Building on Zircuit A Hackathon Introduction I Jan Gorzny

[Music] Uh my name is Yan. I am a co-founder and technical lead at Zirket. I am very happy to be here to tell you a little bit about what we do, a little bit more about the prizes and just you know introduction to to Zirket if you don't know what we are. So without further ado, I will get into it. So welcome to ETH Global NYC 2025 hackathon.

And in this little talk, it's not going to be too much of a hands-on workshop. I do realize it's a workshop, but we're going to talk about a couple things. First, um what we think makes a good project. So, if you're looking for our prizes, um probably want to pay attention. This is in fact the main goal of this presentation to you is to sort of get you the opportunity to maximize your chance of winning.

A little bit about the resources you need from Zirkuit. Pretty simple from a user point of view. So, there's not much there, but I will still cover the details. And finally, what the specific prizes are that we have for this um hackathon. So without further ado, what makes a great project or at least what do I think makes a great project which matters a lot because I'll be judging the circuit bounties or prizes.

Uh I think first of all um and this is a little bit obvious apologies but I think it's worth saying they solve a problem right come to me and say hey we're building this app it's really cool um and we're doing something for some reason we're solving something that maybe other people really care about. So maybe you're making the world a better place. Uh maybe you're solving a market like lack of of solutions. So you know you need to move money or you need to sell something or there's a service. You're building a product essentially.

Maybe you're just solving a technical challenge, right? Maybe uh actually in terms of the uh you know the market, everyone's made enough money, but you really hate hooking up one protocol to another and just like so many calls and and your AI agent can't figure it out. So maybe you're just solving a technical challenge or my personal favorite, maybe you're just solving a problem for yourself, right? Maybe the problem is you're not familiar enough with what is going on in this space and you're here to learn. I think that is a totally acceptable type of problem to solve for um a hackathon like this one.

After all, you only have a couple of days. It's not like you can, you know, build out a massive company. Um so feel free to target problems like this. But either way, identify your problem and figure out what it is you want to solve. Probably you're not going to solve all of these at once.

Maybe you're going to get some experience and try to build a product or maybe you're going to, you know, solve a tech challenge and build your CV. Or maybe you're going to be really good at just building a product. And we'll talk about about how maybe that's just enough at the end. Um, but I think the best projects solve problems that you don't forget about. And I think that's the key takeaway.

I might forget about it. I hope not. I think the really best ones I'll be thinking about for months or years later, but you should not forget what it is. Maybe you've solved, you know, a good product and you want to take it from just a hackathon PC to, you know, a real company. Or you're going to take what you learned here and say, actually, it was really cool that I, you know, moved some NFTs.

What if I did it quicker? What if I did it faster? Or what if I generated a different way or on a different chain or what if I used, you know, the Python or sorry, the PayPal uh stable coin we just heard a lot about instead of, you know, some other thing. you're gonna want to iterate on it and think about it and it's going to live in your mind for a long time because if you don't forget about it, the person who has the the passion for building it, other people probably aren't going to forget about it, too. And if you can convince me to keep thinking about it for years, that's even even better.

Uh, but you should always be building something that solves a problem. And it should not be something that's just going to flare up and then die, right? Maybe next time you won't build on exactly the same codebase, but you'll use the experience you gained here today or over this weekend to figure out how to do it better next time. So solve a problem, but don't overengineer it. There's a lot of buzzwords in tech.

A lot of them are in this room here representing a blockchain that uses AI. That's two of the biggest ones right there. Uh that being said, a lot of solutions are a lot simpler and a lot more basic than um you know, they need to be in terms of some investors, right? AI is really hype. Use it if it makes sense.

But if a regular expression can do the job, use the regular expression. you know, do the other things that are simpler if it means you can get to market or solve your problem much more quickly. You don't need to have an AI everywhere for the sake of AI. Maybe you do if you want to invest, but I I honestly don't think the investors will go for that. Um, and even on things like blockchain, um, admittedly, we see a lot of apps come through um, you know, prize judging and it's like, why are you using a blockchain when you could just use a a database?

People don't like to hear that feedback, but the reality is sometimes blockchain, you know, is not the right fit. So try to really come up with a problem to solve that actually tackles something where you know there's not trust everywhere where a blockchain makes a lot of sense. And so when you're going to present, you know, your your final submission, justify why you're doing something. If you really need something that's really intense or really difficult or you can't use an offshelf solution, figure out why exactly that is. It might be the case that you just can't afford them, right?

might be a solution provider out there that you don't have a account with and you they don't have a trial, you can't do anything about it. Um, but it might also just be the fact that, you know, you could figure out a better way to do it. That's a lot more simpler. And while you're building all of this, you're going to be here for a couple of days. Uh, talk to other people.

