Why a16z Built a Town for AI People

In this episode, Nathan sits down with three members of the a16z x Convex AI Town project: Yoko Li (Partner, a16z), Martin Casado (GP, a16z), and James Cowling (CTO, Convex). AI Town is a virtual town where AI agents live, interact, and socialize. They discuss how AI Town originated from Yoko’s companion app project, unpredictability as a feature in LLMs and interacting with models like they are lifeforms, and why they chose Javascript and Convex to build AI Town. If you're looking for an ERP platform, check out our sponsor, NetSuite: http://netsuite.com/cognitive We're hiring across the board at Turpentine and for Erik's personal team on other projects he's incubating. He's hiring a Chief of Staff, EA, Head of Special Projects, Investment Associate, and more. For a list of JDs, check out: eriktorenberg.com. 📣 CALL FOR FEEDBACK: To borrow from a meme… we're in the podcast arena trying stuff. Some will work. Some won't. But we're always learning. http://bit.ly/TCRFeedback Fill out the above form to let us know how we can continue delivering great content to you or sending the feedback on your mind to [email protected]. TIMESTAMPS: (00:00:00) - Episode Preview: Intro to AI Town and the idea of AI as companions (00:05:29) - Overview of AI Town and the simulation framework based on the Stanford Generative Agents paper (00:08:24) - Yoko explains how the idea for AI Town originated from the companion app project (00:12:31) - The simplicity and elegance of the AI Town codebase (00:15:47) - Sponsors: Netsuite | Omneky (00:18:25) - How Convex built a server-side game engine for AI Town (00:23:39 )- James emphasizes the power of functional programming paradigms like Convex for building AI apps (00:25:02) - Using simple JavaScript so anyone could understand and extend AI Town (00:30:23) - How AI coding assistants were used in building AI Town (00:31:22) - No open source code for the Stanford paper when they started (00:33:25) - The interplay between programmer and AI model (00:38:01)- Martin draws a distinction between using formal languages vs. natural language (00:43:21) The balance between formal language and unpredictable behaviours in LLMs (00:48:27) Are we living in a simulation? (00:50:38) Advice for other developers in AI (00:54:29) AI Town is a community project to be extended on SPONSORS: Shopify | NetSuite | Omneky Shopify is the global commerce platform that helps you sell at every stage of your business. Shopify powers 10% of ALL eCommerce in the US. And Shopify's the global force behind Allbirds, Rothy's, and Brooklinen, and 1,000,000s of other entrepreneurs across 175 countries. From their all-in-one e-commerce platform, to their in-person POS system – wherever and whatever you're selling, Shopify's got you covered. With free Shopify Magic, sell more with less effort by whipping up captivating content that converts – from blog posts to product descriptions using AI. Sign up for $1/month trial period: https://shopify.com/cognitive NetSuite has 25 years of providing financial software for all your business needs. More than 36,000 businesses have already upgraded to NetSuite by Oracle, gaining visibility and control over their financials, inventory, HR, eCommerce, and more. If you're looking for an ERP platform ✅ head to NetSuite: http://netsuite.com/cognitive and download your own customized KPI checklist. Omneky is an omnichannel creative generation platform that lets you launch hundreds of thousands of ad iterations that actually work customized across all platforms, with a click of a button. Omneky combines generative AI and real-time advertising data. Mention "Cog Rev" for 10% off. Music Credit: GoogleLM

Speaker names marked ? are inferred from speaking time and the show’s known hosts — they are not voice-verified.

• 0:00Speaker 00

  So the previous project we coded up, which is called Companion App, was based on this idea that AI can be served as companions and then as a human, you can, you know, talk to it about your problems or maybe have a conversation, learn things from the AI. The thing I did notice is like when you try these things firsthand, you give these like AI companions life by giving it more details, like what do they like? Do they prefer Diet Coke or real Coke? You know, it's like all the ditty gritty details you would have in like an actual friend or human. And then over time, since we implemented the memory implementation, they really remember everything, like every interaction you had with it. So that project, after we launched it, obviously, the community kind of came together, came around it. And then we started to get questions around like,

• 0:45Speaker 00

  okay, this is great to have one on one conversations. What about having the AI as part of a group? Like what happens when you kind of play games with the AI?

• 0:55Nathan Labenz?

  Hello and welcome to The Cognitive Revolution, where we interview visionary researchers, entrepreneurs and builders working on the frontier of artificial intelligence. Each week we'll explore their revolutionary ideas and together we'll build a picture of how AI technology will transform work, life and society in the coming years. I'm Nathan Lebense, joined by my co-host Eric Torenberg. Hello and welcome back to The Cognitive Revolution. Today, we have a fascinating conversation about how quickly AI research ideas are moving to practical implementation and what that might imply for a world full of AI populated simulations. Programmers and partners at A16Z Yoko Lee and Martin Casado teamed up with James Cowling,

• 1:39Nathan Labenz?

  CTO of development platform Convex, to create an interactive simulation of human behavior using AI agents, which they call AI Town. This implementation was based on the recent Stanford paper, Generative Agents: Interactive Simulacra of Human Behavior, in which researchers introduced a new simulation framework, which was headlined in my view by a novel memory structure, which combines direct observational memories with higher level reflection memories, which are periodically generated over time. Such that the AI agents were able to effectively navigate their virtual town, have coherent conversations with one another, and even make and execute plans together. The open source version of this project now runs on TypeScript development platform Convex,

• 2:23Nathan Labenz?

  where a single shared runtime processes all the many LLM API calls and other asynchronous events that come together to create the virtual town. Coming into this conversation, I imagined that this project must reflect an investment thesis at a 16Z related to AI simulations. But what I learned is that this project is actually just more playful and experimental than all that, because modern development platforms like Convex make it so easy to build such things. It now often takes months, weeks, or even just days for academic research to start popping up in applications. And I do see this as a sign of things to come. It's been maybe a year since LLMs really started passing the Turing test.

• 3:06Nathan Labenz?

