ETHGlobal Taipei Opening Ceremonies 2025 I Kartik Talwar & Vitalik Buterin

[Music] Good evening everybody. My name is Cartik. I'm one of the co-founders of ETH Global and I welcome all of you to ETH Global Taipei. Yes, this is not a conference. This is a hackathon by ETH Global.

We're known for running in-person and async events to get more and more developers into the web 3 ecosystem. And what makes this weekend unique is that this is a hackathon focused on building. The thing that's interesting about this setup is a few values that we like to bring on to everybody attending these events. The focus here for all the attendees is to be on builders, which means anybody in the room here is here to build something that they're excited about and they're passionate about. And we want to make sure we can help them get to the finish line.

This is an opportunity for people to learn new things, for people to get mentorship, and just grow as developers and builders. And the most important part that makes these events unique is that we believe showing is way more important than telling. Which means you're not allowed to just talk about an idea. We want to see things work. We want to see actual code.

We want to see demos. And we want to see people talk to you about how you can actually go from an idea to execute it. And that's kind of how how we're structuring this whole weekend. We started this as an experiment seven and a half years ago in a small town called Waterlue in Canada and since then it's only grown bigger and bigger as a massive movement of getting more and more developers into the web3 ecosystem. We've been able to go across six different continents and brought this experience and got hundreds of thousands of developers, not thous hundreds, just over a hundred thousand developers into this space.

And last year was a blast where we did a whole tour to a lot more new places than we've ever done before. And we started by going to London, went to Sydney as a further event in Australia, Brussels with ETCC in in the summer. Uh we did a massive event in Singapore, San Francisco, and the massive end to our year was Bangkok with Devcon. And now we're here in Taipei. This is our first in-person hackathon of 2025.

We got five more to go and a lot more async ones. And now also our first time in Taipei. So we're super excited to meet all of you, welcome all of you in this community. And the best part is that there's so many cool things that are a first for us for this event. The thing I want to check here is I want to see who here has been to an ETH global hackathon before.

Nice. See how not loud that room was. That's awesome because our goal here is to get existing developers from Web 2 to come to web 3. And there's over 700 attendees here this weekend. 550 are hackers.

43 countries are being represented. And the most interesting part is that 42% of them are existing web 2 developers trying Web 3 for the first time this weekend. And that's incredible. Like give yourselves a massive round of applause. We of course have representation from 40 different countries.

But another interesting fact which may not be as a big surprise that this is the first event in four years where we have five instead of six continents being represented. That's kind of cool because we got so many people from just nearby to come in and it really kind of brings in the local communities together and this is kind of the goal here which is it's actually big enough of a movement that we have as Ethereum enough of a footprint to not require people to go all across the world and and that really no spots in South America is actually a quite wonderful thing. But of course that means you have a lot more people from the rest of Asia. So many countries, so many nearby Southeast Asian representation and uh it's going to be even more awesome to see what all of you are excited about and all of you also who came into Bangkok. I'm happy to see a lot of familiar faces.

Not only just the hackers, we also have 98 ecosystem partners. These are founders, builders, developers, engineers, devil engineers, product managers, designers from all of these different companies that are here to help you su be successful and 15 dedicated mentors who are going to be here 247 to help you debug any of your projects and help you with any of your technical questions. There have been 13 workshops here that have already happened and are available for you to watch on demand. And because this is not a conference, there are exactly zero talks. Our goal here is to give you the entire weekend to build.

And that's what we want to do to help you be successful and go from an idea to something real in just a short amount of time. The best part of all of this is that there's going to be over $200,000 US of prizes to be won. So, you can win a lot of cool stuff by just integrating a lot of different protocols. We're going to thank so many incredible companies that are helping make that possible. So, please join me in giving a massive round of applause to our 18 incredible partners.

Let's start clapping. And I want to thank 1 in Worlds, Cello, Circle, Citria, Curve, ENS, Flow, Hash Keychain, Hyperlane, Noted, Polygon, Rootstock, Self, Metal, Saga, Uniswa Foundation, and Zirket. All of them are here in the building. You can go talk to them. You can find them.

You can reach them on Discord. A lot of their employees who are not physically here are also available to answer your questions directly on on Discord. And we want to make sure that you are getting the help you need to make your ideas a reality. So talk to these projects and don't reinvent the wheel if you can build something quickly with the SDKs that they already make it easy for you to integrate. But the next 10 minutes I want to cover some important things on how this weekend is structured and then we're going to go on with the rest of the schedule.

Uh we're going to talk about all the venue related things, all the activities we have planned, how you can get help, workshops and talks, just how to catch up on that. If you're still looking for a team, we have good news for you there. We're going to especially talk about how judging and project submission works, which is a very important thing that you need to know. And of course, there's going to be a keynote by the founder of Bitcoin magazine, Vitalik Bdderin. Let's go into the important things here.

Internet. We've been doing these things for a lot of long lot a lot of years. And uh the most common advice we can give you to have a great experience here is do not use hotspots. They interfere with the Wi-Fi bands and it makes it harder for all of you to get good service if somebody's sitting next to you with a hotspot on because it affects everybody in that range with signal interference. The cool thing about this particular event is that we're actually experimenting with Wi-Fi 7 um as a standard which is a very first thing that we're trying for the first time.

