← The Digital Executive
The Digital Executive
The Future of Software: AI, Composability, and Breaking the Code Silos with Co-Founder Jonathan Saring | Ep 1041
00:22:24
In this episode of The Digital Executive Podcast, host Brian Thomas speaks with Jonathan Saring, co-founder of Bit, a platform transforming how teams build software through composability and AI.
Loading summary
Transcript15 lines
- [00:00]
A
Foreign welcome to Coruscant Technologies, home of the Digital Executive Podcast. Welcome to the Digital Executive. Today's guest is Jonathan Sering. Jonathan Sering is the co founder of bit, a cutting edge platform that redefines software development through composability and AI. With a commitment to building the future of enterprise technology, Jonathan has successfully led BIT to become a vital resource for over 250,000 developers and countless Fortune 100 brands. His role as an advisory board member for multiple startups in the R and D space, along with his experience as an angel investor, positions him as a thought leader in the tech community. Well, good afternoon, Jonathan. Welcome to the show.
- [00:48]
B
Thank you, Brian. And thank you for having me here today.
- [00:51]
A
Absolutely. I appreciate it, my friend. We're doing a podcast internationally today. I'm in Kansas City in the United States and you are in Tel Aviv, Israel, which is always fun. Jonathan, I'm going to jump into your first question. BIT has become a go to platform for over 250,000 developers. What gap in the software development world did you see that inspired you to build this?
- [01:15]
B
So I think that the most basic reason that we set out on this journey was that software development today is just plain hard. It's complex and it gets worse as you scale it. And as more technologies emerge and the need of the product that people build for their customers are getting more and more complex, and the speed of delivery required today from organizations, even small organizations, let alone enterprises, becomes very challenging in the way that we develop software today. And today, many teams need to deliver more applications, more user experience, more features. And they need to do it faster, they need to do it more efficiently, they need to cut down time to market. And at the same time, they have to ensure that the quality of the software that they're creating is of the highest standard and that the experience is consistent across every application, feature or touch point that they build for users. And that is incredibly hard today. And the reason that it is incredibly hard is that the tools and the infrastructures that we use were built for projects. If you think about it, everything that we use today to build software, the basic foundations, the way that we build software is all about projects. If you look at Git or CICD products and so on, when an organization, for example, needs to build, you know, hundreds or thousands of features and compose them into a huge number of applications and experiences, then trying to build them and deliver them in projects and maintain them in projects becomes closer to impossible as you scale. And that pain, that underlying problem or gap is becoming painful in more and more areas of software development and eventually impacts the end experience of users. But composability solves all of. So the basic idea was that instead of building, you know, these software silos, which in turn build organizational silos, we can lean into composability and build software in components and build modular software, build software where you can build features, you can build applications, you can build complex backend functionalities, and you can turn them into building blocks, into pieces of Lego for the organization or for everyone you want to share it with, and you can use them to build anything you want. That was not possible when we started this journey. And I think that the pain is so great that even if you are, you know, just encountering a very specific instance of it, for example, we're trying to make the design consist the UI consistent across the organization, or we're trying to improve the speed of delivery of a new feature. Or for example, how do we make sure that we can create a feature that all our applications will use and we can easily, you know, update everyone with changes. And those kind of problems led more and more people to seek solutions that will allow them to build differently. And building differently means building composable. So our goal was to build the tools that will let, and I'm talking pre AI era developers build and organizations and teams build composable software together. And basically that was the major change, that was the major shift. And I think that the market was the industry, the people building software were actually not only ready for it, they already begun this process. If you look at organizations today, you see a lot of APIs, for example, which are the interface of composable design and are by definition reusable and something that many different entities in the organization can work with. And you see libraries and packages that now consist, according to the latest report, 60 to 70% of the average organizational code base. You see microservices in the back end, you see components in the front end. So the technology itself was already leaning into composability. There was just no tooling, no platform infrastructure that will allow this form of software development. And I think this is where Bit came in, because we were the ones that said, okay guys, it's okay, the pain that you're feeling are real. And now let's try to think differently and solve them by not creating them in the first place. Let's build with Lego. That was the main thing. We see a variety of use cases for companies to start with. And now when AI comes into the game, I think that the combination of AI and composability is an exponential game change for the way that we build software.
- [06:14]
A
Thank you so much. I appreciate that. Jonathan, I love how you talked about this idea of being more agile, using modular type applications with composable design to really scale, be more efficient, still have the quality, and just be faster at that development. So I appreciate that. Jonathan, jumping into your next question here, if you could briefly share, you've been vocal about composability being the future of software. Can you break down that composable software really means and why it matters so much right now?
- [06:45]
B
Yes, thank you, Brian. That's a great question. So why right now? I'll give you a practical example. I'm a big fan. I've heard that you, you've been working on technology infrastructure for the health industry before, am I right?
- [06:57]
A
Correct.