  And already, people spend hours a day on character AI, where the number one use case, by the way, is romantic role play. And meanwhile, Inflection's Pi has exchanged 1 billion messages with users. That's roughly 10 million per day in its first 100 days. Now, we're seeing speech synthesis and video deep fakes that are sometimes legitimately hard to distinguish from the real thing. So with simulation frameworks like AI Town becoming more accessible, it seems almost inevitable that we'll soon have a wide range of simulated AI-populated worlds to explore. As always, often for better, but in some cases, for worse. Now, if you want to understand the technology underlying these developments more deeply, and you haven't already, I really suggest you check out my AI scouting report, which

• 3:49Nathan Labenz?

  is available on the Cognitive Revolution YouTube channel. We made this episode a YouTube-only feature because the visuals are really critical. And while I think this is some of the most valuable content that we've created, and our reviewers seem to agree, part one just hit 3,000 views, which is significantly less than our typical podcast downloads. So please, check it out. And if you have a friend that's trying to get up to speed on AI as quickly as possible without relying on misleading metaphors, maybe send it to them as well. Now, without further ado, I hope you enjoy this glimpse into the rapid development of AI technologies with Yoko Lee, Martine Casado, and James Cowling. Yoko Lee, Martine Casado, and James Cowling, welcome to the Cognitive Revolution.

• 4:35Nathan Labenz?

  Thank you. Excited to talk to you guys. So just to set the stage, Yoko and Martine are partners at A16Z, where you guys focus on infrastructure and AI. And James, you're the CTO of a company that A16Z has invested in called Convex, which is a development platform emphasizing TypeScript and the incredible ease of making some magic happen. You guys have kind of collaborated recently on an implementation of this project that made a huge amount of waves, at least in my part of the world, not too long ago, which was this simulated town. And people have probably seen these little images or seen the story of, oh, my God, researchers

• 5:18Nathan Labenz?

  set up this thing where all these little virtual people walked around and talked to each other and kind of made plans and then got back together and executed on their plans to some extent. And it is kind of crazy to think, boy, it's getting to the point where we can simulate some pretty sophisticated behavior. And I think this just has so many different angles that I'm excited to unpack with you guys today. So maybe for starters, I'll ask Yoko and Martine, it seems like if I understand A16Z correctly from an outside view, one of the firm's big ideas is that you're going to take huge swings on kind of unpopular theses, try to get to things like before they go mainstream, and that

• 6:02Nathan Labenz?

  some of these might look crazy at the time that the investments are made, and some might even still look crazy later on downstream. But it seems to me that you guys are kind of working on a big crazy idea here, which is that we are headed for a ton of simulations and kind of virtual agent populated environments. I guess I'd love to just get a little bit more of your deep thinking to get started on like... What is it that you kind of see coming for us in this vein of simulations? And why are you so interested in starting to build the foundations for such simulations?

• 6:36Speaker 02

  So the FERP kind of has two, I think, underlying aspects of the philosophy. One of them is it's always really focused on operators when it comes to hiring partners. And so a lot of us were software developers or founders or, you know, were associated product lonely. And the second one is there's this general feeling that consensus is risk averse, right? So you certainly want to go to areas where there isn't a lot of consensus. I will say when it comes to this AI move, it's probably worth knowing that Marc Andreessen, who was a founder of the firm, did Netscape, which was the browser, which is a very similar type movement, which at the time it was created, people were like, oh, this is not useful,

• 7:20Speaker 02

  like this is wild, it's just a bunch of kids and they're on the internet, talking to each other, playing video games or whatever. It was kind of treated like a bit of, you know, it was like kind of a hobbyist thing. You know, I'm not sure, maybe James remembers, but like Eric Schmidt, when he was the CTO of Sun, he actually banned the browser because he's like, people won't get, people won't get work done, right? And so, you know, very obvious in these new waves, you know, it's like the enterprise and kind of the, you know, I think the more mature aspects of the industry tend to issue it and whatever, but then you end up having these kind of very interesting, you know, often creative movements that arise up. And so we saw this very, very early with the AI wave when we've been involved.

• 8:02Speaker 02

  Now, I want to make the point that actually the creation of AI Town itself was actually a hobby project by Yoko. And so I thought maybe it would be worth Yoko going through kind of like, you know, where this, this came from, because it's turned into a big thing, but originally it was like, you know, Yoko, you know, at home.

• 8:20Speaker 00

  Yeah, I can shed more light on how this project kind of got started. I mean, it's easy, so kind of like what Martine said, all of us are developers. So really, we kind of spent weekends and nights kind of like at home coding our projects. So the previous project we coded up, which is called Companion App, was based on this idea that, you know, AI can be served as, you know, companions. And then as a human, you can, you know, talk to it about your problems or maybe have a conversation, learn things from the AI. So we created quite a few personalities in Companion App. And then one of it is called Alice, and then I text Alice all the time from the app I built.

• 9:04Speaker 02

  So we were making this app for everybody to play with, right? And Yoko got so attached to Alice, her friend, that she didn't want other people to be able to talk to Alice. And so she took Alice, she took Alice out of the app.

• 9:14Speaker 00

  Yeah, so I ended up creating like a male version of Alice, which is Alex. The thing I did notice is like when you try these things firsthand, you give these like AI companions life by giving it more details. Like what do they like? Do they prefer Diet Coke or real Coke? You know, it's like all the ditty gritty details you would have in like an actual friend or human. And then over time, since we implemented the memory implementation, they really remember everything, like every interaction you had with it. So that project, after we launched it, obviously, the community kind of came together, came around it. And then we started to get questions around like, okay, this is great to have one-on-one conversations. What about having the AI as part of a group? Like what happens when you kind of play games with the AI?

• 9:59Speaker 00

  So that's where I started to think about, okay, how can we get like greater environment? So the AI companions can talk to each other as well as talking to human. So my husband and I, my husband is also like a programmer. So we started hacking around this on the weekend, built a prototype using Phaser JS. Thank you Phaser. And I'm building like a client-side version of the implementation of generative agents, which is like a really great paper fleshed out by Jun from Stanford. Everyone should check it out. And then the next question that came to us was, okay, so now it's great to have a single-player mode where agents can walk around, talk to each other. What's the implementation for a multiplayer mode?

• 10:44Speaker 00

  How can we enable this as like a product or a platform that other people can, you know, fork it, build their own AI town around it, and then even have multiplayer already? So that was the genesis of the project. And then later, kind of Martine and James and the whole Convex team kind of came around it and helped us to kind of take the project to the next level. And we open sourced it. It's MIT licensed, so you can really open a fork it and then use it for any commercial project, what have you.

• 11:15Nathan Labenz?