It's a very new thing we're trying for the first time and so far it's been pretty awesome because you have a massive range of frequencies you can do bands on and a 3x capacity than before from Wi-Fi 6. And so far I don't see any complaints on Wi-Fi. So let's hope it stays that way when you're actually hacking it in. Um, also we have Ethernet available for you and most tables. We want to make sure that you don't run into just Wi-Fi only issues.

So if you need to plug in, you can do that in most tables. If you need a Ethernet to USB adapter A or C, you can find that in your swag bag. Security. This venue is open 247. And also, if you haven't noticed, this is a public place outside.

Um, want to make sure that all your stuff is safe and everything else stays secure. So, please make sure you have your wristbands and your badge on you at all times. This is why we're also making sure that anybody who's allowed to be in the building is the one who can come in and there are checkpoints at the the doors. And these badges serve as an easy way for everybody to understand who is allowed where. Very simply put, if you don't have your wristband, you will not be allowed to enter the building.

Doesn't matter if your bag's in there. Doesn't matter if you lost your wristband or your friend's in there, you want to do something real quick. None of that counts. We have to make sure that the experience is better for everybody else here. So, please do not use your wristbands.

And they are waterproof. So, don't worry about taking a shower with them. And we really encourage you to take a shower in general. So, keep the wristbands until Sunday. And because this is also 247, please do not and open, please do not leave your belongings uh unattended.

That's it. I don't know. I don't have too much to add on that one. And don't make a mess. This is a really important and incredible building.

We want to make sure we keep it in its pristine condition. We do not want to do any damage. We don't want to make it dirty. And if you are eating snacks or drinks or whatever, please try to find the nearest garbage can and dispose those items. And let's keep the building the way we found it.

And uh make sure that we respect this uh this cultural building. On to other stuff about the venue and activities. There are three warehouses that we have scheduled for you for this whole weekend. This is number four where the ceremonies are going to be. This is also where we're going to come back on Sunday.

The other two is the hacking spot. And across the street is where your uh food stations are and that's also where judging is going to be. And in case you haven't found where the bathrooms are, you just have to go outside and take a right turn. It's right next to the last building. So the bathrooms are outside, not inside the the halls.

We're going to make sure you're all well fed so you can continue working on your projects. You've already seen dinner happen. There's going to be a midnight snack conveniently 11:30. I know. And tomorrow is going to be brunch, dinner, and another midnight snack.

And Sunday will be breakfast and lunch. And then we're going to do some small cocktail hour at the end of the hackathon. And there's going to be coffee, snacks, and just other hot and cold beverages available to you at all times. So you can go at the end of any of those snack stations and make sure you are well fed. There's also other cool stuff that our operations team has done and that's offer a massive and incredible really cool selection of local food that all of you can enjoy.

So this is what today's midnight snack looks like. I'll just let you read it because I don't want to butcher this. And this is what tomorrow is going to look like. So hopefully all of you stick around for a few more hours and really enjoy all these cool things. Not only that, uh there's also come cool things we have planned for you.

There's a photo booth that you can take selfies on. It's also digital, so keep that uh memory of being here at the hackathon. We're going to do some Tai Chi sessions tomorrow. That's not the real guy. I checked.

There's going to be two of them. And then we're also going to be doing a a bubble tea station, which is coming from that's the thing that gets a clap. I mean, to be fair, this is this is the right thing to clap. So, that's tomorrow from 3:00 to 5:00. So, make sure you're in the building.

And if you haven't already picked up your swag bag, there's an awesome collection of cool things that we have for every single person here, which includes a shirt, this amazing back, the sticker sheet, reusable utensils, a bubble tea holder for tomorrow. a tissue pack, a phone strap, and pineapple cake. A lot of lot you need instructions. That's a good point. We'll uh we'll post some videos for all of you to enjoy on how to consume the pineapple cake and the bubble tea holder and how to open the tissue pack.

I think the utensils do not need instructions. The t-shirt, too. Yes. in case you put it on backwards. We don't want that.

But pick it up if you haven't gotten your swag back. I like being heckled. This is interesting. This is a new one. First event of the year.

Got to cut me some slack. Jordan, come on. Communication. Uh, everything is going to be on Discord. There's a lot of you here.

It's very hard to shout everything and make sure it gets to every person. So, we're going to streamline this by posting it on a Discord announcements channel, which means anything that you need to know will be posted there first, and then we'll go run around and shout in case you don't get to see Discord in time. Everything here on Discord is also managed specifically for the attendees in this building, which means these permissions are automatically set. Only people in that channel are the ones that are in the building at this event, including our partners. And these are all private channels.

So you get to not worry about will I disturb others or the 50,000 other people in those in the Discord. Nope. It's all about the Taipei attendees only. And this is also where you can find direct access to our mentors and all of our partners. So you can ask them questions.

Um and especially when you don't see them at their tables or it's late at night, you can still ask that question on Discord. We encourage you to do so because chances are someone else from your team who is remote will be able to look at it and address those questions. Everything else is on the website which means what the website says is going to be the source of truth. If something is running late, we update the website first. Then we announce it on Discord and then we go shout and tell you to look at Discord and the website.