I think it's really important you talk to other people. You share your idea. Um, this is not cuz I want your idea to be stolen. In fact, I hope none of that happens. But you should, you know, hear feedback from what people tell you.

Now, it's critical, I think, that you tell people you are and you hear, but you don't necessarily incorporate all of that feedback. Some people will hate your idea and they will be wrong. Others will hate your idea and they will give you good feedback as to why they think that idea is not worth solving or your approach is not good. But listen to them, figure it out and then narrow it down so that whatever you end up solving is solved in the way that you think satisfies the the best sort of set of requirements that you've heard over the couple of days. because you might start with a project which you might see is very technically difficult.

Um you know as a researcher I've seen this lots of times like oh what if we did this what if we did that and I'll think about it in my room for 3 days like it'd be really cool. Okay I'm going to go and build it and I'm going to go tell tell my peers and then within three minutes I'm like actually that's solved it's right over there like ah crap now what do I do with my life? Um you know as a researcher that's annoying because I have to write a new paper or do something different or I have to build a new product. At a hackathon you just move on like cool let's get a new one go for it. So listen, talk, but use your own discretion to um to actually build the thing you want to build because at the end of the day, you are building something for yourself.

Time is limited. If someone does say there are services that can help you build something quicker, use them. It's almost certainly not better to reinvent the wheel, but sometimes, you know, there is a use case. And finally, I think the best projects are built on Zirket. What a surprise.

I work for Zirket, so uh we're here sponsoring it. We're giving out some prizes. Uh Zirk's a great place to launch a web three um project. We're a lawyer to roll up on Ethereum. We're zero knowledge.

a fast finality at the end of the day. But we've got some other cool stuff. We got quicker block times on Ethereum. We're certainly a lot cheaper than Ethereum. We're full EVM compatible.

So everything you can write and deploy on Ethereum itself. You'll have no problem deploying on Circuit. If for anything if for any reason something does go wrong, please come find me. But I'm it's not going to happen. Uh and finally, we have one at least one special feature which is this AI um in our blockchain.

So as we go through building blocks for our layer 2 network, we actually have an AI that determines if transactions are malicious or benign. And what we're doing is trying to prevent all hacks from happening on our network. So if you're building a DeFi protocol and you don't want to get it audited, I think it's bad advice, but if for some reason you don't, um our AI can actually still prevent um people from stealing funds from your users on our chain. Conversely, if you're a user, you can use a D5 protocol without worrying about if it's audited to some extent. It's an AI, so it's got really really good returns, but you know, it's not a not a formal proof and it's not a guarantee.

Uh nonetheless, we'll work with everything to make sure that the hacks on our chain go to zero, and there's none. So um that's the mo the most unique thing that Zirket has. We call this the sequence level of security. There's no way to sort of incorporate it as a developer. So you can't sort of change the rules of the AI or convince it you know in some way that um what you're doing is a hack or not a hack other than just you know doing stuff.

Um it's not a prompt thing. So you can't sort of try to to game that. Uh but nonetheless it's there and it's going to protect people uh your users or you or both. And if you're, you know, interested in trying to figure out how to change that, come talk to me. We can talk on a B2B level.

And of course, there's prizes. That's also why you want to build on circuits. Probably why you're in this room right now. So before I get to the prizes, I will give you a little bit of the technical details. Uh the quick start stuff is not going to be super intense and I will tell you it's, you know, pretty straightforward.

uh as an EVM compatible chain, most of the same sort of tooling and infrastructure you'd use on Ethereum should work out of the box more or less on Ethereum or on Zirketch change. Your best starting place is docs.zet.com. You'll find all the links there that you need.

Um the RPC endpoints for our mainet, our test net, uh hints about our architecture and design if that's important to what you're doing. um probably you know it won't matter that we're a ZK roll up other than sort of speed of finality and it probably won't matter um that we have this SLS unless you're using it to protect your stuff but there's no way to sort of directly talk to it anyways um there's in some cases some very rare limitations if you really want to shove a bunch of hash functions out of the chain the provers might not like that that's going away in a couple of weeks we're upgrading your ZKVM but unfortunately it's not not here just yet nonetheless all the details you need are in docsircuit.com so if you can only remember one URL this is the one you should take Otherwise, connecting to circuit if it's not already in your wallet is as simple as changing an RPC endpoint and changing the chain ID um as well as our um block explorer. We do have a custom block explorer which I'll talk about, but it should be straightforward. Put the RBC endpoint in, change the endpoint, everything should just work out of the box.

Uh and I think in a lot of cases we're already pre um populated in a bunch of things. And if you need these specific details, docs.zcircuit.com/buildoncircuit circuit.com/buildoncircuit um for the prizes and this hackathon you know you're free to use either test net or mainet to submit prizes if you need to verify code on chain obviously test net is cheap and there are faucets I'll talk about that so use that but if you want to go mainet also fine with me the functionality is the same uh if you don't want to block block explorer we have a customuilt one this is largely in part because we have this feature called sequence level security where you know sometimes uh transactions are put into quarantine if they're malicious So, we need to build our own stuff, but it should more or less have all the functionality you need and want from a standard block explorer.