  So there's so much to follow up on there. I mean, I think there's kind of two levels at least for this conversation. One is kind of the behavioral, psychological, like, future of, you know, relationships. And then there's kind of the technology that underpins that. And I think both are super interesting. But it is striking, you know, right off the bat that, you know, we're so early in this paradigm and, you know, there are so many kind of fundamental things that are still in the process of being worked out, like memory, you know, which you mentioned, that obviously, language models by themselves don't have these kind of intermediate-stage memories that are so important to our ongoing sense of like self and ability to relate to each other. But that is starting to be worked out with projects like this.

• 12:00Nathan Labenz?

  But even as early as we are, you know, it's extremely compelling to people who are not like extreme people in other dimensions, right? I mean, I think people are all too quick to sort of assume that this is like a fringe phenomenon. And I'd love to just hear more about kind of what value you're getting from this, what kinds of interactions you're having. Because as much as like the future of the technology, I think the future of ourselves and just kind of imagining how we might also start to follow you in interacting with these AIs is super fascinating.

• 12:31Speaker 03

  I'm not going to make it through any podcast without talking about simplicity and how much I admire the concept of simplicity and elegance. And I think people should go look at the source code for AI Town. There's a lot of ideas in AI Town that in themselves are not complicated, but they're sophisticated, right? So the way memory is represented in AI Town is not complicated, right? You, and we can talk in more detail, but you have a conversation, you record it in a log. Every now and then you ask OpenAI to summarize that memory. And then you kind of use vector search and embeddings to come and, you know, use these to prompt future conversations. These are relatively kind of simple building blocks and simple kind of elegant ideas. Now, something being simple doesn't mean it's easy to come up with, right? It's, it's, you know, and props to Yoko and the team for calming up with these abstractions.

• 13:15Speaker 03

  I think what's particularly interesting about something like AI Town is at its heart, it's a very simple project, right? It's doing a few basic things. And then you have something that looks like, you know, emergent human crowd behavior. And, you know, we can insert human actors in that as well. So I think what's just really, I think, anyone who spends time building AI apps, I spent the weekend doing it because I was, you know, feeling excited. You really struck with how quickly you can build something really shockingly useful, impressive, based on relatively simple primitives.

• 13:52Speaker 02

  These models really are kind of life forms in their own way. And so if you're building a project like this, there's kind of two considerations, right? And we should talk about both of them. The first consideration is like you have these life forms and like, you know, it really is a brain and you're interacting with this brain. This is the model. And so much so that like they surprise you all the time. So for example, I once like missed some quotations on my string that I was passing to it. And I actually accidentally passed it some code as opposed to like a conversation. And I started commenting on my code, right? You know, like you don't really know when you're putting in these things like what you're going to get out. When, so I, I actually paid my way through college, um, building the new games. So this is kind of like, you know, I did game development back a long time ago.

• 14:36Speaker 02

  Like it's been a very long time to say 20 years ago. Uh, and at the time, everything was an algorithm in the code. Like, how do you go from A to B? Like, how do you interact with things, whatever? It's like this kind of, you know, big if else tree. But when you're dealing with these models, they're like these big brains. And so, so they do the work. So you can, you can, you can tell the model like, what do you want to do? Like, how do you want to get over here? Like, how do you want to interact with these things? And so it really changes the way that you interact with software. And then you get these merchant personalities that you actually get attached to, just like Yoko did. And so that changes the way you think about software. But what it doesn't do is it doesn't change your need for a lot of the basics. Like, you still need transactions, you still need to like have the state go globally, you

• 15:21Speaker 02

  still need to be able to kind of, you know, work with, you know, multiple users in different places, you need strong hair and teeth, you need all the mechanics, right? You know, even though that AI is relatively simple, apps, the bands are really real. And that's kind of where Convex came in. So I think maybe for the listeners, it's probably worth James just going through the background of what Convex is, and then we can explain why we thought it was the best platform for this.

• 15:47Speaker 04

  Hey, we'll continue our interview in a moment after a word from our sponsors. If you're a startup founder or executive running a growing business, you know that as you scale, your systems break down and the cracks start to show. If this resonates with you, there are three numbers you need to know. Thirty-six thousand, twenty-five, and one. Thirty-six thousand, that's the number of businesses which have upgraded to NetSuite by Oracle. NetSuite is the number one cloud financial system, streamline accounting, financial management, inventory, HR, and more. Twenty-five, NetSuite turns 25 this year. That's 25 years of helping businesses do more with less, close their books in days, not weeks, and drive down costs. One, because your business is one of a kind, so you get a customized solution for all your KPIs in one efficient system with one source of truth.

• 16:30Speaker 04

  Manage risk, get reliable forecasts, and improve margins. Everything you need, all in one place. Right now, download NetSuite's popular KPI checklist, designed to give you consistently excellent performance, absolutely free, at netsuite.com/cognitive. That's netsuite.com/cognitive to get your own KPI checklist. netsuite.com/cognitive.

• 16:52Speaker 04

  Omnikie uses generative AI to enable you to launch hundreds of thousands of ad iterations that actually work, customized across all platforms with a click of a button. I believe in Omnikie so much that I invested in it, and I recommend you use it too. Use Cogrev to get a 10% discount.

• 17:10Speaker 03

  Absolutely. So Convex is a full stack development platform. And the Convex founding team and a lot of the Convex engineers have spent a lot of time building very large infrastructure. And there's a lot of commonalities you see in very large infrastructure and the certain ways to build infra that just works and certain ways to build infra that gets complicated and falls apart over time. Convex is a platform for folks who just want to build applications, front-end developers, empowering them to build scalable industrial strength applications in TypeScript. And so my pitch is always going to be, if I can, if I see a domain where there's a really good idea, and AI is such a classic case, because so many AI apps I see, the hard part of the AI app is not necessarily the AI, it's the app.

• 17:54Speaker 03

  Hey, I have a great idea, I'm going to fetch some embeddings and search for them. The hard part's building the rest of the application. And so where Convex comes in is, hey, here's an amazing idea from the Stanford paper and Yoko turning it into a real live simulation. How can we turn that into a real scalable application really fast? And that's where Convex came in. AI towns built on Convex, that's the framework it uses. The data is all stored in Convex, Convex scheduled actions run the application.

