. Yes, I think it's a very empowering message for lots of people to show them that you don't have to be technical to get started. The only thing you have to do is start though. Most people get stuck at starting. They think about, I got this app idea, I want to build it, I know how to do it. But they're like, is this the right thing? I think too many people get stuck in planning. And like, should I use this or should I use that? You hear this person talk about this not being scalable. I think the best thing you can do is get started. Even if you start in the wrong direction, at least you know which direction you shouldn't go anymore. I think that's something very valuable that we got from their story because if you just start, you'll eventually get to the goal.

I agree and I want to add—

oh, sorry, Ray, go ahead. I, I was just going to ask a quick question of Tara. I'm curious whether, like, if we sort of rewind the clock, like, back to being in school or earlier days, did you ever have any coding experience?

No, zero. But my mom, my uncle, my entire families, uh, they're like, uh, very old hard coders. They coming from, uh, like the— they work with IBM and Microsoft. And my mom always told me, I don't, I never ever want you to get into the software industry. Cause I remember back then they were dealing with the AS/400s and all the logics and like they, they got a phone call and they're like, oh my God, we have a bug, we have to fix it. And they got stuck in the office for maybe 2 days, 3 days nonstop. And she was like, no, I don't want to teach you anything about it. Just find something else for you. And to be honest with you, this is a very special thing for me, but I want to share with you guys. My mom, um, she— we are from Turkey, and she was working for, uh, she's the one of the first barcode systems and the CRM systems who built in Turkey for the biggest companies, like, and she was working with the IBM and Microsoft And my uncle, she is one of the training, training person in the IBM. So when I first started getting to this journey, because I don't have any code background, 2 months ago, she told me that you're never able to do that. You're never ever going to do it by yourself because you don't have any code background. You always need someone. You're depending on someone. She highly discouraged me. And for 2 months, I didn't mention anything about it. I didn't show the product I made it. And now she actually— maybe she still doesn't know, but now I know the database, I have my own product, I can write the Lambda functions, basic ones I can do it by myself, but the complex one I need to rely on the AI. Now I have something in my hands. And yeah, so zero, zero background, nothing about code.

That's very impressive.

Thank you. So it was in the water, but not in your training. Around you, you're surrounded by code. But like the— no, I think that, but that's such a— first of all, I love the I love the story. And I think also speaks to like the demographic moment we're in, right? We're like, hey, sure, we haven't actually been trained in this stuff on a more formal basis, but you've had the idea that this is a possible thing, right? It's not something that people like you can do as opposed to many of us who are, you know, coming like, oh, well, tech, that's done by people over there, right? And like, you know, this was sort of, you know, inside. I think this is, and the technology that sort of allows us to fulfill that promise is what I find very exciting about this. I really appreciate you sharing your story.

We actually have— Thank you very much. I've been following Tara's progress and it's impressive. To do what she's done in 3 months, you need at least 2 years of learning code. That's Daniel who was also in our—

Daniel, thank you.

I think you're doing amazing. Then someone, Daniel again, 100% spot on. And then he also said, oh, love these guests.

Thank you, thank you, thank you, thank you. Uh, one thing that I wanted to add, guys, uh, I think this is going to be very important for entrepreneurs as well. Um, one thing I noticed because I failed so many times, which I'm grateful for, uh, now I can say, uh, There is one quote that really touched me. I think that was by Warren Buffett where he says, "It matters not how hard you row, but it's a matter of which boat you're in." So before I had ideas that it was just like, you know, I wanted to build something. The MVP wasn't feeling like it was enough. So that's why I needed to get more people involved where it was supposed to be more technical. Then I noticed if what you're solving is truly valuable that people would really pay money, they would really want to get into your space, like Ray's community or Yasin's community, because the alternative is incredibly expensive to get help from someone else, to have technical person in your side. I mean, you're not going to think twice, even though if the interface is shitty, it doesn't look nice, whatever, which I'm not saying it does look good for you guys, but even if it was like simple as, I don't know, like a basic HTML div there, like join, people would click. So Sometimes we get confused that we get hyped, oh, I should use the best tool, this and that. However, one thing I notice is if it's a bus, if it's a car, or if it's whatever, Ferraris, Lambos, whatever, you just need to get to that space and don't get stuck so much in the, in the tool that you're going to use, but focus on the end goal. Maybe this is going to come like a super business-oriented person, like a mentality, and I don't want to say that. However, I also made a huge mistake by jumping into the fancy schmancy shiny object syndrome products. Hey, let's use this. Hey, let's use Make. Let's use n8n. Let's use Manus because it's like the DeepSeq is the new thing. Just use the tool that you understand that is simple and then move on.