You can verify contracts, which is important if you're submitting prizes. You can, you know, see all the actions that you're taking and all the contracts. Explorer.circuit.com.

It's all linked. Getting on to and off of the network. Pretty straightforward as well. We were built on the OP stack. So, if you have any tooling that's familiar with optack withdrawals and deposits, um, it's going to work.

Just change the address to the ones that we have for Zirket and Zirket testnet. um our test nets on Sapoleia so you can you know work with that you can get faucets I think it's the next slide and otherwise pretty straightforward and there we go the faucets ETH global has one set up that's great on our test net so you can get e um on our Garfield network right there or you can get on and then bridge it over and I encourage you to get the hacker pack in order to get all of the stuff for free I think you can probably find some other stuff it's not super interesting to talk about URLs publicly so uh that's the technical details or that's the sort of the quick our details um covered and I'm making record time here. So uh we'll get into the prizes which I think is the most interesting part. Anyways, first some boilerplate stuff. Uh requirements are I think reasonable.

If you don't think they're reasonable, please tell me why at the end of the talk. Um but essentially we wanted to see some smart contracts and we want them deployed on Zirket. What a surprise. Uh test net or mainet. Testnet's cheaper.

Use testn net. and then you know a GitHub repo that I think all the people will want and it would be nice to see a description of what you're solving, how you solved it, why you chose Zirket, um or how you use Zirket. It might just be as simple as well, we needed a network and we picked yours. Um the description of the team love to learn a little bit about you and what you guys are doing. Uh and uh some other little stuff.

If you want to give us a, you know, a demo or a video, great. we're not going to be necessarily investing in you. So there there's no need for that but it would be nice and um instructions to test it so that we can see that it actually works. In order to win I we do need to see some code deployed with one caveat and that caveat is we have two different categories or prizes um at this event and the first one is best project deployed on circuit. So for deployed on circuit I need to see code on circuit or surprise.

The second one I do not and I'll get into that in just a minute. So best project deployed on circuit is pretty open-ended. this is intentional. We want to see what people can do. Um I think it's a you know pretty standard ask and because we're fully EVM compatible chain we don't want to pigeon pigeon hole ourselves into any particular app category or you know do anything specific.

Um so just build anything make it good make some part of it work depending on the complexity of of the app. You know, not all use cases will be totally finished, but there should be some things that work and it work really well and well enough to demo to us sort of at the end of the the hackathon. Whether or not you do that in person is up to you and your schedule and if we're around, which we should be. Um, but you know, something should work and it shouldn't just be an idea. We do like new new stuff.

It's more exciting to see brand new stuff on the chain. But honestly, if you come to me and just say, "Look, I made, you know, some protocol that we all know and love better, faster, or easier." That's also acceptable as long as there's notable improvement. Um, and you know, there's there's not just forking something from one chain to the other. I don't think that's quite good enough.

It might be if you're the only one who submits, but probably not. On the other hand, we tried this project, this prize, the last um hackathon in Cans, and this one was really, really interesting. We got a lot of feedback on this and I expect to see a lot of interest in this this time. Here we don't even want functional code. This is a little weird for a hackathon.

You probably should be hacking something, but really I just want the best app ID. I'm not going to steal the app ID. I'm not going to go build it myself. But what I want to see is a killer app that will get people who maybe aren't in this building today into blockchain or not. Maybe it's just other blockchain people.

but something that will actually bring people onto let's say Zirket, you know, week after week, day after day or month after month. Um something that might just be, you know, an interesting twist on a stable coin. I'm not really seeing how that could be super impressive, but maybe there's one out there. Uh or something else that will just get people coming and it just happens to use a blockchain under the hood. Whether or not that's implicit, you know, they don't necessarily explicitly sign transactions or something where it's really technical and they have to get in there.

We don't need to see code. I absolutely do not see need to see functional deployed verified smart contracts on circuit or any of the networks related to it. But what I do need is you know some I don't know uh some understanding that you spent some time thinking about it over this week, right? Uh not just like oh it's 20 minutes and I better try for this prize. That's not the goal.