• 18:25Speaker 00

  Convex team actually kind of created their own server-side game engine for this product. So when we got started, it was like a very simple client-side thing where all the logic are happening in your browser. So the problem with that is that you can't possibly run like a multiplayer world with it. You just have to keep getting updates from Phaser, which is like a great framework, but we wanted something like where the logic is shared globally. And Ian, who's the developer advocate, Ian Mark Kurtney from Convex, he came in and then we kind of bounced ideas on, okay, how do we architect this? So that everyone don't have to, you know, build their own game engine,

• 19:07Speaker 00

  can just take something simple, use it, maybe even extend it one day. So Ian really created this service side game engine that kind of, it's, at the end of the day, it's a loop. All game engines are loops. This loop was backed by some, like a very powerful database, which is Convex. My takeaway of this whole experience is that writing a game engine was so easy with Convex, because all the transactions, all the journaling, it was taken care of by the backend. So all we needed to do is to run a loop. So my simple JavaScript brain, I mean, I'm a Python developer and like a backend developer too, but like recently, I just really love JavaScript. My JavaScript brain is that, you know, like, anyone can create their own game engine nowadays with Convex, because you don't have to worry about this transactional semantics.

• 19:51Speaker 00

  And what's more interesting to me is that for the game engine, you can really, you know, like, just start extending it on top of the Convex. So say tomorrow, you want to create like a cat town. You want a different set of interactions. You can just fork the repo, add more loops on top of it, and Convex will just take care of the rest of it.

• 20:08Speaker 02

  What AI Town is, is it's a virtual town, a virtual environment. You know, it's like a level imagine, you know, and it's got like, you know, potentially houses, whatever you want to put on it, and it's got these characters. And each character is the front end to a model. And the character has a backstory, right? Like, your name is Bob, you're grumpy, you're a gardener, you don't like talking to people, or, you know, your name is whatever Paul, you've got something to hide. And every once in a while, it slips out, you know? And, and then these characters just walk around with their backstories. And whenever they get close, then they, they start talking and they start interacting. And then they have these conversations, and the conversations are not in any way, you know, written in the code. This is just what the models will say to each other.

• 20:52Speaker 02

  And then as they have the conversations, there's two things they do. One thing they do is they'll save the conversations, so that they can kind of have some memory of interactions. But they also reflect on what has happened and they reflect periodically. And then when they reflect on things, they come up with kind of like higher level ideas about it. And so anybody that wants to build one of these simulations could pick up AI town. And as Yoko said, they could like change the tile set, change the level or change the characters, change the character backstories or extend it in any way possible. So it's really intended to be a very general framework for anybody that wants to experiment and build like an interactive AI simulation. That's the basic idea.

• 21:35Nathan Labenz?

  Yeah, I think this is pretty important to highlight too, because a lot of folks who are coming from a research background, certainly in ML or who have kind of come in as like the AI engineer, you know, profile over the last couple of years and largely picked up a lot of Python repos and started to build on top of those are very accustomed, and this definitely includes me, very accustomed to working in a sort of script type format, a notebook, and everything there is very like blocking and kind of synchronous by default. Whereas this paradigm, you know, the JavaScript, TypeScript paradigm more generally, and you

• 22:20Nathan Labenz?

  know, this with being just a super clean, you know, convenient implementation of it, does really lend itself more to kind of a functional form where things are kind of naturally able to happen at their own pace and like when things are blocking, you know, that's okay, you can kind of wait for them to come back. The whole town doesn't get frozen, you know, because one inference is kind of hanging. And so what's been added here from an architectural standpoint is basically like that long lasting kind of shared runtime, right? It's it's now not just a single, you know, back and forth that you can kind of pick up where you left off, but there's a shared. Runtime where, you know, in theory, all these different language models, you know, they could even be different providers, right?

• 23:04Nathan Labenz?

  They could have different sort of latency profiles, whatever. And you can easily support all that. So I think this is something that is really worth people studying a little bit because if, depending on what kind of experiments or what kind of experiences they want to create, it is, you know, it's almost like the whole world is flipped because, you know, I'm fairly old enough to remember, you're going to tech history a little bit, but I'm, you know, it used to be that Python was like the, you know, the functional, you know, programming default, right? And now that's like totally flipped to JavaScript. I think that whole thing is extremely interesting to me.

• 23:39Speaker 03

  I just think functional programming is really having its renaissance. It was going to have its renaissance again. Convex is based on functional primitives. I mean, Haskell, for example, is a huge influence on the design of Convex. Actions that get triggered and run and they, and they trigger other actions and subscriptions that dynamically fire in response to data changing. You know, it's funny the way you framed it because, you know, the real world doesn't stop because I'm thinking, right? The real world doesn't stop because I'm having a conversation with something. Everything just keeps on running, right? So, you know, now if, if, if we're, if we're writing the, simulating the world as a, as a, as a single loop that blocks, yes, that's, the application is going to block. But, you know, if, if you design something based on functional primitives,

• 24:22Speaker 03

  yes, the characters walk around every now and then they just have nothing to do. So they stop and reflect and you see a little thought bubble show up. And they're talking to, to OpenAI and they're saying, hey, take some of these memories and, and, and say something with previous conversations and synthesize them into memory so I can reflect. That's happening while other characters are walking around and having a chat. And, and it just works, right? And I think, yeah, I think that this, this method of designing things and building things in, in functional ways really is a, is a, is a, is a pretty close analog to how things work in the real world.

• 24:56Speaker 00

  Yeah. So when we were first starting to build this, this project, one thing that was clear to me is that I want this to be like, The most common denominator, so anyone, maybe like someone who didn't study computer science in college, maybe someone who is just interested in AI and I want to learn about kind of deploying apps, how it worked. I wanted to have like the ecosystem that's like supportive and having enough resources for people to pick up and run. And I don't want front end and back end to be different languages. It's just higher barrier to entry that way and it's a lot of mental barriers. It works probably for a large scale app, but we're getting started. That's just not the case.

• 25:38Speaker 00

  I've always been a fan of TypeScript, like started using it since twenty fifteen. Like the design choice of like what the language is gonna be, what's the ecosystem I want to appeal to was pretty natural. But later I also found that like since the new year's research, it's all based on Python. But Python, at the end of the day, it's like a great tool for kind of like coding up things in your notebook. But when you want to deploy an actual application, it's usually JavaScript these days. Most of the developers we talked to, and we talked to a lot of them, that when they get started, like even for like a non-technical SaaS product, it's always just based on JavaScript, like have a backend like convex that can handle your transactions. So you don't have to take care of it. Not all of us are DevOps at the end of the day.

• 26:22Speaker 00

  And for someone who used to be very deep in the DevOps world, I think most of people shouldn't handle that work. So for JavaScript, we really wanted it to be something that people can easily fork. And even maybe, you know, start to learn about how programming works. So I think that's just the perfect choice for the world.