- [06:58]
B
Correct. All right, so I remember that one of the first today on the platform you can find Dozens of Fortune 100 companies from across all industries. We can check the logos on our website. You'll see names like AT&T and Red Bull and Moody's and many others that are pursuing the composable transformation. Gartner recently stated in one of the latest reports that organizations that will adopt composability and lean into it will outpace competition by 80% in the next five years. That's already happening. And as we mentioned, the technology being built is already using APIs, abbreviation packages. And if I go back to the healthcare example, I remember one of the first enterprises to onboard the composable development platform and use it practically and instantly after onboarding was one of Canada's largest healthcare organizations. And there was a very smart architect that said, okay. He was asked by leadership to double the number of applications that the organization is going to deliver in that particular year. Mainly due to the COVID crisis, which transformed really accelerated digital transformation for healthcare organizations because they had to provide healthcare services through applications and digital interfaces. And so they had to double the number of applications given the same resources. And they had to cut down the price or the cost of developing each new application and had to make sure that everything is consistent and that everything is standardized and of the highest quality. And so that was at the time before bit, I think, nearly an impossible task. And what they did was to embrace a different kind of thinking a little bit, and try to think what are the experiences that they're going to build? And then they built all these experiences with BIT as components, as building blocks. And they used these components to compose all the applications that they needed. And one year later, when they ran the study, they found that they delivered just a little in the first year alone, which means including building the components, which is a one off investment. They are later reusable, so you don't need to build them again. You can iterate, compose, build new ones. But building all of these components from scratch in that year and they more than doubled the number of applications, cut down by 60 something percent the cost per app and increase the consistency by over 80%. Basically because they were just reusing the same Lego piste everywhere, so didn't even have to work to make sure it is consistent. So that's a very strong, I think, example of why now, of why the need now. And today more and more organizations are facing this exact need. So if we're going back to technology, they're already working and thinking composable, everyone is trying to become composable because it's just, it makes more sense, right? If you go into a car factory, you don't see a team building my car and a team building Brian's car and a team building Joe's car or Deborah's car. You see a team building with a domain ownership over a domain like wheels or engines or electricity or whatnot. And they're building components and they're serving them to assembly line to the rest of the organization, which is a little bit like building microservices and serving API. So software, this is much more efficient because it's faster, it's cheaper, it's consistent, delivery time is much faster. If you want to update a component, the update goes to all of your products and software should be built the same way. And now the world has changed once because composability has become a simpler option. It was hard, it was very hard before bit because you had to implement all kinds of stuff and your own solutions and integrate all these tools and even then, and now composability becomes a much simpler choice. You already have all these APIs and packages and libraries and you already have all these features and you want to build new ones, just build them like pieces of Lego, build them as components. So I think that this is the moment in time that we're seeing this transformation already happening. And the second thing is that AI has dramatically changed the future and even the present. We're already seeing a drastic impact on the present of software development. And I think that this future, this combination of composability and AI is going to radically change the way that we're seeing software being developed by people from indie hackers to entrepreneurs, all the way to the world's largest enterprises. I think that we are in that moment in time right now.
- [11:42]
A
Thank you so much, Jonathan. I appreciate that. And I love the example, of course I've been in healthcare a lot of years. But you know, the fact that you come into a large healthcare organization and cut down the cost by 60% while keeping the quality consistency at 80%, you doubled the apps. You're able to do this using composability or your platform bit, which is awesome. So thank you for sharing that. Jonathan, my next question for you. BIT has gained traction with Fortune 100 companies. What are these large enterprise doing right or wrong when it comes to adopting composable and AI driven technologies?
- [12:14]
B
So Brian, I think this is maybe the most hardest question in the world, I think in our world right now, of software development. How is AI going to change the way that we're building software as people and as, as organization? So if we look at what's happening today, we're basically seeing two types of AI for software development, for building applications or any type of software. The first type is code generation. I think Cursor is a great example. It's an incredible product. Like once you try it, it's very hard to go back to what you had before. It really helps you write more code faster, makes the process easier, it even makes it more fun. But at the same time, tools that help us generate code also create huge problems for the organizations. Firstly, developers are now able to create a lot more code. So the number of reimplementations and duplication across your code base. And if we're looking at a big organization, you're looking at hundreds of thousands of different products or projects. And now it becomes nearly impossible to track or maintain and update all of those AI generated instances of functionality or features across the code base. The rate in which the code base grows becomes exponential. So all these problems that we already had before composability are now becoming 5 times or 10 times or 100 times worse because you're creating more code, you're creating the same code everywhere and then you're left with a huge code base that is largely AI generated, which also means that you're not consistent and you're not sure about quality, about standards, even about which text you use. Am I testing this react component with this testing technology or with another? What's the standard of the organization? And so it's very easy. It becomes much easier to lose complete control over your code base and it becomes more and more hard. Let's imagine that you are using AI to Create some feature, payment, a header, yes, something simple, right, for your applications. And now you have an unknown number that grows every day. Implementations of this functionality across your code base. You