How AI Is Changing Developers and What They Build—with Ray Deck

Is software dead? Are engineers going to disappear?

In this episode of Futureproof, Xano CEO Prakash Chandran talks with Ray Deck, founder of State Change and longtime technologist, about how software creation is evolving in the age of AI. Ray explains why the future isn’t about replacing developers or building the right SaaS — but about rethinking how we define both of those terms. Together, they explore the evolving profile of a developer and why software might be ephemeral, but the data and customers behind it are constants.

• AI as amplifier: AI helps shorten the path between concept and creation, expanding the definition of what a developer is, emphasizing the importance of AI literacy and putting a premium on creativity.
• Data Is the foundation: Your most durable advantage isn’t code — it’s the data and customer context that power your products.
• Think big, build small: Big ideas will increasingly be delivered through targeted, temporary tools designed to solve precise problems.

Episode ID: 18126513-how-ai-is-changing-developers-and-what-they-build-with-ray-deck

Subscribe to Futureproof wherever you get your podcasts.

From Xano - The fastest way to create a production-ready backend for any app or agent. Xano unifies AI speed, code control, and visual clarity, so you never trade reliability for velocity. Sign up for free today.

Chapters

• 0:00 Meet Ray Deck
• 4:00 The Last 5% Challenge
• 10:00 Who's a Developer Now?
• 13:30 AI as the New Literacy
• 17:30 How to Think With AI
• 20:00 Data Readiness is AI Readiness
• 25:00 From Apps to Agents
• 33:00 Ephemeral Software, Enduring Impact
• 37:00 Advice for SaaS Leaders
• 40:00 Tools & Techniques
• 43:30 Looking Ahead

Transcript

Ray Deck: 0:00

My my spiky opinion is that SaaS is dead. The idea of SaaS is you have an app. People are going to come use your app. They're going to put their data into your app, and then they're going to get value from having to put data into your app. That's like the classic sort of crud structure. The agentic aspect of what you were just describing is what really changes things up. I don't want to go to your app. I want your app to come to me.

Prakash Chandran: 0:19

Hi, my name is Prakash, CEO of Xano.com. Today I'm joined by Ray Deck. Ray is a veteran technologist and founder of StateChange.ai. With over 25 years in software, he got his start in low code back in the day when it was called rapid application development, later training as a data scientist and leading implementations for organizations from Fortune 500s to NASA. Along the way, he's founded companies, run large consulting projects, and built a reputation for helping others become software creators. At State Change, Ray now focuses on the hardest 5% of building software, the tricky challenges that still remain even in the age of no-code automation and AI. Beyond all that, I'm happy to say that I consider Ray a friend and a mentor and a former mastermind member. So Ray, thank you so much for being here.

Ray Deck: 1:25

Thank you so much, man. I'm so glad you're doing this podcast.

Prakash Chandran: 1:27

Yeah, it's gonna be a lot of fun. Um I thought we'd maybe start the conversation for those that don't know you on a little bit of background, your origin story. So I'll let you go ahead and kick it off.

Ray Deck: 1:37

Uh yeah, sure. So, I mean, I went to school for what was supposed to be political science and then got uh, you know, roped in by this guy named Ed Tufty, who I did not know was gonna be one of the founding fathers of a science we would later call data science, uh term I didn't really get to learn for the first 10 years of my career. Um, but learned pretty quickly coming out that like uh I did some consulting, did some of this and some of that, really got bit by the startup bug, but wanted to go build my own businesses, and then just started building one after another over the course of my career. Um, and the um, you know, built a pretty successful uh uh legal tech company um that focusing on you know large, you know, helping large law firms. Um and then uh really, really around the pandemic, which is around when we met, you know, was trying to figure out what my next act was going to look like um and did uh did a little bit of a of an experiment and um discovered a couple different ways that like, you know, that people were finding value out of technology. Crypto was one, AI was one. That's actually what led me to the NASA thing. Uh, and then uh and also this world of you know, low-code, no code, which sort of ties into something I've always really enjoyed, which is teaching people and growing people and like the helping uh you know people who saw themselves as non-developers and non-technical uh become you know successful and be able to turn their ideas and their the the engines and their mental models into uh into products that they could have in the in the world. And uh I've done some of that with Xano, some of that with some other tools, um, and sort of that's what I've made into a practice that called State Change now, which is a combination of a uh community mentorship service and uh and absolutely and now AI bot, uh, that is about helping people, you know, apply these mental models that I've learned over the course of the last quarter century. I feel old when I say that out loud, uh, to um to so they can, you know, move faster. And sort of our line is it's you know, 80% of this stuff is easy and 10% of it seems hard and 5% of it is impossible. And then we're about cracking that last 5%.

Prakash Chandran: 3:34

Yeah. Um, you know, I think I got the benefit um and the joy of watching you kind of build state change. And um, I think that you have a unique ability to bring like to bear all of your experience as a computer scientist and an engineer and an architect and distill down those learnings in a very like understandable and accessible way to help people get over that 5% hurdle. So maybe let's talk about state change a little bit before we talk more broadly about AI. How many people do you feel like you've served at this point? And are there any common themes around what uh problems people uh tend to solve in that last 5% mile as they're building applications or business value for themselves?