• 26:40Speaker 02

  It is interesting. Like I feel like we're like seeing kind of like this confluence of like Python and JavaScript now. And we'll see how it turns out. So listen, I'm a longtime Python programmer. So like actually my first job out of undergrad was like doing computational physics and these like large, you know, you'd use like basically Python to drive these large C++ programs. And so like you got really deep in the language because you're embedding it. So that has always been kind of my go-to language. So I've never known JavaScript because Yoko has been so enthusiastic. I'm like, huh. Maybe I'll try it out. And I mean, listen, the learning curve is high because of all of the libraries and stuff. You've got a kind of end framers who've got a deal with, but once you get this incredibly sophisticated language with this got incredibly sophisticated, you know, support for things that I wouldn't, I mean, I wouldn't expect in a scripting language.

• 27:23Speaker 02

  And then with TypeScript, I think you can make these kind of very secure applications that are quite safe. And so it's been a great learning experience. But the biggest thing for me is just how much support there is in the broader ecosystem for JavaScript and TypeScript if you're building an app. So like when we first built like the AI Starter Kit, which were precursors to this or the companion. If you're like, hey, I want like a global queue that I can just store stuff in. Like, I don't want to write my queue. I want a global queue that's, that it'd be like, oh, we'll use Upstash, right? Or we're like, oh, I want authentication. You know, I want to be able to authenticate using your GitHub account. And we just like, oh, we're just going to go ahead and use, you know, Clerk. And so it really does feel like when you're in the JavaScript TypeScript ecosystem, that like there's kind of like the serverless microservices dream.

• 28:06Speaker 02

  And I know these are super buzzwordy things, but like it actually is kind of real. Like it's a couple of lines to add these very major services that are totally stateful and globally available. And so it just turns out with a few lines of code, you can write these pretty serious things. Then there always is the question of the backend. You know, listen, I think that's the most important decision you make if you're building a global app. And I think that's where for something like AI Town, just Convex made the most sense for us just because it's not only a runtime for JavaScript TypeScript. Like you can actually run transactionally blocks of code in the backend, but it has like all the searches and queries and stuff you want from the database.

• 28:39Speaker 03

  Isn't it so interesting that, you know, once upon a time, like C++ was the real language and everything else was like the fake languages?

• 28:45Speaker 02

  It's a great, it's still a great language.

• 28:47Speaker 03

  Yeah, I mean, I don't know if it ever was a great language. We can have an argument on another podcast about C++. But certainly if you wanted to do like the real industrial strength stuff, like the access, the gateway to the, to the real stuff was C++. And now, you know, the, the amount of resources and libraries and frameworks available for Python, but also JavaScript and TypeScript in particular is amazing, right? And this is just, you know, to, to kind of. Parnell Martin's point about, you know, microservices and SaaS, as much as these might sound buzzwordy, this is just the evolution of the industry. Once upon a time, you had to build everything yourself, right? And so you had to build everything yourself and

• 29:29Speaker 03

  you were building it, yeah, whatever it was, in C or whatever. These days, you know, the incentives for an application developer are to build as little as they can because they want to build an application. I mean, most of AI town was done in two weeks.

• 29:43Speaker 00

  Oh, yeah, two weeks. The prototype itself was like two days and we just wrote the whole thing in two weeks.

• 29:48Speaker 03

  And so when you want to get stuff done fast, you want to reuse components. And it just turns out like Convex is like a JavaScript TypeScript first platform. And you might say, well, why did we even choose those languages? Well, one, you know, they run in V eight and we can determinize them, et cetera. But the main thing is that's just what the people want, right? And if the people, you know, the web has won, everyone's using JavaScript and TypeScript. And anyone who has a platform has to put out a library for JavaScript and TypeScript. And so this is a lot of power at your fingers if you're an application developer right now and looking to use, you know, third party services to accelerate what you're building.

• 30:23Nathan Labenz?

  I would love to get a little bit more into kind of the details of the process of doing this. I mean, you're in an interesting situation here where you had a reference codebase, right, in Python from the original authors out of Stanford. And then you kind of maybe elaborated that. I don't know to what degree, you know, it was kind of a point for point thing. But I'm struck by the fact that with language models, what language you're coding in, in some ways like matters less than ever, right? Because if you can express yourself in any language, clearly, including like English, for example, you know, you can get a huge jump on the code. So I'd love to hear how you, you know, took

• 31:08Nathan Labenz?

  advantage of kind of AI coding assistance in this process and just your view on to what degree this is all kind of blurring together into a, you know, a fundamentally new reality. I would love to see us move past the language wars personally.

• 31:22Speaker 00

  I guess one thing to point out is like when we kind of started the prototype and the project, there really wasn't an open source yet. Generative agent wasn't open source at the time. That was actually one of the motivations for us to kind of open source it. Cause we're like, this is a great, this is one of the most interesting people will have read in this space. And then there's nothing open source on it. So that's the origin of it. But later, June kind of like open sourced it. And then we were able to kind of take a deeper look at it. Cause it's like very Python based. And I was telling June that together we cover all the bases. There's the Python developer and the JavaScript developers. But at the end of the day, I actually think the patterns are more similar than we had thought. But the, the reality is it's just the ergonomics of how to work with things.

• 32:07Speaker 00

  Like in Python, you can have non-blocking things. You just have to implement global async I O, which is very painful. It's doable. In Python, you can implement multiplayer stuff. You can implement web stuff like, you know, Django has come and go. But at the end of the day, it's just what's the tool that's the easiest for the programmers to reach for? And we do want to unlock that latent demand. So someone who has, you know, very little fundamentals of like how web works. Like I didn't know, I didn't always know TypeScript. Pick it up after college, because college never taught us anything about it. I want someone like that to be able to pick it up and then see the patterns and being able to kind of onboard from there.

• 32:43Speaker 03

  I think what's so interesting, you can look through the AI town code base and use a whole bunch of JavaScript, a bunch of TypeScript in there. And then you look at the interaction points with OpenAI, for example. And it's a sentence. It's an English language sentence that says, you are an agent and your name is blah. And here's some memories. And here's what person, someone just said to you, what should I say next? And it's so interesting that like the, you know, the lingua franca of, you know, interacting with models has become regular English sentences. So fascinating to see that. It's not very often for someone who's been coding a long time just to see an English sentence in the middle of a, you know, a code base that's not that perfectly formed. It's just a regular old sentence, you know?

