Product Strategy in the AI Era: Learnings from Cloning X with AI Tools with Tamas Kokeny - Productized 24 TalkInTen

Productized 2024 is our third conference of the year, and we’re excited to bring We’re back with more exciting content from Productized 2024, our third conference of the year! This series continues our TalkInTen format, where we bring you short, impactful conversations with industry leaders who are pushing the boundaries of product and business agility. Packed with fresh insights, innovative frameworks, and real-world examples, these episodes are designed to help teams and organizations thrive in today’s fast-paced world.

In this episode, we’re joined by Tamas Koke, Technical Co-Founder of Bobcats Coding. Tamas shares how he used AI and no-code tools to rebuild X (formerly Twitter) in just three weeks. This fascinating experiment highlights the power of AI-driven prototyping and the growing role of no-code platforms in product development. Tamas discusses how AI can streamline the prototyping process, reduce development time, and help product teams iterate faster than ever before.

Tamas on LinkedIn - https://www.linkedin.com/in/eggdice/

Here is the synopsis of Tamas's Talk:

Curious about how new technologies have impacted product development, Tamas set out to push the limits of AI and no-code tools by recreating X/Twitter in just 3 weeks. In this talk, Tamas will share how this bold experiment reshaped his approach to product strategy and explore the exciting opportunities and challenges of AI-driven development.

If you enjoy the show, please leave a review and stay tuned for more great episodes from the Product Agility Podcast!

Use code PROD24 for 15% off training courses at Sheev.

Sheev - https://www.sheev.co.uk

Host Bio

Ben is a seasoned expert in product agility coaching, unleashing the potential of people and products. With over a decade of experience, his focus now is product-led growth & agility in organisations of all sizes.

Stay up-to-date with us on our social media📱!

Ben Maynard

🔗 https://www.linkedin.com/in/benmaynard-sheev/

🐦 https://x.com/BenWMaynard

💻 https://sheev.co.uk/

Product Agility Podcast

🔗 https://www.linkedin.com/company/productagilitypod/

💻 https://productagilitypod.co.uk/

🖇️ https://linktr.ee/productagility

Listen & Share On Spotify & iTunes

• Spotify - https://open.spotify.com/show/0lkwAYJzVSuk5zfJ1vIDZq?si=4c691fb12f124a56
• iTunes - https://apple.co/3YvTX8p

Want to come on the podcast?

Want to be a guest or have a guest request? Let us know here https://bit.ly/49osN80