Ray Deck: 4:18

Yeah, sure. So uh I think state change itself as a practice has served maybe 500 different logos at this point, um, you know, teams of people within institutions. Um, most of them probably like, you know, entrepreneurs who are sort of you know transforming themselves, uh, as well as, you know, some, you know, medium-sized and some, you know, larger, uh, larger institutions as well. Um, and the um, and but but the but the nature of the problems they run into have a lot more in common uh than they are, you know, super different. And it really kind of boils down to machines aren't like us. Uh and there are aspects of building software that are very intuitive because that's the part where the machine is more like us. Or part of Barrier to say, this is where we're bending the machine to our will. And a lot of front-end development looks like this, right? Where we're trying to create things that a human being will know how to use, and there are issues of user experience and design that are actually fairly intuitive when you just sort of lay them out for people. Um, like, you know, keeping simplicity, white space, organization, though those kinds of things. But then there's a design for machines. And machines don't think the way that people do. Uh, even though they are thinking things and we are thinking things, uh, the way that humans think or animals think in general is just very different from the way the the machine works and the way that a machine computes, right? Because we we call them computers. They do math of a very specific kind. Uh and the and when someone tries to attack it as if it were just sort of a person doing it, you wind up getting uh getting a little bit uh sideways. So it is when we think about that in terms of like front end versus back end, it's often about the back end. It's often about like how does data get managed? What you know really drives memory? How does a computer work that it can go faster and what causes it to go slower? And you know, a lot of these things boil down to you know mental models we built along the way that we also have inside state change. But that but being able to deal with the alien when it's acting in its most alien way. And AI has sort of you know redoubled that because uh, as much as natural language processing looks normal, it actually is a very weird beast underneath. Uh, and so getting our arms around that becomes an interesting source of uh of challenges, you know, as well. So I find that's usually where the hardest 5% is, is getting people to think more intuitively as the machine does. Although I will say that over the course of the, I guess about three years I've been doing state change, the um the number of people who come in and sort of intuitively understand more about how the machine works has really increased as I'm seeing more people who are coming in who have, you know, maybe they did a bit of coding, uh, they learned how to do JavaScript or had a Java course or something back maybe when they were in school 15 years ago. And so they can have an intuitive sense for what could be going on here, but not quite how to articulate it to the machine in a way that's gonna allow them to make the kind of progress they want to.

Prakash Chandran: 7:05

Yeah, this actually leads me to one of the questions that I wanted to ask you, um, which is well, let me frame it. I, you know, I think that especially with the introduction of these uh vibe coding tools, a la bolt, lovable, et cetera, we've seen like the widening of the aperture for uh these new types of builders that are able to build, prototype, and express their ideas. Uh, kind of amongst them, we kind of have people that are trying to tackle that last 5%, a lot of the people that come to statechange.ai. In your mind, like what is the definition of like a developer these days? Like, how would you how do you think about uh the new type of developer, the people that are coming to State Change? And uh how do you see that changing over the course of the next couple of years?

Ray Deck: 7:53

Yeah. Uh so there were there was this line, you know, when the iPhone first came out, um, or or shortly after when the App Store first became a thing, the iPhone. There's an app for that. And now pe and I think it's worth noting that the iPhone's been around long enough that if it were born the day it was introduced, it'd be old enough to drive today. Yeah. People who have had it for their entire professional careers, their idea of software has gone from like my grandfather's idea of software. He actually worked with like these these big mainframes and insurance companies. Uh, and the um, and like so his idea was like these machines that would fill like the floor, right, of a big skyscraper building. Um, you know, my my my my dad's idea of computers, you know, he was actually a little bit of a pioneer like first early on to like using like the original, you know, Apple Macintosh and even the one of the first versions of the IBM PC that came out. And his idea was like it's a relatively heavy piece of metal, right? With an ugly screen or whatever that's sort of sitting in front of you that you are that that you're then trying to bend yourself to make work and communicate through this very loud, clackety keyboard. But the people who have been working with technology for the last 15 years have had it in their pockets and it responds to touch. And more recently, in the course of the last decade, it responds to voice. It is something that they have command over. So now instead of a, oh, it's that big thing that I couldn't possibly hope to understand, I might only ever hope to wrangle it a bit. They say, no, not only there is an app for that, but there should be an app for that. And that's a small thing. I should be able to make the app for that. That should exist and be able to serve me rather than it being the giant colossus that I need to serve. And that attitude towards technology, what Malcolm Gladwell calls entitlement, that I think is the big difference in what I've been seeing, even just increasing over the course of the last few years with the way people have been uh uh using and approaching technology. They come in with this idea that there should be software that does this, uh, and that they and that they themselves are empowered to do it and they should be able to make it happen, and there should be tools that they can have command over that will allow them to do these things as well. That change in attitude is, I think, what defines a developer. A developer is someone who can make software to do their will. And that um, and and then that describes a much broader set of people, partially because of what's been in the air, the technology that the you know, the the the technology over the course of the last you know quarter century of this um, you know, the start of this millennium. The uh the way it's become so much cheaper, so much smaller, it can be fitting in our pockets. The way that um the fact they've had some training with this, usually in school, if not professionally, the fact they're using these things every day, both at work and personally. Uh, and then they say, yeah, sure, I can do that too. And that's um, and and that idea of um again, you know, using that word entitlement, I think that is what allows many more people to become developers. And then the question is, you know, who's gonna be able to help them be able to uh to realize and execute on that vision.