So if you have any questions about when this something happened, whatever the website says is going to be the truth. But remember, you're in person, so talk to each other. If you have any questions about how to use a protocol, if you realize how do I use this React thing or I don't know how to solve this bug. Very good chance that the person sitting next to you will actually be able to help you. So, we really encourage you to uh get to know each other as well.

Meet people from other countries, other areas that they're excited about and uh make some new friends. I also want to talk about being onchain. A big part of us bringing you on chain also means that we get to do cool experiments on getting ourselves to be more on chain. And there's two things that we have we want to talk about quickly. One is the NFC bands.

So every wristband here is conveniently a Ethereum wallet. These are special chips that are P256 compatible. And that means you can sign Ethereum transactions with it. You can receive anything on this address. And this is a full-on SDK.

If you go to arx.s.org, you can see the documentation. You can hack on it. You get to keep this.

And there's a lot of cool stuff with this that we want you to use. The other piece that's interesting is that it also has context about who you are as an attendee. So, if you scan someone else's wristband, you get to connect with them directly and it'll keep all that information as a double opt-in. So go to a website on eat global and you get to say I'd like to connect with you and they get to decide what they want to share with you uh directly and you get to get a record of who you met and how you can follow up with people especially our partners or our mentors and all the cool things and the best part is that because all these things are default on chain all these interactions will be published as attestations on EAS and we get to do a record of how things have been evolving for our events. Don't worry, none of your private information is posted, but the general aggregate information about what happened at E Global Taipei will be made available on chain and that's super cool.

So, you can either go to eglobal.com/connect on your phone or just scan any wristband, including your own. It'll take you to this link and you can see the history of what's happened here and find that on chains. The other thing here is hacker pack. This is a custom ERC 721 contract that we've made that lets you claim a lot of cool things onchain and offchain.

Every hacker who is here should already have this by now. If you haven't logged into your dashboard, you will automatically get this. But the cool part of this is that we also want to make development a lot easier for you as a as a hacker. And if you have a hacker pack, you also get access to our developer faucet which lets you get tokens from 13 different chains. And after this event ends, you will be able to automatically get a free ENS domain name, a free lens username and a forecaster account and a lot more perks.

That includes anywhere from access to different conferences to discounts from different products. We want to make sure that if you are a developer who's continuing to build stuff, we get to make that whole journey easy for you. So all this stuff is on our website as well on the navbar eatglobal.com/packs andglobal.comfet.

Now, let's talk about how the hackathon is set up. Hacking officially begins now, which means you can start working on your projects and you have about 36 and a half hours to work on your submission. You can work on a project with up to five members. You can also work on it individually, but the limit for how many people you can have in your team is five. You cannot work on two projects.

You cannot do be listed as member on two teams. You have to pick and choose. and submissions are going to be due at 9:00 am on Sunday. The rules are very simple. Anything that you build this weekend must be done from scratch.

You cannot add a feature to something that you've been working on before. You cannot extend your own site project. You cannot just copy something from the internet and deploy it to a different chain. Those things don't count. You must show proof that you started this thing today.

Of course, we don't want you to start everything from scratch. So, if you have some open source project that you're forking or trying to clone and adding improvements to it, that is allowed. We encourage you to look at existing open source projects, which means anything that's properly licensed, has a lot of other contributors and uh just overall is something that's usually a front end to popular contracts that we've seen or scaffolding code or SDKs from some of your partners. All those things are allowed, but you cannot do something that you've been working on in the past. And of course, if you have not RSVPd, which means if your team members are not properly listed and it doesn't show they're confirmed and checked in on the dashboard, you will not be able to win a prize.

So, please make sure that you are all registered as a team on your hacker dashboards. Of course, that's just some of the info about basic setup, but there's so many of you who are doing things for the first time, and we want to make sure that it's easy for you to get building. There's a lot of content that we have that you can look at. Uh the 13 workshops that just happened are all available on YouTube, eglobal.tv or eglobal, youtube.

com/eatglobal. You can watch them on demand. You can zoom in and out. You can pause when you need to. And the best part is all the speakers are still in the building.

So, if you have a follow-up question, go find them at their table and just ask the question directly. Not only that, we have 2500 videos on our YouTube on how to use every single protocol that's out there that's ever been part of our events. And you can literally learn how to use anything that you can imagine at this point and find a 30-minute tutorial on it. And we encourage you to just literally Google things on a protocol and put a global next to it. very high chance that you'll find the exact tutorial you're looking for.

So if you have any questions about how to use DevTools, Foundry, VM, five other tools in there, scaffolded, all those things, there are nice tutorials available with code. And if there's anything else that I didn't talk about yet, find me, find anybody on the global team, ask the questions, ask them on Discord, send us an email, ask them on Twitter. We're checking everything. want to make sure we can help you and we will be glad to uh clarify anything that that you uh want to know. Let's talk about being a finalist.

So, at submission time, you'll be asked to make a big choice. Whether you want to be a finalist or you want to just do a normal judging track. What does it mean to be a finalist? Well, we're going to have up to 10 teams that are going to be considered finalists for this hackathon. And these are going to be the most interesting and the best projects from this weekend.