• 33:25Speaker 02

  Yeah, I, I think there's a very, very subtle and important point in all of this. And Nathan, I think this is the right question that you're asking, which is like, where, where does the programmer end and the model begin? For me, it's been very much a revelation working with these, these models because they're so smart. Part of me is like, why, why, is like, I almost feel like writing code around it is like wrapping an abacus around a supercomputer. I'm like, you're so capable, why don't I just give you a keyboard and a monitor and a mouse? And I'll just tell you in English what to do and you go through it, right? Like there, like there's part of you that when you're working with this, he seems wants to do that, right? In which case formal languages go away. So I'm going to say something that may be a little bit controversial, but like, I, I think I've gotten to a point where I don't believe these models will ever replace formal languages. I just think that, I think that AI time is a great example of that.

• 34:10Speaker 02

  Like anything that has to do with navigating the worlds, you want to abdicate to the model. Interacting with objects, interacting with each other, making decisions, any sort of like in simulation you want it to do. But to think that it could generate the code that, you know, is going to make the trade offs between like scalability and correctness and all the subtlety of building a distributed system is just not designed for that. That's why, I mean, listen, you know, when I first met James, he was a PhD student at MIT working on databases and distributed systems. Like this is some of the most complicated systems in the world and that I think you kind of need a form, that you will always need a formal system where you, you can very clearly articulate the trade offs that you're going to make in order to scale.

• 34:55Speaker 02

  So I don't know where the line comes down. Like, I don't know where the program ends and the model begins, but it feels to me as like creating the laws of physics, you want a formal language and then working within that, that's what the models will do. But I don't think one is going to take over the other. Like, meaning like, like, we're never going to write a program that'll become a model. And I don't think the models are going to ever be able to create the laws of physics because you need strong guarantees.

• 35:18Speaker 03

  I obviously agree because that, that, that's my outlook on the world. Yeah, I've been doing a lot of AI enabled programming lately, just to, just to try it out and it's been helping a lot, especially when I get a new code bases or new languages I'm not that familiar with. And then, so the question is, do I feel like that threatens my job? Like not in the slightest. And you asked a question about the end of the language wars. I think anything that can. Can teach software engineers their job is not being a code monkey, their job is thinking is a good thing, right? And so, yes, it's going to get easier to write code. That's great, right? Is it going to get easier to exhibit judgment about the rest of the world? Maybe not, right? There's always room to think architecturally, to have good design principles,

• 36:02Speaker 03

  to have good common sense, and I think that's always going to show up. And there's always going to be a need for platforms, hey, like Convex and and vector databases. And, you know, I don't think we're at any risk of someone just throwing out databases as a concept and using models instead.

• 36:18Nathan Labenz?

  Yeah, it's interesting. Certainly, we've had, you know, software eating the world and it's taken a good, you know, a few bytes out of it at this point. And then the next thought was AI is going to eat software. And it seems like there's definitely a few bytes to be taken out of it, but the question is kind of like, ultimately, how much, right? I wonder how you guys would think about that. I could imagine like saying, you know, in the future, there's a lot of different measures you could put on it, but it does seem like there probably are fewer lines of code in future applications, right? I mean, just like that one natural language sentence, like that would have been a real mess of heuristics until not long ago. And so it does seem like there's like a some significant shift where code that used to be explicit

• 37:03Nathan Labenz?

  is no longer. But I would also agree that like, we're not getting rid of databases anytime soon. There's some, there's also like a tool use paradigm here where I don't want the language model to be my database. I want it to maybe use my database, though.

• 37:18Speaker 03

  I mean, the abstraction models increase over time, and that's that's just the march of technology. If you're baking a cake, you don't go out and and farm some wheat and and mill the flour, right? You go to the grocery store, even if you're a chef baking a cake. And if you're, if you're building a web application, I'm hoping you're not writing your own database. And that would be very silly and and probably you should be using a managed platform because you get to focus on the part that matters. And as these tools mature, they allow us to not do the things that we don't want to do and to focus on what makes our application or our task uniquely special and to exhibit our common sense and our judgment and and things get done better and they get done faster. And that's that's just a good thing. It's a strictly good thing. I think that's just the march of technology.

• 38:00Nathan Labenz?

  I think Nathan's I seem very or making a.

• 38:03Speaker 02

  I think a very important point for like the future of software here. And I think it's worth kind of just digging in a little bit. These models are so powerful. How much can you advocate programming to them? Right? And here is one person's view, kind of having stared this in the face a bit, which is there's a minimum amount of complexity in a program. Where no matter what you do, it's not going to get any less complex. And that complexity is describing exactly what you want. And that description is a set of trade offs. Right? And it's not about syntactic. It's not about kind of boilerplate code. It's literally like here are the trade offs. Here's what I want things to happen. And you've got two ways of doing that. You could do that in a formal language where it's not ambiguous. So you know what's going to happen.

• 38:47Speaker 02

  Or you could do it in a natural language where it is ambiguous and you don't know what's going to happen. And like in natural languages are naturally, they're known to be ambiguous. So there is no way to describe something and get a predictable output. Right? Like, for example, a dog brought me a ball and I kicked it. Right? You don't know if you kicked the dog or the ball. This is like one of these classic cases. Right? And so, so formal languages are great when it comes to things where you've got to describe clearly trade offs. And I don't think you could even use a natural language. I just, I don't think that's possible. And so I do feel like we're hitting an event horizon where code where there isn't inherent complexity or trade offs. That goes to models.

• 39:28Speaker 02

  That still leaves you with everything where you've got to describe clearly what you want in a set of trade offs. And listen, we're in an infinite world with an infinite amount of work. So that just means that there's an infinite amount of code to write. That's now not, you know, cut and paste. But it doesn't reduce the amount of code there is to write. It just, it just makes it so we're not so annoyed to kind of like, you know, being like, I got, I've written this like the 50th time. Okay, it'd be really interesting. Like, I know you've been kind of working with these models a bunch too. You know, and you've mentioned that it's kind of changed your development style.