What I want to see is a a clear and detailed problem statement. So what you're trying to solve, who's the market, who's your users, you know, what are you doing? um a compelling solution that uses a blockchain and in this case should be zirkuit um you know why is it you need a blockchain why can't you just do this with a database why you know why is this a web 3 native thing or what advantages do you get you know even if you could use a different one and then a pitch so you know 2 minutes 30 seconds whatever something like this that says why I really think this matters why my users will care and um who's going to use it and ideally you know some other technical stuff like okay I'm going to have a blockchain here and I'm gonna have a microser there that does something. I'm not expecting you to write these things, but just an understanding that if you were to go build this, that you'd have a pretty good idea of what your first couple of weeks or months of building that would look like. Um, and that's a really, you know, big ask, but it's something to convince us that really you've thought about this and you care about it and you think that it's feasible because also, you know, you could say something that's going to be really, really hard, right?

So just coming up with like I'm going to do general AI that like is really good probably isn't a good pitch. But if you apply it in some very specific area and you figure out exactly what's going to change I think that's that's where the value is. I'm going lightning fast so I apologize. But that's the end of the talk. So I'm happy to take questions about circuit which I didn't talk too much about.

Didn't think it was appropriate place to shield too hard and uh about the prizes which I did talk a little bit more about. So I saw two hands at least and I think I saw yours first. No, I mean we're a zero we're a general purpose chain. So there's nothing that's inherently circuit uh specific. That being said, our sequencer level security, this thing that prevents hacks, you know, it's very easy to understand what a hack is when there's financial loss.

So there's a bias towards DeFi in the sense that like we know what it means to have something hacked or like you know, if you steal an NFT, that's that's a clear hack. What does it mean to be a hack on a web 3 game? It's a little tougher. The AI might struggle admittedly because we don't have a lot of training data because there's probably just not a lot of really popular web 3 games. Now, if that hack stole money from the game, we'll catch that.

No problem. Uh but there's there's nothing that's inherently, you know, zirket powered that's super special. uh there's you know we have products being built on top of it from inside and with and from outside people but uh it's more or less just a EVM chain.

Yeah.

Oh uh like overall of my hackathon experience or just like today in general uh I'm sure I have an answer for you, but I I have a hard time. Um,

yeah, I mean, we had uh we had this idea for the the the ideas um tracking last time in France and there were there were three clear winners. One of them was the idea to actually take uh the SP1 prover that we're using behind the scenes in in Zirket and add a co-processor to it and do something. So the killer idea was actually just making Zirkuit better. Uh I don't know if that's got like a great product market fit and I didn't promise them that we'd build that out. Um, and then there were a couple other ones that used account abstraction which were which were very nice, but they just felt like natural extensions of wallets and I like I really like them and I think there's a lot of really good places for a good account abstraction to cut all the bad UX out of this space.

But like I'm also it sort of just feels like par for the course or at least it will be in like a year a year and a half whenever account distraction really becomes like really powerful. And then I think there was one for period tracking for women and I thought that was really really interesting. I think they had a lot of really technical challenges to overcome before they make that a reality. But it was something that like actually that's just not that's not DeFi. That's like oh good you're using a blockchain for something that might matter to a lot of people and they had compelling use cases for exporting the data or keeping it encrypted or doing whatever it is that they wanted to do with this data.

And that was really really exciting because it wasn't just look I'm getting 5% more yield because as much as I love money like someone will figure out that 5% yield and it's not quite as interesting. Personally, I would love to see more games in the space. Um, so if you're looking for ideas, I think the best way to sort of get lots of users on a chain, for example, is to build a game that under the hood uses a blockchain and maybe you give someone a wallet that they can export if they care about their coins, but otherwise you just let it sort of do its thing. And people don't even know they're using a blockchain until suddenly they can, you know, transfer their skins out of the game or their stats or whatever it is that they're doing. And I haven't seen one.

I think the problem is everyone wants to build a blockchain game and so they try really hard to get the blockchain to work and the games are not fun and I think that's the wrong way to approach a game and I could just be, you know, not playing the right game. So maybe there one is out there. I'm not trying to say it doesn't exist, but I would love to see a really fun game where I'm like playing it non-stop and there happens to be a blockchain.

Yeah, go if any any follow-ups. Okay,

I believe we have layer zero. Uh, I have to triple check, but I'm pretty sure we do. I'm on the tech side, so I don't know who our integrations are. I believe for oracles, we have at least Redstone, so they should be there. I don't know what other pricing oracles or data oracles there.

If you go to circuit.com, there is like a logo wall of all of our partners. And so, you know, there should be at least one oracle there and you can figure out how that works or come ask me and I can, you know, figure out how to overcome that. But there should be at least one. Any other questions?

All right, then. I won't stand here. Um, I will be around. Feel free to ask me. Come down to the booth downstairs if you haven't got a Tory cat plushy, please.

Oh, okay. One question. I I'm not sure there is. I you know I wouldn't be surprised if the BD team doesn't you know go there first, but uh go build one then. I would love to see it.

I'd love I'd love to see be a gaming ch. Beautiful. Yeah.

Excellent. Okay. So, I'm going to end it there. Thanks, [Music]