Going on this track does not have any impact on any prizes you can win from our partners. But it does mean that there's a slightly higher set of expectations for you to submit a project that meets a certain level of bare minimum criteria. The finalists get to win a lot more than just the prizes from our partners. And what it means in practice is that to be a finalist, of course, you have to make sure your project is started from scratch. But you must have a properly auditable GitHub repository, which means you cannot just commit two commits with 100,000 lines each and call it done.

You have to be a proper user of source control and git, which means you want to be able to see what changes there were. Try to use pull requests. If you have multiple members, you should all be committing. You should not be uploading zip files. We've seen a whole thing set of things that people have done which will disqualify from being a finalist.

So please make sure you're using Git properly. These are also all automated checks which means that we will flag automatically the second you push a commit to your GitHub repo and point out that you are currently not eligible for this. The project must also be open source and deployed and live in some form. You don't have to buy a domain name. It's fine if it's on some test website where is a very common place people deploy on but it must be accessible externally so somebody can look at it and play with it and actually do this as a judge.

It cannot just be local host and you have to physically be here on Sunday to demo your project to a judge because they will be looking at whether you are one of the best projects from this weekend or not. you get to win the Taipei finalist pack as part of being one of those top 10 teams. What does that actually mean for being this? This means that you get a lot more perks than just those prizes. Uh that includes ETH Global Plus, which comes for a full year.

We're also going to be giving additional $1,000 to these teams. You get a $500 flight credit to the next hackathon you're going to attend. You of course have an onchain proof for being a finalist. We do a lot of events with other founders in the space. So, you get to do direct Q&A with them in in a very short and intimate setting.

And over $15,000 in credits, whether it's AWS or account abstraction related tools or any services that you need from RPC to productivity apps, we get you set up with that. So, if you want to make this a company or more, you are well suited to do so without having to worry about upfront costs. Also, we want to make sure that you get the right help here. Anybody with this vest on is an official mentor. They're not doing construction in this venue.

Don't worry. You can tap any one of them and ask them any questions. They're all technical. Whether it's Solidity, front end, backend, you'll be able to get help that you need and really make sure that you are not stuck. Um, they will also be looking at the Discord channels.

There's also an actual physical station in the next building that's called the mentor station conveniently. You just walk up to them, ask anybody, I need help with X, and they'll be able to help you. Now, we talked about finalist, but let's talk about what judging actually means. At 9:00 a.m.

, you'll be required to submit your project. All of this happens on your dashboard, and you get to see all the questions that you need to fill out ahead of time, so it's not going to be a surprise. We recommend you look at these questions tomorrow afternoon so you can see all the information you need, whether it's screenshots or specific description or specifically what other APIs did you use or other tools. And when you do submissions, you're going to then have to choose the finalist or the partner judging track. Both of these things happen at the same time in parallel from 9:30 a.

m. to 12:30 p.m. So, it's a three-hour window. Everything is happening at the same time.

When you pick a partner judging track, that means all the partners will automatically get a list of your project name and they'll be they'll start looking at your source code, your demo video, the screenshots, the description, all that stuff on their own. You don't necessarily have to go up to them to present. It's nice if you want to. if they have time, they'll be happy to look at this, but you get to put everything forward and they get to see the submission you filled out directly on their dashboards and they will decide on their own if you qualify for one of their prizes, which means that if they have any questions, they'll reach out to you directly and clarify, hey, how did you do this or can you point me to where this is happening? they get to ping you on Discord or email and they'll ask the right questions that they need to make a decision.

The other qu key piece here is that you can only have three selected prizes for this event which means you cannot apply for all of them. You must pick at least or up to three partners that you want to apply for the prizes for. This makes sure that you get to focus on a project and you don't try to just integrate everything and at the end realize that you couldn't get one of them to work. That makes everything fall apart. That's the partner side.

Finalist part is that once you do that, you are already automatically eligible for all the partners, but then you have to go present your project to another set of judges that will determine if you're one of the best teams from this weekend. And that means you'll go there and you'll physically plug in your laptop to one of the judges and they'll get to see a demo and they'll ask you questions. I'm sure you're going to have a lot of questions on this thing. So, we'll clarify a lot of this stuff on email and Discord. But the key thing to remember here is that we can only know that you wanted to apply for a prize if you selected it.

So, please make sure that you don't forget to click on which partner you want to apply for because without this, neither us nor the partner knows that they're expecting you. And then if you're like, how come I didn't win anything? Well, it's because nobody knew that you applied to them. And we don't know how to help simplify that unless you uh select that particular prize. All this also happens in parallel, which means that if you want to be a finalist and you also want to go present to one of the partners, you can go in any order.

is going to be first come first serve which means the second you're in front of the line you have to be ready to present and the partners are going to decide based on the submission not the demo that you show them which means that if you do not do a good job at making the code easy to understand or explaining how something works in the description or not having enough screenshots they're not going to have enough context because they're not going to be able to remember enough in their heads. So they need to always be able to reference your submission page to know what you did. So a lot of stuff has been said, but the thing I want to have you remember is that if you do the finalist judging, you have seven minutes. Four minutes for demo, three minutes for Q&A. And don't do this thing last minute because you'll be required to make a video of your project.

So it takes at least 30 minutes to make a 4-minute video. It's crazy, I know, but you want to make sure that you are being very precise with what you're doing, why, how it works, and try to fit all of that in a 4-minute video. So, try to get ahead of all of this and do this thing before or around 8 a.m. on Sunday.