don't know where. And now you want to make an update to all of them. You know, good luck, that's just not going to happen. And the second type of AI for software development that we're seeing is tool that let you build software or applications without coding. This is now referred to as vibe coding, which I'm sure you've heard, but it's talking about that. And you're seeing tool like lovable and Bold and other tools that are now letting hopefully everyone use simple natural language to create applications, to add features, to add functionality. And the thing is that these vibe coding tools that we're seeing today are not the solution that organizations need or even a single entrepreneur that's going to build. A big company needs to build software with AI. Not to write code, not to generate code, but to build software. First of all, the architecture, you know, you're just getting us really getting one chunk of code. You're not going to build the next Airbnb on top of that code. Secondly, you know, it does not build full real systems and applications. You basically have no backend. Some tools offer pre made integrations or you know, make your own integration to databases such as Supabase and others, but it does not have a real backend to it. So you are pretty much unable to build, you know, full blown platforms. You can't really build a big company with these tools. If you try to use them within an organization, they are completely out of context of the organization. We talked about the fact that the organization is already becoming composable. You have APIs, for example. These tools are completely unaware of the APIs that you have in your organizations. They cannot use them, access them, use them to write or get information. And so what you're building is really kind of an isolated, out of context prototype that is not part of the organization's code base or system or technology. And lastly, the quality of code is not something that you can really use in production. If you're a single organization, you need very high coding standards, you need code that is tested, linted, documented, so on. And that is just not the case right now with the code that these tools create. So that pretty much turns most of the tools into prototyping tools at best within organization. While I think the future holds a lot more for natural language AI software development for organizations, the potential of what organizations and large enterprises, even small Startups can do with this is a hundred times larger if we just build it right in the way that fit their needs. And that is what we're doing at big right now. That is Hope AI, which I'll maybe tell you a little bit more about in a second. We are working on building and we're already releasing in the next couple of weeks the first GA version of an AI that is composable. So the way that all of these problems are solved is by building, you know, a better AI is by building an AI that builds composable software. And if you go to Hope AI on Bit Cloud, our website, you can ask Hope AI next week you can ask Hope AI to build Ready or to build booking.com and it will design a complete architecture for you. Bit can scan. If you're a big organization, you'll need a five minute chat with us for that. Will become familiar with all the existing APIs, packages, libraries in your organizations. You can set your standards, it will create the feature or the application that you requested. And it will create an architecture that is composable. You don't need pre made integration. You can write software, you can create new integrations inside and outside organization, internal or external tools and APIs. The code quality is of the highest enterprise grade standards. It comes with test coverage, LinkedIn documentation, everything you need. It can be consumed in any way, each component into any project. It can be shared with the rest of the organization. And it builds complete systems, including backend and everything. So now everything changes. Because now, for example, the organization can. You can ask. Even if you're a huge enterprise and you want a new app, just ask for the app that you want in a single prompt. And it will reuse the components you already have. It will access and use the APIs in your organization. It will lay down an architecture that you can see and review before you build it for a composable design of the feature application that that you want. And then it will build all the components for you. And it can even take your design for example, and your visual language. So you're not limited even it comes down to stuff sometimes that is just as simple as that, right? The design is not limited to some pre built library that this specific tool chose to use. You can use your own design, you can ask for any design you want. And eventually you can start building new features, new applications, update existing one and collaborate on your technology within the organization in context with the highest quality code, complex capabilities, backend everything. And it's all being done essentially using natural language. So when AI meets composability, then we get a nuclear explosion in what becomes possible.
- [20:18]
A
That's awesome. Thank you for sharing. I appreciate you unpacking that. And you did break out some of the issues today with having co generation type software, right, like cursor. Obviously you'll end up having a lot more applications you need to manage. It'll be a headache. But also the big hype right now around vibe coding, right, no code, low code platforms like lovable, etc. You're not going to be able to scale an enterprise company with that. Absolutely agree. Jonathan, looking ahead, what's your boldest prediction about the future of software architecture or developer tooling? Think that you'll surprise people.
- [20:52]
B
I think that the role of developers and other professionals and executives and stakeholders in the organization are going to completely change. And that's not a bad thing. We should not be worried, we should not be afraid on either side of this change. I think that AI, when it meets composable tools, composable technologies like HOPE and bit, will let everyone become partners in the process of software development. So now we can literally build together. I can review your things, you can review my things, I can use your things, you can use my things, we can all share, we can become partners. And a lot of the broken communications and the challenges of collaboration are going to go away. This is going to be something that is much better for everyone and I think that we're going to see it in two, three years from now. I think this is one of our personal aspirations and goals in bit and we're going to make it happen. We're all going to witness a change and everyone can be a builder and that's a good thing for everyone.
- [21:54]
A
Awesome. Thank you so much. And I truly think that as well, the future is truly building together when we can leverage BIT and composable software with AI. That's just awesome. So thank you. I like to focus on the future and what's emerging out there. Jonathan, it was such a pleasure having you on today and I look forward to speaking with you real soon.
- [22:12]
B
It was my pleasure. Brian, thank you for having me today.
- [22:15]
A
Bye for now.