Prakash Chandran: 11:04

Um, and then the the person or the persons that would help those that are entitled, like this new type of developer that wants to kind of will that app into existence. That's kind of at least what I think about as like software engineers, people who have had all of the education and literacy to be able to kind of take the initial spark of creation and then get it over the hump. Would you sit, do you see the world in that same way?

Ray Deck: 11:29

Well, I think software engineers, as we might have understood them before, people who were trained to do this and says software engineering is my job, right? And is it was how I define myself. And like, you know, the what is going to be used for, that's someone else. I but my specialization is I know how to make software, right? And the the gag is that the new developer says, I understand this business process, I understand this domain, right? And I can bridge that understanding to software, maybe mediating it. Previously, you have said I have to mediate this through the software engineer who will then be able to take my ideas and express it in the form of software. And now it's I can, I can more directly express it as software, you know, myself, maybe through something as simple as doing it through an Excel formula, or something as complicated as, you know, creating a, you know, big app or automation or AI agent or whatever it is, right? Uh, but that they don't that it I mean, this this is a term that showed up a lot in like the dot-com boom, right? The idea of disintermediation. We don't need the middle person anymore to create the software. We can instead say, I have an idea. I can then express that idea as software and have that come directly. So the, the, the, you know, you I think you're right to say that like, you know, if we were talking about this a decade ago, we would have said software engineers are the people who are doing this. They were trained to do this and they learned how to do this. They could put in the time commitment to do this. But now the uh the ex ante training required is less and the time required is less, which means software engineering goes from being a, you know, a definition of a professional to a thing that I'm doing on Tuesday.

Prakash Chandran: 13:07

I think that makes a lot of sense. Um, I want to touch on something that you you said earlier, which is around kind of, you know, the people increasingly that you speak with have a more intuitive sense of how to kind of wrangle uh the machine. And this kind of touches on like the general topic of like AI literacy and um what you need to know now in order to be relevant and to, if entitled, be able to use these machines in the most productive way. Let's talk a little bit about your thoughts on that.

Ray Deck: 13:41

Yeah, sure. Uh the the I I I I think I've I've made the analogy before of AI to literacy, right? Literacy really changed the way that you know people could uh create value, you know, in society and the the way that ideas could create value in society. Because previously, an idea was only one that you could share with somebody, and if they didn't remember it, the idea died, right? Then you have the ability to write it down, the ability to read, be able to build build on other people's ideas. And I sort of see AI providing that same kind of value into society, particularly like the the uh natural language uh you know flavor of it. Um I'm teaching my children, you know, I homeschool my children, um, and we're working on some fairly, you know, complicated ideas, sometimes ideas I don't totally get. And, you know, I would say five years ago, in order to understand this, I would like go to Google and I was okay, where's a good article or something on the subject? I'm dependent on like who might have read uh uh put something together and like I have to read a bunch of pieces in order to sort of understand the topic. Now I can go to AI. AI will take those, whatever those resources are, distill them into the aspect of my question, trying to help me understand just the thing that I need to know at that moment. And my access to ideas as a result is really much more refined. And I can learn a lot more and work with a much bigger set of ideas in a much smaller space of time. And that time thing, like with sort of using my Tuesday joke from before, is the reason why I no longer need the is the difference between needing to be a lifelong student of a discipline and being able to incorporate that discipline into whatever it is that you're doing today, right? And that ability to remix ideas and disciplines is what I think is really um, you know, transformative here, right? In terms of being able to apply AI to create a whole lot more value. Like one thing they they were just researching, like, you know, debt equity ratios and how it relates to like, you know, the largest companies in the world. And like the the the insight they were able to get very rapidly was just astounding to me. One, of course, they're spark kids. Um, but because they could, you know, make use of these resources, they had access to a lot of data. But rather than having to sift through all that data themselves, they could use the leverage of the AI on top of it, and then use the leverage of their own insight on top of that to come up with like really remarkable understanding in a very short period of time, which means they can combine it with other kinds of understanding along the way, which allows for much bigger intellectual unlock. And that's that that's sort of the the uh the the literacy aspect, I think, of that, that gets me really excited.

Prakash Chandran: 16:18

Yeah, that's fascinating. But a part of that is understanding like how you can properly interrogate um the AI to kind of narrow in on exactly what you're looking for. And this kind of speaks to whether it be software development or just kind of more broadly, a paradigm or framework around like how you can work with the machine to get the results that you need out of it. Yes. I know you have a lot of frameworks around how you think about this. Maybe there's one that you might share for any like technical builders or people who are really starting on the frontier of leveraging some of these tools to work with them.