So, in summary, make your submission good because it only helps increases increase your chances of building a great project and winning any prizes. And it's also a great memorabilia for others to look at because anybody else on the internet will look at and see how high quality of a project submission there was and the time and effort that you as a team put in to make this incredible. And of course this is an inerson event. There's people from all over the world. We want to make sure that everybody kind of follows the same rules.

Everything that you need to know is on eglobal.com/ruules. The gist of it is please be respectful. We don't want to have any situation where somebody feels unsafe and harassment in any form will not be tolerated. And a simple rule as a key point to take away is that anything that you build here is going to be owned by you as a team collectively.

Which means neither Eve global, none of the partners nor an individual on your team has the IP rights to the work. If five of you built a project, you are all equal owners of it and you can do whatever you want with it, whether it's commercialization or something else. you have full authority over your work. And if you have any other questions, find us and we'll answer them for you. Those are all the logistics.

We know you're all waiting for the keynote. But what we're going to do is we're going to bring on two incredible partners here to tell you about who they are, what they're excited about, and how you can get help from them for this hackathon. So, we're going to bring him on, and I'm going to start things off with Matt to tell you all about One inch. So, please give Matt a big round of applause. Hello.

Hello everyone. I'm Matt and I'm a core contributor at 1 inch. And for those of you who don't know, 1 inch was actually started at hackathon back in 2019. And our two co-founders, Anton Bukov and Sergey Coons, stayed up for I think it was like 40 hours straight coding out the first iteration of 1 in. And since then it's uh it's grown pretty substantially.

Uh we've processed over 600 billion in total volume and have over 150 core contributors around the world working on this project. So it just goes to show that what you build here this weekend um if you have the right idea could actually turn into the next big unicorn. And there's some serious potential here. Um so bounties this weekend we have three different bounties. The first one is the Fusion Plus integration.

And Fusion Plus is our latest protocol. It's intentbased crosschain autonom atomic swaps. And it's kind of like a mouthful, but basically it will allow your users to have a really nice experience swapping on chain. Um, they won't have to worry about MEV. They won't have to pay native tokens for gas fees.

And also, yeah, they can swap while maintaining custody of their tokens. So that's the best part. Our second bounty is our dev portal APIs. Um, actually, wait, is that it? Oh, yeah.

Our dev portal APIs. We have a whole suite of APIs on our dev portal to access onchain data and we have anything from like a web 3 RPC API. We have token balances. Um, you can get NFT data. So, whatever can help you with your DAP, you can use that and qualify for that bounty.

And finally, we have our portfolio API. And this is really cool because it allows you to track the wallet performance uh for your users. So you can track P&L um and just the entire history of like profit and loss. So you can let your users see like how much money they've been winning or losing trading their meme coins. And yeah, so that's our three bounties.

And if you have more information or want more information, go ahead and scan this QR code and that'll take you to our bounty page. And we'll also be at the booth all weekend. So feel free to stop by with any questions. And yeah, happy hacking. Thanks, Matt.

Next up, I want to bring on Ian to tell you about World. Please give Ian a big round of applause. Thank you, Cartik. So, I'm Ian Dillik. I lead developer relations at the World Foundation, and we are here to answer this one very big important question.

How do you know who or what is a real and unique person online? You use World ID. World ID is our proof of personhood protocol that lets you prove totally anonymously that you are a real person doing something only once without revealing to anybody, not even to world, which unique human you are. This is especially important as AI gets better and better at well everything, but especially the tools we use for this today, like captures. The entry point to world ID is this fancy thing, the orb.

The orb's job is to actually verify that you're a real human. Takes a photo of your iris. Uses the pattern of that to check and make sure you're unique. And it does this really privately using a bunch of fancy cryptography that I could spend hours rambling about, but I only have another minute up here. So, if you want to learn more, can ask at the booth.

World is built to be privacy first, decentralized, and open source. And there's a bunch of different use cases for this sort of tech. I don't need to read all of these to you. Customer incentives is one of my personal favorites. I think we could be the thing that brings back the Netflix two week free trial.

I'm constantly pushing for that. And our focus here is mini apps. Mini apps are third party apps built for the world ecosystem used in World App that tie in with the World App wallet and with World ID with an amazing user experience. Right. Pretty much everything that we've built from the developer side can be done in a mini app.

And it will be a better experience for you as a dev and for your users than just about any other way you could build a DAP. And you get distribution to our 25 million users of World App, almost 12 million of which are verified with World ID using New York. We've got some big bounties for building many apps. Thank you. And if you want to get information on everything for our presence here at the hackathon, world.

org25. /type25. Thank you. Thanks, Ian. All right.

Now, we are ready for some really, really interesting points that you're going to learn today. Uh, and that's our next keynote talk. I want to bring on Vitalik to tell you about the future of layer 1. So, please give him a big round of applause. Yeah.

No, um, thank you so much uh to everyone here for coming and definitely thank you cardik for giving me the honor of uh giving the third of zero presentations at ETH global today. Um, so damn it let's uh talk about the layer one. Yeah. Um, so first of all, where is Ethereum in 2025? Right.