• 39:59Speaker 00

  You know, previously while I wrote like kind of, you know, tiny games, it was always kind of hard code the game state into whatever logic you have. Right? Now, it is interesting. I do feel like the the state management, like the state of the game, the state of the app, that layer is different in that if you're in a position now, you don't really, you don't need a trade off for a transactional system. You could give it the control back to the model and then it will do something, you know, magical and different, unpredictable. And that unpredictability is the feature, like for any user we can imagine on AI town, what they love about it, it's unpredictfulness. It could feel dopamine rush, right? Like our other portfolio company was Idogram, who's like, you can like easily, you know,

• 40:44Speaker 00

  typing a sentence, it will generate like actual words and then a really pretty picture. Every time I generate something in the deep of my mind, I'm like, that's a dopamine rush there. It wants you, like you will want to hit the button again and do it again. Game state, application state, it's similar. You really have to find that knob. Where's the right layer to implement this? So you have a bit of things that you can control, but deliver joy. I was like a human. I like to see something that's not exactly what I define, but what's the layer you want to have full control? So transactional system, you don't want the server to go down. You don't want the transactions to fail. So that's a layer I don't think you can, at least today, like give the control to models.

• 41:28Nathan Labenz?

  Two things. I want to maybe spend the balance of our time on kind of other applications. And again, I'm really curious about kind of the value you're personally getting from this, you know, Alice persona. But just to highlight briefly before going deeper on that, this interface, I'm always really interested by the interfaces, the places where the sort of this weird new intelligence that is the like highly general large language model, you know, kind of flabby in its way, like super vast and diffuse, then has to somehow actually connect to the real world. And that could be a real world of like hard computing or, you know, even obviously over time, it'll be more and more the real physical world as well.

• 42:09Nathan Labenz?

  There's a really good interface point in this like shared runtime on a platform like Convex that handles all of the like hard stuff in a way that you can kind of represent elegantly so that you can then have the all the interesting interactions, either you with the language model or you know, you and a bunch of language models or a bunch of people and a bunch of language models. And that's, I guess, where we go next. But this this space that's kind of that that shared runtime is a really interesting interface where all these, you know, kind of forces can come together, but also have the like clear, you know, well-defined place to save stuff

• 42:52Nathan Labenz?

  and execute transactions and, you know, do all these kind of fundamentals.

• 42:55Speaker 02

  I mean, what I mean, what way to think of it is, is like, let's say I wanted James to go into my house and build a bookcase. The the reality is I would want to like give those instructions to James in a formal language so he doesn't fuck it up.

• 43:09Nathan Labenz?

  No, I'd do an amazing job. Yeah.

• 43:12Speaker 02

  No, right. No, that's what I would want, right? Because it's like, there's no way in English that you could actually like, you know, it's too hard. Like, that's not what natural languages are for. They're just not. They're just not further. They just don't have it, right? And so unfortunately, James doesn't like, well, maybe James does. Like, most people, you you can't kind of like give like a formal language to and have them execute. The great thing about things like AI Town and the great things about these models is you have both, right? You have the ability to use a formal language to define the laws of physics and define the things where you know specifically what you want and you know specifically the trade off you want. And then within that, you have these models where you get,

• 43:53Speaker 02

  and I think Yoko said it so beautifully, that you've got this kind of these creative, independent elements that will surprise you continuously. Like, even when I'm programming against them, I'm constantly surprised. I'm like, you said what? Like, you did what? Like, you know, they really are these kind of intelligent beings. And so, you know, again, you know, AI Town was Yoko's passion project and then Convex, you know, jumped in to help out, which is fantastic. And then I've basically done very little. But like, you know, it is a framework for playing around with this. For people that are interested, I would just recommend them just like, go ahead and get pull and see how these these two paradigms match. So I actually think for, you know, I've worked on a number of these projects. I think this is the best example of the full autonomy of an AI.

• 44:36Speaker 02

  And a very formal structured system with strong guarantees interacting.

• 44:41Speaker 03

  Martin said something that really reminded me of just engineering in general and construction projects. And it's the idea of kind of compounding approximations. So anyone who builds any like serious engineering project knows that everything's approximate, right? And there's certain times you need things to be exact, like it's a bearing and it has to be the exact right place. And there's parts that just can be approximate. Are you showing up to work at a certain time and stuff? So when you build, for people who are building applications, need to think about these compounding approximations, right? There's certain parts of applications where approximate answers are great. And certain answers where you know, yes, I really need an F sync that has to get stored to an application.

• 45:22Speaker 03

  And it's not that hard, you know, you just go think it through, like what stuff needs to be exact? Okay, I'm going to use a formally specified platform here. I'm going to have an if statement, I'm going to have a code block. Which stuff is going to be great to be approximate? Yeah, use a model for that, why not?

• 45:39Speaker 00

  I think Martin is too humble, so I really have to brag on his behalf. But honestly, like, kind of like Martin said, this is our passion project. We didn't have an agenda over it. It was just, we're a bunch of programmers, we just thought this is fun to build and then we keep building on it. If you, you know, join our Discord, you'll see Martin is every night kind of coding up this next feature, which is super cool. That allows any user to kind of create their own world visually using tile masks, like drag and drop tiles and create your own layout of the world. So you can later run agents on top. I don't have an answer for like, what's a strategy? Cause there's no strategy, we're just a bunch of programmers and nerds who, you know, tinker with these kind of things over the weekend.

• 46:21Speaker 00

  But now kind of we've got so much support from the community. Would you want this to be like a living project we can keep supporting? So at the end of the day, we want to, you know, add more integrations. There's just so much work to be done. And then we want to have the community to be more engaged, you know, open PRs, help us out. At the end of the day, it's an open source. And I think it's just the beauty of having, you know, community of people who are on Discord who never met each other to work together on it.

• 46:48Nathan Labenz?

  So let's take the, you know, last stretch and just inspire people a little bit with your sense of kind of value and other use cases. You know, what sort of things should people have in mind when they hit, you know, the fork button?

• 47:02Speaker 03

  I've got two examples, real quick. One, we have a fork, we have an upstream, it actually hasn't been landed yet, I don't think, with human actors. You can walk around in the game and talk to people. That's cool, you can go and be part of conversations. Now, you might say, okay, the first thing any company is going to think about when you do that is, is someone going to come and spout some offensive content? Right? And are we going to be hosting a demo app that someone's going to come and say something horrendous and try to teach people something awful? And that's the challenge, right? Anytime you build a demo app on behalf of a company, you worry this thing. What you can do is you can just ask the model, is this offensive or not? And it does a pretty good job of this. Now, you might get it slightly wrong, right? Because it's approximate.