Ray Deck: 16:51

Yeah. Uh, you know, the I the um the the the the framework we keep on coming back to most often in state change, and especially in the context of software development, are you will build this again. The idea that you are going to operate iteratively and you're trying to get to the next answer. I run into so many people, both in the world of software, but also just like, and this happens like just in the world of ideas. You're trying to get to the final answer. How can I rifle shot this right now and get my final answer? Um, but the thing is, you're not ready for the final answer. You have ideas that need to get into your head in order to get from here to there. You're not even ready to ask the question yet. And so the the what I find is that being willing to ask more questions that then get you that one step further. And that's another question gets you another step further, is a model that uh I think when people are are afraid it's gonna take a long time to do all this, they become much more reticent to ask those questions. And this is where like AI and the speed aspect really changes that, because now I can afford to ask more stepwise questions and then get to the next place. Because that wasn't the way research worked previously, people are more reticent to do that. They're like, oh, that thing's too far away. There's no way I can get that. I want to go make that somebody else's problem. But you can make it your problem and get to the get to that like at the end of an hour. It is amazing what you can do. I think the the most important skill is going to be the willingness uh to be the the the the willingness and certain amount of technique of asking semi-open questions, right? Questions that like are in your domain, right? You're not allowing things to go completely flop open, but that are uh they're focusing you forward, but that are keeping open the the weird possibilities, you know, that are in front of you. And that's a balancing act. And that's something that I don't think that many people are good at right now, uh, but is definitely one of the things we teach in like the mental models that we do in SageChain, certainly something I'm trying to develop because something provides even more value today than did yesterday. I would say it's probably true that like you for an intellectually curious person should be asking questions in general. Yeah. But I think the technique of asking those questions is shifting as we use AI. The uh the the the speed at which you can ask questions goes way up. The types of questions that's able to answer well shifts a bit, as you might imagine. That's also somewhat sensitive to like how these models work and what have you, uh especially like in the multimodal aspects of it. Um, but the um, but the value you can get from just being able to go all the way down that uh path is just extraordinary. Like another line I have is like, you know, data is like the new oil, right? And that AI is a refinery for that oil.

Prakash Chandran: 19:35

Yeah.

Ray Deck: 19:36

And that you you now get access to more of the value from that oil, that data, that knowledge that your company has, that you have, that like the world has, and that you can put together, you can put to put it to work for you in a much more refined way in a much shorter period of time because of this technology uh that is, you know, growing every minute while we're having this conversation.

Prakash Chandran: 19:58

So you mentioned something there, like data uh kind of as the new oil. And you know, we were at the Gartner conference a couple of weeks ago, and uh there's a lot of talk around AI readiness and how that relates to data readiness. Um, you know, I think that when they talk about that, oftentimes they're talking about data quality and there's a cultural component of that. I'm curious as to your thoughts around, you know, the broader term of AI readiness, the importance of data as it kind of relates to kind of getting organizations to leverage AI in the right way, to get the answers that they need. Um, yeah, talk about your uh your thoughts around data or AI readiness and data.

Ray Deck: 20:39

Yeah. So um, you know, this is not our first turn of the wheel for thinking about data and these larger organizations. Big data was a topic of the day a decade ago. Right. Uh that had a lot to do with the fact that these big companies, like Amazon's a good example of this, right? It has tons of data about how people engage, go through their stores, what they buy, um, and they try to use it in various, you know, semi-competent ways to be helping people buy more, find the stuff they're really, you know, looking for, um, et cetera. Uh, and the the trick is that so much data is terrible, right? That the the the it is it is um you know, it's it's exhaust data, and sure we have access to it, but like what of it really, really matters in here and what is driving um value. So a lot of the the data I think is is most exciting is data that was not exciting in the big data era, but like is for what we now call maybe you know uh unstructured or semi-structured data, you know, like text, things that reflect knowledge. Um and the and and I think one of the when when I think about like AI readiness for these larger organizations, it's where they suddenly realize, oh my goodness, there's all this stuff that we're not capturing that is actually valuable, right? That we're capturing the wrong things and we're keeping the wrong things. Um, I'm imagining uh, you know, that you, like many sales professionals, are doing things like recording your meetings, right? And then you're running through them through analysis. Now, the analysis is fine. That's the refinery part, that's the AI part, but the way you get value from that is by recording it in the first place. And those recordings would have been verboten in most of these companies coming into this. Like, oh no, how do we get value from this? How do we get value from all of these interactions of our knowledge workers that we've been doing all the way through here? Previously, like big data was really about what were our customers doing, uh, maybe what were our line employees doing, things that could be turned into numbers relatively easily, because that that kind of structured data fit easily into databases and also could be run through more traditional statistical techniques. But one of the neat things about AI, something that's sort of you know discontinuous about it, is that this unstructured data, images, video, um, you know, the words people have written, or probably the main media that we usually think about, um, the sound video and text, um, those those create a lot of value and allow us to figure out, okay, what else was this customer doing? And when were they doing? How can we associate that with time? Uh and the uh and what was what were what were we doing? Can we clone our people? Can we take some of our people are really in the business of like just restating corporate policy? Do we need people to do that? Can we have a machine that is able to just restate that policy and be able to help the client, you know, work through our policy to figure out how to be getting the most value that they, that they, that they can? All of this requires capturing more of that unstructured data. So when I go into a lot of these companies, what I'm seeing is that they have some of this data and then a good fraction of it, they just don't have. And the most important thing from an AI readiness point of view is being able to catch the data that when they last did this a decade ago, they weren't catching, right? Because all of a sudden the relative value has shifted between these things. So the and and like the it's it's a little bit like, hey, if you got oil in your field, you got to put in, you got to put in the pumps, right? We need to turn it from just latent value into actually captured value. Because if you haven't pumped out the crude, it won't do any good at the refinery.