I think uh it's important to talk about scale. I mean I I think scale has been probably the biggest thing that we've been concerned about basically since the space started. So Ethereum has about 15 TPS at layer 1. Recently the gas limit was increased from 30 million to 36 million and so that increased the TPS a little bit. We have roughly 250 TPS at layer 2.

So this is uh at least among the layer twos that are using blobs. Uh so actually if you include uh things like stark net it it goes higher there and there's all kinds of constructions that go even higher than that and uh there's large blob count increases that are planned for this year. So the petro hard fork will double the blob space which doubles the amount of TPS that Ethereum can hand out too. And then for Saka which comes after that it's likely to increase that by somewhere between a factor of two and eight um even further and then the plan is to have continuous rapid further increases going from there. Um so a lot of uh progress happening on scale um especially scaling layer 2 um this year and the reason why this is important is basically because there's a lot of both existing and potential applications that really need the space right so uh you know world got up to 10 million users right and that's like impressive and congratulations right but uh it also means that we in Ethereum have to like actually support you Um and for every one of these existing applications, there's also potential applications, right?

There's lots of large uh corporations, um even governments, um all kinds of enterprise users kind of this is the like underwater part of the iceberg, right? People who have not launched an application at all precisely because they know that if they did, they would just fill up Ethereum's current blob count immediately, right? And so like we do just have to increase the amount of space for layer twos and do it quickly. And this is the thing that basically everyone in Ethereum core dev understands as the number one priority. Right.

In addition also EIP7702 which is like a first step toward in protocol account abstraction also enables some really interesting stuff. Uh so in particular 7702 enabled wallets will be able to do unis swap trades with one click or one transaction instead of two. And so Ethereum becomes two times less annoying. Yay. Um and so a lot of things happening, right?

But like this is kind of roughly where we are this year, right? And so I uh at like ETH Taipei a few days ago I basically talked about where we're going to scale L2 further from here, right? And scaling L2 is really important. There's a huge amount of activity there. But I think another question that's worth asking is basically like what to do with the Ethereum layer 1, right?

And basically I think since this whole layer 2 concept started in uh 2019 I mean it really started in 2017. I think there's been this debate of like what is the role that L1 should have right and there is the L1 maximalist position which basically says that L1 does everything that any application wants to do or at least L1 should grow to do anything that any application wants to do. And then there's the like L2 maximalist position which basically says L1 is really there as a settlement layer. It's like this anchor layer and a few hashes go into it. Regular users should never touch layer 1 and it's uh like much more of an institutional thing that layer 2 is hook into.

And I think recently there's a lot of demand for like actually using the layer one more. And uh I mean there's a lot of things that benefit if if the layer one is more powerful, right? And uh I think one of the important things that's valuable to think about is like even in a world where most activity happens on layer 2, there's a lot of value from scaling L1 as well, right? So basically most L1 L2 applications heavily benefit from at least some L1 use at least in uh some cases, right? So first of all, L1 censorship resistance guarantees, right?

So one of the things that the L1 provides censorship resistance. You can reliably get a transaction included and reliably get a transaction included quickly even if um say large block builders are uh are trying to delay your transaction in order to like exploit you in some DeFi protocol, right? And so layer one censorship resistance guarantees are something that needs to be a reality for users. And with layer twos, there are things like force inclusion channels by which they can be. But for those things to actually be useful, like you need layer one to have enough capacity to actually handle these transactions.

Moving assets between layer twos, right? So for large capitalization like things, so like assets like ETH, you could we can totally have a world where average users move with intentbased bridges and like market makers are just doing um actual cross layer two movements. But if you want to move an NFT from one layer two to another, like guess what? Like you can't use market makers to like front another copy of the same NFT for you because well there's only one copy of each of each NFT. mass exits.

So if a layer two breaks or if a layer two takes some governance action that's really unfavorable but users need the ability to move all of their assets out of that layer 2 into other layer 2s on mass. Um also we want the ability to issue assets on layer 1. Also key store wallets. So this is wallets that have updatable logic and they do that like by keeping the logic for ver for how transactions are verified on layer 1 and then synchronously reading that from layer 2's. So that also requires some volume of a layer one transaction.

And finally just like submitting proofs to layer one and submitting state roots to layer one. Right? So even in an L2 centric world there's a lot of these use cases for layer 1. And I have a post from about a month and a half ago where I actually tried to go into the math and see like what level of layer one use is actually justified by these uh use cases. And it turns out that to be comfortable for the long run, like this actually requires scaling the L1 by roughly 10x.

Um, and like I think in general, right, there's this kind of trade-off curve that you get where if you push too hard in the direction of like not using the L1 and and trying to like make everything L2 only, then often like you actually you do end up getting a lot of these L2 layer 2 specific risks, right? I mean I think like I mean Bitcoin and the lending network are probably a more extreme example of this right and uh like there are these problems where I mean like there was like El Salvador tried to onboard the country onto Bitcoin and then ended up know giving people light lightning wallets and then that ended up being custodial and then like it ended up having challenges right and uh I mean like in Ethereum like the the kinds of L2s that you can make are somewhat more powerful but like even still you have like a lot of the same issues, right? And so like the thing that that we'll learn is that well actually you know if you want users to have both a good experience and a an experience that has things like censorship resistance freedom to easily move to another provider openness then like a strong layer one that users can affordably use at least sometimes is like actually a really good like a good ingredient to make that happen, right? So that's why scaling layer one in an L2 centric world. What if we scale layer one further, right?