I think one thing that is important among all of those shiny object suffering people that suffer from shiny object syndrome, I know lots of people who've been jumping from frontend to frontend to frontend to frontend to frontend. You just jumped to FlutterFlow actually. So you're jumping all the time too. Did you ever consider jumping away from Xano?

No. Tara is probably going to talk better because as we see, it's going to hit you if you say yes.

No. So that's my baby, Xano.

And sometimes Yeah, there you go.

So, so no, no, even sometimes Jack wants to add something inside the Xano. I'm like, no, that's my territory, you cannot get into it. That's my function, that's my logic, and I'm trying to explain it to him and like he cannot understand. I'm like, I can understand it. That's— no, don't jump into my territory.

Two, two more weeks and he's going to understand it. Because it's so simple to understand. But also I think Xano got to level up a little bit more on the collaboration tools so you can kind of like merge a little bit like WeWeb actually now has it where you can see the code edits and all that kind of stuff. I think there's some improvement for that. Maybe Supabase has a little bit better role management system going on. Um, but yeah, I think this, this shows what's possible with Xano, with that technology. And thank you so much, Tara and Jack, for being here. We really appreciate you. Thank you. And yeah, Ray, do we want to bring this to a slow ending for now?

I think I really appreciate the, you know, your all perspectives today. And yeah, I think that's a real high note for us to end on.

Wonderful. Let's thank you so much all for joining. And Ray, so let's get this to an end here today. And with just offboarded Tara and Jack. Thank you all for joining here. But we got some comments from Ellie, a StateChange and No-Code Pro Code member, because there isn't really any alternative. I'm not completely sure.

For no-code, I think for backend is there are Alternatives, um, I think Xano right now is, is very much best in class. Um, we talked about N8n before. Uh, there's another product out there called Buildship, uh, which is, uh, you know, good for that kind of backend logic. Um, and but then once you get outside of that, you really are talking much more about, you know, putting either code together or stitching together like an Airtable backend or what have you that has a different, uh, quality to it. Uh, yeah, so, uh, yeah, the, the, and And I think that this, it's a business challenge, you know, like the, the vibe coding front end, woo, it looks fancy. I can get lots and lots of customers and they, they, they, they just climb so fast, right? There's lots of money to be made. Whereas on the backend part, like it only works, it, it only works as a business if you're gonna ship the product. Uh, like all these little don't really make any money. And the, um, and that I think is a tricky thing, right? For being able to be successful in this, uh, you know, the, the backend logic business. And yet it's where all the value is if you do ship.

Also, people really love this topic. What's up everybody? Psyched to see this convo happening. I think Xano versus Supabase has been something people have been asking all the time. Nobody has been doing it before. I mean, Xano wrote a blog article about it, but it— I mean, even I have to say it's a little bit biased, and you know, I'm usually the one who's a little bit biased. But I think this has been a great conversation about highlighting what's possible, how Xano adds value to it. Now my question to Ray is, you've been working on the hardest parts of no-code. Lots of people who are in state chains use Supabase actually. We now have the success story with Xano. What are the success stories and the use cases where Supabase would make more sense?