Prakash Chandran: 24:01

Yeah. So the importance, there's like where data may have been uh not as important. You don't, you didn't have leverage over it now becomes exceedingly important. So if there's one takeaway, it's get the data, you know, uh start attaching as much data as you can because the more context that you have, the more you can leverage the machine to help you.

Ray Deck: 24:20

Right. I I I think that that that's the key. For most folks, I have found they focus a lot on how can we refine it better, how can we have better models, how can we have better prompts, whatever. But it's all a multiplier, a multiplier on the data. And when they're not collecting it and they're just letting it fritter away, which many are, uh, that is, I think, the uh, the the key missing part. That's where like a lot of you know opportunity is for, you know, uh, you know, the robots to be able to come in and be able to make sure the stuff doesn't um uh doesn't float away because traditional data warehouses haven't been solving this problem so far. Not because they're structurally incorrect, but because they've been saving the wrong stuff.

Prakash Chandran: 24:57

Fascinating. Um, I want to talk about basically maybe accessing the data and maybe through the lens of um just kind of something that I think we've seen just shift dramatically. So obviously, search behavior has changed quite a bit. We've gone from potentially using something like Google to now just asking Chat GPT or uh, you know, perplexity or something, give me the answer. And it's amazing across the industry how much top of funnel has dropped, intent has increased, but wow, how that has shifted. And we're it's not really far away, and you can already see this happening. It's not just asking, hey, give me the answer, but do this thing for me, right? In this world where we are slowly moving to like when we talk about agencera, I'm wondering what you think about that modality, like the new way people are kind of going to consume the data and consume the information. And what does agentic mean to you in that regard?

Ray Deck: 25:55

Yeah. Um I the the my um my my spiky opinion is that SaaS is dead. Um the um because the the idea of SaaS is you have an app, people are gonna come use your app, they're gonna put their data into your app, and then they're going to get value from having put data into your app. That's like the classic sort of crud structure, right? Of of one of these SaaS applications. Um the agentic aspect of what you were just describing is what really changes things up. So I'm not, I don't want to go to your app. I want your app to come to me. I don't want the the the environment in which I work with software would have been uh, you know, 20 years ago, it's your desktop, it's your Windows desktop, right? Because Mac is sort of nowhere 20 years ago. Um in uh 10 years ago, it's your browser. Right. And you know, the way you access software is by opening up the browser and going to a certain URL and then getting to there. And that's one of the reasons why search engine marketing made so much more sense because people are looking for websites and now you're able to find the websites, and that's what SEO and that's what you know, PageRank stuff and whatever all that stuff was about. But now I don't want to use any of that. Like I want to just get something done. And how do I get that thing done? I get it done by working with my AI, right? I can go just talk with Claude. Hey, Claude, go take care of this problem for me. Or Chat GPT, or Dia, which is actually uh an AI that's built into a browser, is what I'm talking to you on right now. Um, or Perplexity, which has an offer called Comet, which is its own built-in browser to be able to integrate these things so that you can be doing the work. And that idea of of bringing the workdoer to the professional or to the consumer is, I think, the the the process that we're still in right now. Uh and that is the and and that, and if that's gonna be where work gets done, then that's probably also where commerce gets done. Uh there's a there's there's a great book called Uh When Machines Become Customers. It was a really pressure book that was written a couple of years ago about this subject, about how AIs would be able to start buying things on behalf of people, you know, to to to do work for them, right? Uh and the um and and and you can just have the AIs do these things for you. And the question like, how do we give them permission? That becomes a problem to solve, right? How do we give them a wallet to work with if they want them to do commerce on our behalf? How do we give them access to the tools that we need? And the uh and the standard right now that sort of is most exciting in this regard that allows these AIs to go talk to these various services is called model context protocol, MCP. And MCP is, at least right now, I think the the most exciting thing in agentic AI because it exposes tools and resources and prompts. Those are the three big things in MCP, to um to empower an AI that's connected to a person to do things on their behalf. So that my chat GPT isn't your chat GPT. My chat GPT is chat GPT plus access to my notion, plus access to the books I read, plus access to the My Gmail or other things I care about. And it's able to both bring me knowledge that I care about, you know, either that is private to me, contextual to my company or from the world, and be able to do things on my behalf that uh through through my being able to issue commands. And now that becomes the way that I'm interacting with software. Instead of from an app on my desktop or a site on my browser, it becomes a tool that I can be interacting with, you know, through the agentic interface, which could either be in the form of straight text or because of a newer technology called MCP UI, which is a really exciting development from 2025, um, being able to start to bring, you know, micro apps that are then being uh uh uh interleaved in with the um with the experience of working with the AI and the um and then like software that's just made on demand to help me solve this problem right now.