So there's this uh like philosophical equivalence that you can make between like basically a techniques that are used to scale layer 1 and the techniques that are being used or sorry the techniques that are used to scale layer 2 today and the techniques that can be used to scale layer one. Right? Basically layer 2 is the exploit an asymmetry between production and verification. Right? So if someone makes a block and that block contains 1 billion gas, the amount of gas that's need that's needed to verify it is much much less than a billion.

Right? If you use a sonark, it's 500,000. And then if you use a fraud proof, it's like 20,000 in the normal case and then it goes up to a a few hundred thousand or a few million if someone like actually submits incorrect data and you have to play the fraud proof game. Uh and L1 itself could actually do the same thing, right? like the same kind of techniques where you have more powerful block producers but you have very light verifiers and you have things like bypass channels to ensure censorship resistance like L1 itself could do all of these things right and so this is one of the possible worlds that's being actively considered by some researchers right like basically scaling the L1 much further right and I think one of the questions that's worth thinking about is like basically what would the L2 be doing, right?

And so three answers there. One is hypers scale, right? So basically scaling even further than what you can do with uh like native Ethereum data availability. And so inmax has been doing some interesting stuff with their kind of hybrid rollup plasma thing that also also provides privacy. Um potentially you could go even further in that direction.

Another answer is sequencing, right? So uh Ethereum sl a lot times will never be as fast as layer as layer 2 sl a lot times and even pre-confirmations are never going to fully close that gap right and so sequencing a value ad and then a third uh property that layer twos can provide is uh okay what's actually happening here um the sorry um the okay the slides disappeared. That's fine. Um, okay. So, yeah.

Right. And, uh, okay. Amazing. Um, great. Right.

B, different different VM designs, right? So, experimenting with different programming languages. I mean, like diff like potentially to do with supporting parallelization to do with supporting different kinds of cryptography in a more friendly way. I mean, there's efficiency like there's a lot of reasons to try to do different VMs, right? So like this is sort of one of the like things that's worth exploring but I think like one really important points to make right is that there's a spectrum right like basically there's a spectrum of uh you know using L2 for everything using L1 for everything but then there's like a very realistic world that's in the middle right a very realistic world where L2s do a lot but at the same time L1's also do a lot and or VL1 does does a lot and the L1 does uh a lot more than what than what it's doing today, right?

And in particular, I think also a world where there's like a much better cross L12 communication. And so you can actually have applications that do a better job of kind of mixing and matching L1 and L2 use, right? So L2 is being able to read L1 with things like L1S load. One good example and even like just much faster deposits and withdrawals are one good example. Right?

So one of the big reasons why I think we we need to switch over to using things like ZK rollups for uh validation is that we want layer 2 to to move away from having a 7-day exit window and move to having a 1 hour exit window and then in two years move to having a 12-second exit window, right? And for these things to happen, a lot of work needs to get done. And like I actually think there's a pretty credible road map to having it this year. Um but like there is is like it is possible but it it does actually take quite a bit of work right so this is all on scaling right and so concrete proposals for in 2026 um bas there's this interesting misconception right where like everyone when they see this problem they immediately say oh the EVM is singlethreaded that's bad let's parallelize the EVM right and then it turns out that like about five separate groups started going in this direction and then they realized that well actually the EVM is not the bottleneck, right? And so what is the bottleneck?

It's IO, right? Reading and writing from the state. And so what one of the probably the first item on the list is uh block level access lists. It's where a block can specify what are the things that it reads and writes uh from the state like what are the accounts that it edits, what are the previous values, what are the post values. And what this lets you do is it lets you parallelize IO and then it lets you do EVM execution and all of the other stuff in parallel delay then delayed execution, right?

So executing a block um or one slot after the block appears instead of during the same slot. And this lets us increase the safe amount of time that we can spend executing from maybe 200 milliseconds to like maybe two to four seconds. You can't go all the way to 12 seconds because if you go all the way to 12 seconds, you have no safety buffer and like a node can't actually sync, right? But two to four seconds very possible multi-dimensional gas for call data. So having separate limits for the call data in a block versus the execution in a block and this will just reduce a bunch of worst case situations, make it safer to have large blocks.

Gas repricing um history expiry. So this is one big one. Basically, instead of every node having to store the entire Ethereum history, we want to move toward a world where history gets stored in a peer-to-peer network. And uh I mean, Portal is one of the projects that's working toward this. I think uh there's uh a lot of room to work in trying to improve all of this stuff, but basically like we nodes should not be responsible for each node storing the entire history forever, right?

like it should work like bit torrent where each node stores a few things but then and then the entire network as a whole is like highly resilient and stores the whole data. Also fossil which is a censorship resistance mechanism basically it ensures that um a large number of builders can individually propose transactions that are required to be included in the block. And this is like one of those ways to sort of ensure that the little guys continue to have power and can continue to ensure transactions get included even if block production becomes more centralized. And then also account abstraction, right? So this is basically continuing the trend and of uh like supporting smart wallets, supporting different signature algorithms like making wallets and user accounts more customizable and doing these things at proto um at protocol levels.

later on statelessness. So Ethereum nodes not having to store the state. Um potentially a state expiry. There's about three different versions of this. Also ZKVM, so actually using zero knowledge proofs to verify Ethereum blocks, right?