Yeah, so Supabase is gonna make a lot of sense, particularly if you are scaling hard. Because it scales well upward. I mean, look, there's a story where Supabase is great if you're super, super tiny because it could keep all your costs low because it's all serverless. But that's, those stories aren't that interesting to me. Like I think you're building a successful business, like where Supabase is very fast, it's very scalable. So you can like start running up a bigger database, supports things like sharding. It does that stuff really well. And then also supports more advanced features. That are built into PostgreSQL. Like it's much better for full-text searching. Say Xano has a little bit of support for full-text searching, which is basically based on multi-word selection. Whereas, you know, Supabase supports something called trigrams, which is a bit more technical, but basically it's a, that way you can look up part of a word and it finds it, you know, more like tokens. And so it has better, you know, search engine type capabilities. And the, And people have talked about like, you know, the RLS, people really like the way they do that configuration as being a good way of doing things. And its SDK is fantastic. That's a key difference right now between Supabase and Xano is that Supabase is programmable. And actually using Xano to program Supabase is a really interesting use case. But that programmability and that idea like, hey, you can tell your Supabase what to do and like how it's gonna set itself up and run these big things through not through going through its user interface, but through using its APIs and SDKs is I think a real competitive advantage. Another thing we've seen in state change is as people grow, the amount they need no-code goes down because they're comfortable with expressing things directly to the machine. JavaScript and Python, they're not that complicated as languages go. So you, they were designed to be easy to use. And so like as you climb that curve, you find yourself, okay, well I could use code. So now instead of it's like, I can do this with no-code versus code, it's like, what is the best way for me to do this? And that, and when we start having those kinds of questions, now we're creating real value. Yeah.

And I think the way you get there most of the times for many people is by starting with something out like Xano. Like, like your poster child is Cam here.

Yep.

Uh, right, Cameron's a member of our community who, uh, also came in here like one of our next podcasts if we can get that arranged, because he is like prime example from no-code to code. Yep. He's not using Supabase though.

He's not using Supabase, he's using— but he's using the, the old competitor for Supabase which is called Firebase, which is a Google product. And the— and but even then he's like, well, I need to move off of this and considering whatever, because as he has, you know, and he's running a real business, he's like, he's a financier. He's, you know, doing things for like doing, you know, a product for like, you know, leverage buyouts, kind of things private equity firms do. And he is— and it's all about the math, right? And trying to orchestrate being able to do this complicated math so that people can make better decisions about where they're going to spend tens of tens of millions of dollars.

Yeah, we got another comment from Ellie. I think once they introduce cheaper pricing to compete with Supabase, then the conversation would be completely different.

Yeah, it's not really going to compete with Supabase. What's going to happen is they're going to offer a new low-end pricing, right, which will allow you to break in and make a very tiny app, right? And so the— and then the, um, and then the— to my understanding, the high-end Xano hasn't changed its architecture, right? So it's not going to a serverless model where it will scale elastically, right? It's still gonna have this big step function thing going on. It's just that they're gonna offer a tinier, even lower capacity first tier, and that will be cheaper, right? And that'll be more affordable. More people will be able to do it. And I think that's a, there's a reasonable point in there, but unlike Supabase, which can offer a completely trivial starting point, Xano kind of can't just because the way that the architecture works. And an interesting question to me is whether them dropping the prices the way that at least I've gotten the hint that they're dropping prices— I'm not totally read in on the initiative— is going to meaningfully change the way the economics work.

I think so too. With Xano, you have to— you've got to get your machine, and they will have an easier entry to get your foot into the door of having that machine. Which for most applications, if you are, if you're right now with Supabase, because you can be on the $20 plan and you are having like a, like 50 users on your application and you're in space because you're concerned about scale, which I think in itself is funny. Because, you know, like, let's first get someone to the application before we worry about scale. But of course you want to plan for scale. Makes sense. I think if you're in that position where you use Supabase because it's $29 or $25 and you say, you know what, I could save myself some time, I could save myself some AI headache. I don't like to scream at ChatGPT anymore when it does my Lambda functions. Maybe when they start getting the robots walking in the street, they're going to put me on their hit list because I'm always screaming at AI or something like this. If you consider having an alternative that's similarly priced, I can't talk about the pricing, but it will be, you know how much Supabase is costing, expect to pay the same on Xano moving forward. That would be the lower entry point. That is if you got 100 users, 200 users, you're using Supabase right now 'cause it's cheap and you're on the paid plan because you want some of the extra features and you want the convenience from Xano, that's good. The moment you're at 2,000 users or 1,000, you probably wanna upgrade to $99. But then you probably also wanna upgrade with Supabase to get some features like GDPR, all that kind of stuff. So at the end of the day, it just provides you an easier way to get your foot into the door. Supabase has been doing this, but the way they scale allows them to do it differently 'cause it's serverless. They have one big server and it all goes through there. Xano is just giving you a little tinier piece of the server to get your foot into the door. But it's actually what most beginners that start their applications, they don't need. They're lowering the barrier to entry. I think that is essentially what's going on. Then the nice thing about Xano is if you implement things like load balancing, make your workloads workflows more efficient, you can run, I think, a far away with the $20— I can't talk about the pricing— with the cheaper plan.