Transcript

0:00

Welcome to a very special series of episodes of the Product Agility Podcast, broadcasting for two days, direct from Lisbon Portugal and product ties 2024. This year, we're bringing you more exclusive bite-sized wisdom with our Talks in 10 format, where we're going to be diving into actionable insights from some of the best and brightest minds in product leadership. And attendees this year are being spoiled with talks and workshops from the likes of Radakadat, creator of radical product thinking and Rich Morinov, author of the art of product management, helping us all find some joy in what we do. But it would be a pretty shitty conference if it was just two people. There are so many more people here, and they're going to be getting as many of them as possible on here to share their talks in 10. Now before we begin, a huge thank you to our sponsor, Sheev Limited. Sheev is the company which has bankrupted this podcast pretty much since day one. I want to take an opportunity just to share with you and make you aware we do some awesome stuff. Whether it's training your product teams or coaching your product teams with clarity and alignment, or you know just a simple thing of actually making OKRs work in organisations, these are all things that we are very good at. So do head over to www.sheev.co.uk, see what we do and get in contact with us. Also check out the show notes for a tasty little discount code over any of our courses. Grab a notebook because the next 10 minutes are going to be packed with action tips from the best in the business. And here begins a talking 10. We are joined by Thomas Koke. OK, yeah. Great name. Thank you. You are what is your job title at Bobcats? Chief Bobcat? Yeah, we could put it like that. Thank you for having me. I'm the technical co-founder of Bobcats coding. That means I'm managing the technical teams and I'm doing a bit of technical research. And what is it Bobcats do? So it's basically an outsourced company and we have started it beginning of last year. So because of that we have to do something about AI. No way. Not heard of that before. I think we only had two or three clients that haven't started to talk about. They are right away when we started to even have a conversation. Oh wow. Yeah. So your talk tomorrow? Yeah. It sounds like a really cool story. Which was how you used AI and low code tools to rebuild Twitter. Yeah. AKA X. Twitter was what it was called when it was more valuable, wasn't it? Yeah. Yeah. Yeah. Yeah. That's not worth less money. I saw that. 80%. Yeah. Yeah. But he's managed to wipe off more of the bad even not the financial crash or something. Something crazy like that. I still refuse to call it X. Yeah. Yeah. Yeah. Okay. Twitter. And how you manage to be basically copy it. Without infringing any copyright or putting yourself in the way of any kind of lawsuits. I'm sure. Actually, we're kind of afraid of that. But I think we are small enough that this is not going to be a problem. Well, my company name is Sheve, which is a Disney trademark. Oh, really? Well, because Sheve is the forename of the emperor from Scotland. Oh, I see. You see. So if you Google Sheve, go ahead and everyone listens to Google Sheve. I mean, you look down there, but third or fourth result is Sheve, making agile and product pretty nice. That's my company. Well, I'm on the, I've done really well. I'm just waiting for the cease and desist, you know. You're going to Disney now. Who is this? Who is this clown? Yeah. But I don't think I'm big enough. Who can I think you'd be all right? So you say how? How? Say, I can kind of understand the premise as to why you'd want to rebuild. Yeah, yeah, yeah. Right? Because it'd be fun. It's a nice challenge. But like how like talk to me, tell me the story a little bit. Yeah. It's obviously in about the six minutes we got left. So we wanted to choose a platform that's known by everyone. And and also that has in one way, simple, but also complicated features. So because of the multimedia stuff that you can put there, also the connections and so on, it's it's not a trivial app, but it's also not a really complicated app. And I think these type of apps are really good example for an early prototype. I think if your early prototype is more complicated than this app, then you have spent too much time on your early prototype. And and basically that's why we choose this to show that with these new technologies, you can deliver an early prototype way sooner than it was possible before generative AI tools. And I think that's the biggest difference in software development, not how you're able to accelerate the development time with tools, but rather how you can rapidly prototype and do while while you testing right away. So this thing you mentioned good AI, whatever it is, like how what exactly were you getting AI to do and also what what form of which model were you using as part of the process? Yeah, it's a good question. Actually, in we have written a white paper about it, and we want to paint the holistic views. It's not just about creating the product itself, but rather create the brand, create the logo, create everything. Similarly, how would you do it with an early stage startup? And there was majority of sorry, a vast amount of tools that we are using there. So it's not just a single tool. And also, I think most of the people are thinking about, okay, what's the future of AI? Will we have the ultimate tool? Is it I don't know, cloud now or JPGP or whatsoever? But I think where we will have really interesting stuff is all the small AI features of conventional tools that we are already using. So actually, in this case, the majority of the app have been built with a bubble IO. And bubble itself is not really an AI tool, but it already has a good amount of AI features that are cutting corners amazingly compared to, okay, I just do something with a website builder or something like that. So for example, you can create full front ends like a page just with prompting and that's it. And I think that really makes the difference. It's like tangent, perhaps. And this made the sentiment of data and we can just move on. But I was talking to a friend of mine who is at the large scale Scrum Conference last week. And one of the talks there was talking about kind of AI and basically what the future means in organizational design. And I don't know if it's specific to the talk, but it got me thinking that for a long time in agile teams, as an example, in product teams, there was always the argument you couldn't have all the skills you needed on a team. And so as a consequence, you had to kind of create these kind of fake delineations between teams, on the maximum number of schools you could have. And if you were in a really complex environment, which was maybe unnecessary complexity, but it was still that complex, you couldn't have teams own the end to end. Yeah. Yeah. Too much going on. Do you think that with AI really picking up speed and these neat tools coming in, it kind of maybe goes back to making that dream viable? Yeah. I think so. I think, yeah. Actually, I will have a talk about this tonight on our meetup that I think basically what's going to be the long term difference that people that are T-shaped, that they are expert in something and have a bigger knowledge in other areas, I think the T, like the horizontal part of the T is getting bigger and bigger and bigger because of the AI tools. And it's, I think, true for rapid prototyping like this, because a single person can do it and you don't need an additional designer whatsoever. But also, I think it's true for like traditional product management and product development in general as well. For example, with full-stack developers now, I really don't get the argument that, oh, I'm better with front-end, so I'm not going to touch some, I don't know, infrastructure code or something like that. If there is a well-established technology that you're, there is here for, I don't know, for five years, then AI is amazing to generate it. Yeah. And they're really fascinating. Thank you for indulging me on that a little bit, because I think it's really fascinating. It does, I think it froze up a lot of interesting questions and stuff we just haven't ever do historically. I think it's really fascinating. But going back to you copying, creating something very much inspired by Twitter. Were you able to do it? Yeah, actually. And I think that's also the, so in one hand, it's the AI features of bubble. But also, most of these tools are having amazing built-in features already and connections or plugins with other services. So quite a few features that are granted in any kind of B2C application, like you have to log in, you have to have a profile whatsoever, are there. You really don't need to do it anymore. And I think for prototyping, if you're spending time on that, that's just waste of the space. Yeah. There's, we've got a product coming out and we decided, came back a year or so ago, but just to do it in WordPress because a lot of the stuff we need with WooCommerce, it just does it. I said something button. There's no need to build. I remember 20 years ago building checkout experiences and actually learning them. Yeah, it was just a pop. Yeah, yeah. Yeah. It was a nightmare. Yeah, but it wasn't a nightmare. It was surprising. It was easy, but it was still very much involved. Yeah. Now it's just like, I'll just get that and it works. And it works. And speed things up so much. And I think maybe their future is the art of balancing what you code versus what you don't code. Yeah, exactly. A lot of the time and actually getting that right balance. Yeah. Yeah. Yeah. Ecosystem. Yeah. I think developers that are not getting familiar with no coding tools can I be out of jobs in a couple of years. Yeah. Yeah. Definitely. You look happy about that. You're an evil bastard. Yes. I'm happy about it because I cannot wait that the AI takes her jobs. And I don't have to work that much on rep after the things. So you'll talk tomorrow. What time is it? I'm just the one before the last one. Oh, nice. I was a strong slot. Yeah. I think the George, I think you had a similar slot one time at a scrum conference perhaps. If people want to find out more information about you, tell us where can they go? We have a booth here. But I've realized we were on the call. Yeah. Yeah. But you can find me on Twitter or X. Yeah. I handle his egg dies. But also if you go on bobcatscoding.com, you can find me and anything about us. Excellent. Well, look at that. It's a pretty good timing. Just going to say the same time. In Thomas, we've spoken a lot already today. So thank you for putting up with me yet again. I'll come and bother you later. I really want one of those pink hats, but I've been told I'll have to win it. Yeah, you have to participate or ruffle. Yeah. I participated. See, if you think it's winks don't show in the podcast, do they? So like, yeah, hopefully I'll win. Thank you so much for coming on, Thomas. Thank you for having me. Thank you very much. Listening. If you've loved this episode, please let us know on social media. Thomas, tag myself. We'd love to hear your feedback. Everyone, thank you very much listening. George, thank you being a lovely producer. We'll be back again soon.