Prakash Chandran: 29:49

Mm-hmm. Talk to me in this world that we're moving uh towards and kind of like the foundation that you laid out, what is the job of the or what does the front end look like? Like, Is it a mobile application? Is it a web application? Is it a uh is it something else? Is it just that we're gonna do everything through cloud? Like how do you see the front end evolving these days?

Ray Deck: 30:11

U well, uh that MCP UI thing that I alluded to is actually what I think is the most exciting development in in in user interface. Because now instead uh previously it was a, hey, you either have this flow of text, right, that's coming through the application, uh, or you are dealing with like a traditional app. Or sometimes you have like the co-pilot modality where like you've got the you know, the sidebar of like, you know, I've got my chat going on over here, and then maybe the other 75% of the screen is working with the app. Like that's uh pretty common since like the days of like GitHub Copilot working with Visual Studio Code, and you see that in some other apps now too. Um the but the but but I think that the way we're gonna see it is the is the reverse, uh, where the the primary mode is going to be talking and working with the AI, and the AI is able to bring micro apps in context, you know, to you. So we'd have like a, you know, be able to have a window that maybe could expand up to be able to have the podcast going, uh, be able to have information that's being delivered, you know, over the top, that is, you know, as it has identified opportunities for me, kind of clouly style, uh, being able to, but the but the but but but the the main controller for all this is an AI who's working for me. Then that's important that's got to be working for me because you don't want to be giving up control. But the um, but like I'm imagining I gotta tell you, like when I'm using my computer right now, and I'm a fairly sophisticated user of these machines, um, most of my tabs that are open in DIA are to the chat, not to people's sites. And that's because that's where a lot of the value is. And I think what we're gonna start seeing is through the use of this kind of micro app, you know, integration, uh, which will initially be decried as that's not a real app and that's not a real front end, which was the same thing as what people said about web apps 20 years ago. Uh, I think we're going to see the same sort of you know evolution of like instead of saying, oh, well, the browser is just an app that's inside the desktop, why would you want to have apps inside that? People will be saying the same thing for a while. Traditionalists will be saying the same thing about like things are integrated into the AI. But I really think that's sort of where the where the modalities are shifting. And we'll see that both on, you know, on the desktop, probably with a more textual interface, because desktops tend to have more of that. And then on the phone, uh, we'll see it probably with some text, but I'm expecting voice to become a very important part of that because key typing is a really hard thing to do on mobile, uh, as well as uh making use of the camera. So I'm expecting really the um, you know, the the the microphone and the camera to become uh the most important ways to be interacting with a phone um, you know, heading into the next couple of years.

Prakash Chandran: 32:41

You know, we've um we kind of touched on this briefly, and I'd love for you to dig in a little bit more um around kind of this uh potential around like ephemeral software. Uh in this world, maybe it's enabled by something like an MCP UI, where it's really use case driven um and it serves its purpose. Um, how does the world of software look when you have basically these one-shots of like, hey, we're gonna give you something and then maybe it's not needed anymore? Talk a little bit more about um that because I thought it was a really interesting take from you.

Ray Deck: 33:16

Yeah. So the the why do we use software? We use software to get jobs done, right? And like a lot of the time we're using some piece of software that was written with, you know, a million other people in mind uh in order to do the job that we care about. And then we are the friction points we're dealing with are trying to get through the, you know, the weirdnesses of like whatever Microsoft decided was going to be good for me and be good for, you know, 500,000 other people as well. Right. And that had a lot to do with like the cost of building software. And you know, software used to be something like we would talk about like software engineers being a being a career path, right? And by the way, that career path is not going away. But like, you know, for uh since almost all software had to be built by professional software engineers, uh, that became expensive to build. It would take months or years to build a good piece of software. And then once you set then once you built it, you have to sell a lot of it in order to pay for that back. So that means it's got to be good for a lot of people. And that means instead of being really excellent for you, it's got to be good for you, me, and the five, 498,000, 499,000 other people. So how do we um the the the opportunity is that uh instead of having to be downstream of mass-produced software, we can be downstream of bespoke software. The most expensive, the highest end suits in the world are made on Seville Row. And Seville Row makes each suit custom. It's called bespoke clothing manufacture. And the, and if you uh if you go there, you will get one that fits you perfectly and will be useful for just you. And then you'll probably want to keep it for a while. But that idea of like having software that's just for you or just for a smaller audience is has been sort of a holy grail for software for a while, right? That's what like the rat uh the um rapid application development stuff was really about. It's like driving down that cost of software so I can solve it for a smaller audience that's useful for a shorter period of time. And then we sort of see that happen again with like low-code software is like again trying to drive down the cost, be useful for a smaller set of people for a shorter period of time, basically shorter depreciation schedule is what accountants would say. And then low-code note, and then like, you know, the no-code movement of like 2022, this kind of thing I associate with Xano again, shrinking that down, shrinking that down. And then AI, which has been shrinking that down in like five waves over the course of the last two and a half years, has been just madhouse how fast it's been going. And it keeps on shrinking down, shrinking down. So that now you can be, you know, creating meaningful software in an hour, which might have previously, which in the previous modality might have taken a couple of months. Now, that's not software that's going to be good for everybody. And it's my main software that's gonna be good for you for a really long period of time, but it's based on your superior understanding of this is the job that I need to do, and I need a machine to help me do it. And then I can execute that, and it's good for that job. And at that point, I don't care what's good for after that because we drove down the cost of creation by so much.