Yeah, I'm really curious to see how the new pricing gets used. But even more than that, I'm curious like how the different functionality, how XAML is going to be used in the coming coming, you know, a couple of quarters here.

One said, uh, that's a good— that's actually a good question. If no-code isn't an issue, what's the better option? So if you don't care about code or no-code, Supabase or Xano? They're really trying to pinpoint us down here.

Yeah, well, it's about the job done, you know. Like, the, uh, I, I thought, you know, you know, Tara talking about like her, you know, having grown up in an environment where they're writing in Fortran for AS/400 machines and like the kinds of things like you're lucky you fit into a small room. I mean, it's, AS/400 actually isn't that big, but that idea of being able to work with the technology that makes the most sense at the time, I think is really where the exciting opportunities are here. And if you're, into using, I don't know, like Amazon Web Services, right? I mean, a lot of what you're controlling in terms of managing your infrastructure is through its console, right? There's a story to be told about how a lot of the things that we think of as being relatively high-end are actually very no-code. AAA video games are made using something called Unreal Engine 5, right? Unreal Engine 5, its main language is called Blueprints. It's a no-code boxes and arrows logic building system. So the— so there are lots of environments in which, you know, these no-code tools or visual development tools or whatever you want to call them, in fact, the variances on them are appropriate, right? Are the best tool for that. Not just like easy to use, but like, hey, this is the best way to express. And then you add in that little bit of text-based code to solve a particular problem. And that model is not like unique to Xano or unique to you know, the class of no-code tools that came out starting in 2021. It is something we've been seeing evolving in software since the days of client-server in the mid-1990s.

And OutSystems is making lots of money doing enterprise no-code stuff.

And lots of people—

absolutely. And Appian and Boomi. And, uh, there's a— there's an area called business orchestration and automation technologies. It's called the BOAT. Uh, it's a whole sector for enterprise, what we would probably call enterprise no-code.

How much big enterprise software used by the biggest company out there is actually just no-code.

Yeah, well, it makes sense to orchestrate things this way, and this is really getting to Ellie's question, right? Like, if, you know, what if it's not the issue, what do you use? And the answer is it often makes a lot of sense to be using something no-code-ish to be describing this kind of string and logic. It often makes a lot more sense to be using code for certain jobs. You use whatever the right tool is, but rather than like saying you have to, you know, you have to eat your vegetables, you got to have everything be in code. If you have anything in code or you're committed to using, you know, Xano's weirdo, you know, no-code language or to get anything done, you can mix them up. And that mixture that allows you to put strength against weakness. And that is really how you get a lot of leverage from this. When I think about Tara's story and like how much she's able to do through organizing the databases, right through a no-code interface, which is the way everyone would do it. Being able to express a lot of logic here using this, you know, visual development stack interface, which I think is really the number one value proposition of Xano is in this function stack. And then, and then as she gets into more complicated stuff, being able to bring in that bit of code, like not like having to write thousands of lines of it, but be able to have just the right 20 lines that does value. I know people who make $1 million a year and their job is to drop the right 20 lines of code into these environments for things like, you know, you know, customer success or digital enablement for like large banks to be able to make their websites and their online tools work. And the, and so, and the reason is that it's not about writing thousands of lines or all the weird stuff. It's like what part matters and being able to focus on just that part that matters. That's the promise of these tools.