• 47:45Speaker 03

  But the best way to know whether someone's typed an offensive message is just to ask the model, hey, is this, is this an appropriate message to have in this system? And it'll tell you reasonable accuracy. This kind of stuff is, I just find it so cool, right? You know, you can build an emergent, like you can just use the model to do the stuff that we would traditionally find very difficult. You know, if it was a couple of years ago and we had to build a system to make sure the content was, was safe, it would just do like a keyword match or something. You're really very difficult and very ineffective. I think people can start thinking about like, how can I do stuff now that would have been really hard to do a few years ago? And a fork couldn't add it.

• 48:26Speaker 00

  Yeah, I don't know if folks here or even the listeners have read this book named The Sea of Tranquility. It's like a, it's from the same author who wrote Station Eleven. It's a great book. I'm not going to spoil the sci-fi book for you, but at the end of the day, the point it tries to drive home is that there are simulations. We may live in the simulation, but at the end of the day, just because it's a simulation doesn't make it less real. You know, like what you did in that world and then the emotions you poured into it. Those are still real experiences. So at the end of the day, what I was thinking in the back of my mind is that I think everyone should have their own world that they run, you know, that they outpour their

• 49:09Speaker 00

  passion and then experiences and memories and, you know, like interactions into. And then the goal of AI towns, at least for me, is for anyone to be able to for care and run their own world and customize and explore. Maybe one day the agents in the world will create another simulation, you know, right now.

• 49:28Nathan Labenz?

  I mean, this, I see just, you know, always with this AI stuff. I see just incredible, exciting potential, you know, whether it's stuff that's like relatively, you know, easy to imagine like, oh, my God, I could have a language learning environment where I could, you know, attend a virtual party in Spanish and like get unrusty in my Spanish skills. Stuff like that sounds fun and, you know, for a lot of people, a lot of little variations on that could be super awesome. I also then do worry about people kind of falling into these simulated environments. And I mean, this is just a, you know, random hobby project on some level. So I don't think you're even, you know, productizing this to the point where you're like responsible for, you know, a community of users. But it does seem like we're headed for, you know, a place where games are going to be

• 50:13Nathan Labenz?

  just like much more easy to lose yourself in. I keep kind of coming back to that getting lost in. It's not necessarily that it's like bad in and of itself, but if you kind of fall into it and, you know, lose track of your normal life, it seems like it can easily get out of hand. And then the fact that, you know, we may be in a simulation is also like super interesting. But I would just love to hear you guys reflect on that. Like what would you, you know, you've spent a lot of time with it. What advice would you give to other developers? What do you think is the responsibility for people that are, you know, kind of whipping up these simulations, especially if they're going to try to turn them into businesses?

• 50:48Speaker 02

  Listen, these models are the largest compute jobs that we've ever run as a race.

• 50:56Speaker 02

  By far, I mean, these are the most sophisticated things that we've ever created. You can explore what that means by trying to interact directly with them. But that's only one modality, right? There's another modality, which you just kind of let them interact with each other. And you let them interact with the world. And I, you know, it's kind of funny because like when it's you interacting, it turns out you're the limiting factor. Like you only can, like, we're not that creative. And so like, you only can say a few things and you know, whatever. But when you have them interact together, the state space that they cover is so potentially enormous. I mean, it's more than the atoms of the universe, right? Like obviously, right? And so like, you fall into an appreciation for like how powerful these are and how large these are.

• 51:40Speaker 02

  And so I think it's too early for us to presuppose where this goes or what the apps are or things like that. And like for companies, that's great and they should do that. You know, my recommendation for anybody's interest in this stuff is view it almost as an exploration of a new life form and go very open minded. Don't worry about all the thought, like they're not going to kill you, they're not going to kill each other. They're not going to like make you a dick, you know, like, like, like, you know, I just feel like we're kind of like, we're almost preemptively like stopping like our ability to wonder of this new thing that we've created. And so, you know, very much for something like AI town, that's very different if you're building software.

• 52:24Speaker 02

  We can have that conversation for like, like a company, but for AI town, which is meant to be, you know, an exploration platform. I would just say go in with an open mind and, and realize that you're seeing something that we as, I think is a race has never seen before. And then you'll get an appreciation for how I use these things. And then just like the advent of the internet, like we had no idea what it would be used for, what companies would emerge. Like we didn't know about Amazon, we didn't know about Yahoo. Like that will all come next, but before any of that comes, I think you just have to understand the technology. And that's kind of the point.

• 52:52Speaker 03

  And this is the, the, the wild west in many respects. This is such early days, right? And this is a really interesting time to go and, and see stuff develop, see ideas originate and be part of it. Yeah. And also go outside. Feel the sunshine.

• 53:09Speaker 00

  I was gonna take a different position on this. That is, I do think at the end of the day, it comes down to what's your definition of for games? Like, life to some people, I mean, at the end of the day, you're looking for things you're passionate about. Like, for example, in a part-time cartoonist, when I draw cartoons, I draw the tablet, right? And that's a form of game for me, and I easily fall into that world. You know, when TV came out, people are like, oh, this is gonna ruin everyone. People are just gonna stare at the screen. But that, at the end of the day, I think it's a net value add. Same for novels. You know, when people are like, you know, back in the days, people were like, oh, if you keep reading novels, you're gonna fall into a different world, you're not gonna think about anything, just gonna keep reading it, and you create this whole new world in your mind.

• 53:51Speaker 00

  I do think games, at the end of the day, are similar in that it's not all or nothing, it is just a different experience. And that's very much like any other experience you have in life.

• 54:02Speaker 03

  To reiterate, AI Town is not a, it's not a project that's kind of owned by us. It's a community project, and the whole point of this is to inspire people to go build. And we're gonna keep making improvements and simplify things and making it more clonable and all this kind of stuff. But the whole point is to get people building stuff and get people inspired. And so if people come out of this podcast and they're inspired to build some applications and see for themselves, I think that's an amazing outcome.

• 54:29Nathan Labenz?

  Yoko Lee, Martine Casada from A16Z, James Cowling from Convex. Thank you for being part of the cognitive revolution. It is both energizing and enlightening to hear why people listen and learn what they value about the show. So please, don't hesitate to reach out via email at [email protected], or you can DM me on the social media platform of your choice.

• 54:52Speaker 04

  Omnikie uses generative AI to enable you to launch hundreds of thousands of ad iterations that actually work, customized across all platforms with a click of a button. I believe in Omnikie so much that I invested in it, and I recommend you use it too. Use Cogrev to get a 10% discount.