Prakash Chandran: 36:17

Yeah.

Ray Deck: 36:18

And the difference between disposable versus long-term. Yeah, sorry, go ahead.

Prakash Chandran: 36:22

No, it's it's interesting in that world of kind of Malcolm Gladwell's entitlement. It's like, this should exist. I'm gonna make it. Okay, now I'm gonna throw it away now that I'm done with it. Um, that's where we're moving toward. No good.

Ray Deck: 36:35

When software is when software is something you create on Tuesday, then you're gonna use it on Wednesday. And on Thursday, some other problem comes up, right? Yeah. Your your your view of it, the the same thing that creates that level of entitlement also creates this ephemerality to it. Because now it's not about I need to go make a piece of software. Like I I've still run into some people who say that my big who who describe themselves as their big ambition is to go make a SaaS because that's what they believe software entrepreneurship looks like. But when I look at people who are more sophisticated about this stuff, those are people who are looking to get jobs done, who are thinking in terms of services and the way value gets created. And then they're asking, how can we apply machines to be doing this? And when you start thinking about it from those terms, a software-enabled business as opposed to a software business, you start to free yourself to think differently about like uh how the software is going to work, but also what software gets created, how much software gets created. And then I think we get to really unlock a lot of economic opportunity.

Prakash Chandran: 37:36

So, you know, there's going to be tactical founders, um, application development leaders listening to this. They might be building SaaS or they have SaaS, right? Like, let's let's pretend that there's a fictitious SaaS company that finds the best restaurants in your area. You're able to kind of make reservations. They have a mobile and web application. They are listening to this and they're asking themselves, what should I be doing right now? I've got a web, a mobile application, I've got a back end. What should I be thinking about, right?

Ray Deck: 38:08

You should be thinking about your customers and your data, uh, and be thinking more creatively about how to be getting leverage from both of those because those two things are assets. And the the software that's been created so far, don't get me wrong, it'll probably still continue to create value for a while. But like, do we think the next turn of the wheel looks like the last one? Is it you're evolving the software that you have today to be useful in like the next couple to be useful more than say, I don't know, 12 months from now? Um, or is it like what you're really doing is creating and nurturing, right, this asset you have of your brand in the market, because you've been providing a good service to people, of your customers who have a certain amount of trust and that are also providing you with data, as well as like the vendor relationships you probably have, like the actual restaurateurs or what have you, right? And then those relationships, that knowledge you have, the secrets you have, the the um the reputation that you have in the market, the value of the brand, and being willing to think more creatively about how you will then uh uh you know filter those through software to deliver those assets to your customers because your software was never the asset. And that's I think a key thing about most SaaS. That's one of the reasons why I sort of am more skeptical about SaaS overall, is when I look underneath at SaaS businesses, it's like stone soup. I don't know if you ever heard that story, but like yeah, the um the the idea is that what it was a lure for the customer to get their data in. The data was the asset, the relationship with the customer is the asset, the value they get out of it is the asset, but the SaaS is usually just a mechanism to go to connect those things. Got it. And the uh and and and and uh when, and that's one of the reasons why Micro SAS has been so successful, because it's not like technical excellence means that one SaaS rules them all. It's going to be being able to conform that SaaS to the particular needs of a smaller and smaller segment of people. That means it does a better job for those people. If you're able to do, and the the promise of AI in a more ephemeral form of software is that it's able to conform around the person in a way that is um, you know, different by an order of magnitude from what came before, that implies that your real value is going to be in the data and relationship assets as opposed to in the software code beasts that you have today.

Prakash Chandran: 40:29

Love it, Ray. Could talk to you all day. Um, but in in uh moving to a close, I wanted to ask you a couple kind of more lightning uh round questions. Um talk a little bit about what's in your like AI toolbox. What are the tools that you're using that you're excited about that are potentially underrated?