And that's exactly what you can do in Xano. You get your variables, you get your API codes, you get your building blocks for the things that make sense. And then where you start doing your secrets, you get your Lambda functions, drop in the 30 lines of code, import any from the millions and millions of npm packages, use it in there, modify it, build your own package, import it if you like. You're good to go. You got it. It's not about having fancy code and it's not about writing code that makes you a good programmer. It's about actually writing less code and writing better code that solves the problems that people are willing to pay for.

We have one more comment in the queue here that I think really speaks to that dichotomy you just raised. Maybe we could have you take on answering this question.

If I had to choose 100% code, the opposite has to be visual. Would you do it for the frontend or the backend? That's a very good one. I think most people have the assumption— we had that in the Discord conversation, it kind of like got a little bit out of hand because it is a passionate topic. People are passionate about the way they build stuff. But if I had to choose 100% code frontend or backend, I would never write code on the backend because the backend is how I— I mean, we're doing the right code on the backend, but I would choose code on the frontend, no code on the backend. And I would choose Xano all the time if I had to choose because that's actually what we're doing, you know, in Despia. In Despia, we have the whole backend built on Xano and we got the things connected to CodeMagic and do all that stuff in code that needs to be done. We'll connect them as a microservice, bring them back in as a function into the Xano interface. Then let's orchestrate only the code that matters, that makes us unique to all the competitors, and it lets us put this code in loops, in conditional statements, in this easy-to-restructure database, in this easy-to-restructure function stack. Because what you don't know about Despia is We're every day, almost every second day at least, almost sometimes even every day, rebuilding the Despia endpoints, creating new Despia endpoints, new versions for them. First of all, our version control is pretty bad because we just keep creating new endpoints in Xano. We need to probably upgrade to Enterprise to get some more features. But the way we do it is because the code Despia generates always has to change. Based on all the updates that Apple has been doing, all the features we're adding to it. So if I were to do this in code, I would always have to rewrite the whole function for all the stuff that doesn't matter, the loops, everything. I got to make sure the syntax is working. Xado allows us to just go into the function stack, get this thing out, put this new thing in, change that file, add that, you know, app clip file, put it together, done, send that to CodeMagic. So this orchestration on the backend, because that's where we're making the money. If we were to put a Despia API out there, people would pay for it. The frontend is not critical to Despia being functional. Now it's different if you take a look at Toddle, like the frontend is a completely different thing for the Nordcraft Toddle company because they are making money by providing you a frontend tool. So their critical business logic is on the frontend that, and they don't care, you know, using Xano or Supabase. Because the critical things, the JSON you're building that builds your application, happens on the front end. With Despia, all the magic is on the back end. That's where we do the code signing, that's where we build the files together. We're actually putting 2,000 files together in Xano, and we're still on the $300 plan, which works great. Like, Daniel was like, you know, we should go to Enterprise. Um, so even though I don't have to, it's just he's making it very look He's making it look very attractive to us, but we don't have to. That's the nice thing about Xano, it's your choice. I would never go away from using Xano for the backend. I would rather rebuild the frontend of Despia in React, bytecode it. Because if the backend is right, I don't have to worry about it. That's my answer to that.

Ray, what do you think about it? I would add that the AI is really good at making frontend code. AI is all about training. Okay, it is a token-seeking parrot. It knows largely what it's read in the past, and the more it's read, the better it knows it. And so when it goes and ingests all of GitHub, which is basically how they work, you know, whatever has the most code in there is the thing that's going to be smartest about, right? So it can do a lot of languages, but like, if like you have some weirdo language, it doesn't know how to do it. But you have a language that it's seen a little bit of before, A little bit can do it, but I've seen a lot of it before. Like, and what, what is in GitHub more than just about anything else, particularly as it's built up over the course of the last 15 years? That's React, right? So like, so it knows React really, really well. It knows Python okay. And the, and the, and as a result, that front end code, it can generate with a great deal of skill. There's a whole lot of Tailwind out there from the course of the last 10 years. Great. It knows its way around Tailwind. You can use Tailwind for a lot of control. So, uh, in fact, when I am building for the front end, I am largely writing code at this point. I don't use that much in the way of no-code builders anymore, even to the point that the StateChange AI site I recently built out using Cursor, uh, to, uh, Cursor and Windsurf together, uh, to go build out a, um, a Next.js site, relatively simple one, because then what I focus on is the writing on it, right? Like what's the messaging that I'm going to be doing for people and sharing the assets, what have you, but the actual, you know, superstructure of the website of it, just allow it to be, you know, if you will, live coded. But I really don't like that term because it implies you're turning off your brain. I mean, you're— but it's just sort of putting the whole— putting the technology to work.