Ray Deck: 40:46

The tool I am most excited about AI-wise is Suno. Um Suno is a software for being able to make songs with AI. Their models are extraordinary. But more importantly, I can take like a meeting that I just did, maybe a conversation like this one. I can have Claude, I can have a have a have a pipeline where Claude is able to, you know, take that uh meeting and turn it into uh identify the key ideas out of it. Take those key ideas, turn them into lyrics, have those lyrics turn into an earworm in Suno, and I will then play those songs in the background. My my kids quote those songs. Um, and and it's feeding the ideas back into my brain. The opportunity to be using different modalities, right, to get the the this the this universe of ideas that surround us to feed them back in so that we can be making better use of them is I think the the under-index part. It's not about developing software, it's about developing ourselves. And that's why uh, and and so like I I know the the Suno one is a little bit funny when I know I've sent you a couple Suno songs before, um, but like that that kind of tech that tech has done, you know, more to sort of you know infuse these things into my brain than just about any of these other uh chat-based systems.

Prakash Chandran: 42:01

I love it. I was just gonna say I've been on the receiving end of some of those songs. It's it's pretty incredible. Um, did not think you were gonna say that, and I love that you did. Um talk a little bit about an underrated like practice that developers should be honing or working on right now.

Ray Deck: 42:23

Um the White Room, uh, I think is the one that it seems it seems like the hardest thing for me to get people to do, which is to, you know, you're working on a hard problem. Don't try to edit your software to do it. Clean room, or if you're if you're if you're in a chat, clean chat, work just this one problem. Um it is the it is something that like professional software engineers get taught in school and then routinely forget. I it I I find myself saying this to people at at whatever level of education they're in, because especially when you're getting most frustrated, it's most tempting to say, that didn't work, try one more time, right? That that's the kind of thing we've done with AIs before. It's something we do with ourselves all the time. But the willingness to say, nah, you know what, this is the one problem, let's isolate it, let's work this problem, and then when we have better understanding of it, let's bring it back in is a hard thing to do from an ego point of view because it's saying that you weren't almost there. Right. But then if you are willing to do it though, you will make a lot more progress, you understand it much more quickly, and you can tack a big complicated problem as a series of small and much simpler ones. And that tends to uh, you know, crumble much, much faster uh and provide a whole lot more value for you. So I think as a mental model and as a practice, the can I take this to the White Room would be the question I actually certainly find myself asking a lot in like state change office hours. And that I would encourage anybody listening here at home to ask, should they be doing that today?

Prakash Chandran: 43:47

Yeah. Um, that's a great one. And then finally, you know, we've talked about a couple of these in the process of our conversation, but any bold predictions um or just kind of a glimpse of what you think the world is going to look like with AI in the next three to five years. And I know it's hard to think that far out just because things are moving so fast. Um, but anything that you might share around uh what you think things will look like then.

Ray Deck: 44:15

In 2012, anybody who knew anything would tell you there's no future in AI. Uh in 2017, anybody and and then uh that was the year, uh I'm pretty sure that was the year the Alexnet came out and uh transformed and just sort of lit the firecracker for the world of of computer vision uh and uh and that little machine learning. In 2017, anybody who was anybody would have told you that like AI is good for computer vision, but has no play in um in in in natural language. We were like working with things like called RNNs. It was like it was really a kind of a go-nowhere segment. It was in that year that um attention is all you need, which was the paper that first defined the technology behind what we now call LMs, uh, was published. Uh and people have told you, like, you know, there, there's you know, people in 2022 would have said, uh, it is so obvious that like the the you know, the use of just sort of creating like this this spacer foam and and chilly stuff uh for you know any of this natural language stuff that is being made out of GPTs. And that, of course, is the year the Chat GPT came out and caused us to really rethink just by uh changing up the packaging and some improvements to the model. Um the world will keep changing. I don't know what's going to be coming down the road, but something, but more things are gonna come, you know, down the road. And it's not gonna be about little stuff, it's gonna be about big stuff. I do think that there's going to be an interesting play for energy uh when I think about the economic implications, because we now have much more demand for compute, and compute requires energy in a way that like SaaS that were built in 2017 did not require energy because they were just about storage, right? Rather than being about, you know, compute. And the uh the thing that caught that that drives the biggest caloric intake in our bodies uh is this brain that we've got up here. And machines are the same way. When they're thinking, they run hot. You put your hand on your laptop, when it's thinking hard, you can tell that's running hot. And that's that's the reason why. And that the the uh a demand for energy is going to be changing, I think, significantly for that. Uh, and and will probably have a significant impact on like the the the way labor spreads uh as well and the kinds of jobs people do. So those are the kinds of things I would look for in terms of like the way I expect you know markets to change and the way that like the next generation of people are even more native, not even to the world of software, the way they were from the previous 20 years, but they're going to be native to you know AI and what Andre Karpathy calls software 3.0, the natural language-driven software, as we head into the next few years. And that's leaving aside like big weird possibilities like quantum computing, et cetera, that could, you know, really change up the way that like, you know, compute is allowed to work. Um, but um, but anyway, though those none of those were like super huge you know predictions, but those are things that I'm looking for in terms of how AI will affect economics in the course of the next, you know, 36, 60 months.

Prakash Chandran: 47:00

Ray, as always, fascinating conversation. I really appreciate your time and your wisdom. Learned a lot, took copious notes as you were talking. Um, thank you so much, my friend. Absolute pleasure, sir.