Yeah.

And I found that I was able to move really, really fast that way. But where I wanted to maintain control and have precision is going to be the logic, deciding what happens when people do X, doing things like— there was an allusion before to like StateChangeVault, which has resources on like from all of our, you know, almost 1,000 events we've done over the course of the last 2 years plus. And all of those are, those actions are being controlled through business logic. That business logic is being run through Xano. And I would, you know, continue to choose to use that kind of visual tool for doing the backend and to use code and allow the AI to be writing the vast majority of code for me on the frontend.

So you're sticking with code on the backend? You're sticking with no-code on the backend, with Xano on the backend? Code on the front end because that's where AI can serve you best. I like that. That's what we're doing too. Most of my total projects are a mix out of random frameworks put together, thanks to ChatGPT, to get things to work that I, uh, that I'm a little bit too lazy to figure out how to natively in Toggl. So, you know, I think my total projects have more to— have more in common with your cursor projects and anybody else's tunnel projects.

You can tell that we enjoy talking with each other because it was almost half an hour ago that we said we should bring this into the video.

I thought we were at 20 minutes or so.

Perfect.

Now everybody watching, take a look at StateChange AI. I think it's definitely worth it. If you're using WeWeb, if you're using Xano, Supabase, BigQuery, whatever you do, if you're essentially, if you're building an application right now or consider doing it, you want to make sure you get it set up the right way from the start. Ray is incredibly knowledgeable when it comes to everything out there. You don't want to miss out on his guidance. You can start getting his guidance as soon as today for the small price of $39 a month. Essentially, you can get the StateChange app on your phone, on your computer, and you can always have a mini version of Ray in your pocket with you to ask him when you have a question.

My goodness, that's a high standard to live up to. But Yacine, can we make sure we put up NoCodeProCode for a second here?

Sure. If you care about making fancy buttons. No, we also do a little bit of backend stuff. I mean, a little bit too much, but I like doing the backend because Xano is so enjoyable. No, but if you're committed into using Xano, if you're committed into building your application, also making what's connected with Xano look pretty and, you know, dealing with frontend npm packages and you want to get some libraries into your frontend, you're using Nordcraft or WeWeb, right? Because I found my way around hacking stuff into WeWeb now, so I'm good with WeWeb. You know, if you consider that, take a look at that. We're pretty frontend focused, but it's more like how do we get the data from the backend on the frontend? How do we get the proper data from the front end to the back end and how we do the things on the back end. Now it's a little bit more holistic approach. It's more expensive to start with than Ray. So I think if you're just looking for general advice on how to structure things, I think Ray is probably a better option. But if you kind of like wanna have, you know, the daily help, if you're a beginner, you have lots of questions, you wanna join our daily office hours, make like Tara the best use out of that, take a look at no-code-pro-code.com. Regardless on if you join or not, we'll keep doing this amazing podcast here because it's so enjoyable. And we can't wait to get all of your questions, comments, answer them, and talk about great tools like Xano and WeWeb and Nordcraft.

And next time we need to have like the Viking hat show up whenever we got to—

I got to see if Party City is open or if they only have open during Halloween. But I'll take a look at that.

Fantastic. You've seen it. Enjoy again, talk to you today.

Thank you so much. It has been a pleasure talking to you. I also want to thank Tara and Jack who are probably watching right now, even though we kicked them off. Um, thank you so much for, for being here, for sharing your story. We really appreciate that. And thank you for, you know, watching us. Let us know in the comments what you think about this format. Do you want to see more of this? And yeah, thank you so much. We really appreciate that. And give us a like And the thumbs up and comment, share, subscribe, whatever other YouTubers tell you to do, we'll just ask you to do the same thing.

Fantastic. See you in a week. Thank you so much. See you.