So basically conclusion so far, right, is like there's a lot of uh pretty cool and pretty fascinating stuff down the pipeline that actually can in like very plausibly increase the L1 gas limit like first by 10x. And like I think we should probably, you know, get to 10x and then stop and then take a breather or verify that like we're happy with the situation and consider any further increases only after that point. Um but like all of the technological bottlenecks that made it hard or that made it unsafe uh to scale to a higher level before they actually are in the process of being solved, right? I mean actually a lot of them have been solved already, right? So if you remember like one of the big ones that we had before is denial of service attacks, right?

Like who here remembers the Shanghai DOSs wars from 2016? Okay, you see a few hands, right? Basically what happened was that there was uh one guy some anon on the internet who kept on finding vulnerabilities in Ethereum clients that caused a few particular blocks like that used a few particular EVM features to take like 50 times longer than regular blocks to be processed. And so they kept on discovering these features and they kept on just like completely like basically causing the chain to slow to a halt. Right?

So for about a month, the gas limit was actually like six times lower than it should be. And like even still, the chain could barely run. And so what we ended up uh like we we ended up going through this like very painstaking process of fixing all of these issues and then there were even other issues that we fixed years later and now today the protocol is much more robust against uh against that sort of stuff, right? So a lot of that kind of progress has been um happening, right? And so basically yeah, you know, all of these upgrades that we're seeing, they're like the culmination of a decade of like actually like five different strands of work that have all been trying to solve all of these fundamental fundamental problems that uh have prevented the L1 scaling uh scale from going higher without compromising security or compromising decentralization, right?

And now like we actually have a pretty uh credible road map by which we can have more scale which means we can have more apps uh on Ethereum and apps can use Ethereum L1 more and transaction fees can be lower for everyone. Um so other than scaling right so account abstraction is one of these really big things right so if anyone is building things here definitely encourage you to look into um EIP7702 which is kind of this first step account abstraction that's uh happening in Peterra and then EIP7701 is uh basically a second step like what it's doing is it's uh kind of more fully abstracting the verification step and so like we can basically we get rid of this whole thing where people like can have to try really hard to make sure that they have the SEC P256 K1 curve or like to to use Ethereum and instead we could just say well for your account you can use any form of security you can like basically bring your own algorithm and uh Ethereum will be able to handle it. This is also a quantum resistance strategy because EOAS are not quantum proof. Um, and uh, I think it's also valuable to make sure this is compatible with censorship resistance stuff that we're doing. Layer one privacy features, I think, is like one of those things to think about.

EVM upgrades are another one of these kind of long-term categories that's good to think about. I mean, both short-term and long term. Protocol cleanup, protocol simplification. Um, a lot of like junk in the protocol just from stuff that we've had for a long time. And so there's a lot of room to just like clean up a lot of things, simplify a lot of things.

Um, changing the serialization, dumping our OP, switching to SS is something that I think we should do eventually. Also VM stuff, right? So if we do do make improvements to the VM, it would be good to like also do something to the existing VM potentially replace it with a smart contract written in the new VM at some point. So a lot of like other things that can happen, right? But but I think scaling is like the number one priority for everyone, right?

And the reason is basically that you know we're finally at the stage where we can build applications that are seeing huge amounts of use and the chain needs to be there to handle that use and the chain needs to be there not just to handle the use of today but also to handle the use that any of these applications would have if they were to actually succeed. Um so also consensus layer right so single slot finality which is basically a rework of the yeah proof ofstake consensus which makes it finalized more quickly it also improves the security properties of the protocol it also makes the protocol simpler also quantum resistance I think uh like we've been really start taking quantum resistance much more seriously in Ethereum and reducing the slot time so researchers have told me that uh we can reduce from 12 seconds to 8 seconds without kind of too much fundamental technical difficulty. But there's like a lot of practical annoyance which is just the number 12 is hardcoded in way too many places and that's like the main actual difficulty. Uh but then also we want to like decrease from 8 seconds to 4 seconds, right? And uh like the each each one of these decreases, you know, like like it provides a lot of value, right?

and that you know we want to make applications that just continue to be faster, continue to be more responsive. I mean even 12 seconds like compared to the blockchains of 5 to 10 years ago it's like already a big improvement, right? And remember when there's a 12 seconds a slot time that means that on average a transaction waits six seconds for the next lot, right? Like 12 basically means a dice roll between zero and 12, right? And so like it's actually like even better, right?

And so four means a dice roll between zero and four. And so we're basically talking about like better UX than credit cards, right? Like if you think about credit cards, you know, you tap the credit card, the wheel spins for a while. Like that like it's actually very far from being instant, right? And so like like we're already pretty close to competitive with that, but like we want to actually be better, right?

And then if there's applications that need even faster confirmations then you know layer 2 is continued to ex continue to exist for that and I think there is going to be value in a layer two is forever right so you know these are some of the kinds of things that are happening right I think uh it's uh it's good it's important to think about these things right because uh you know from an application developer's point of view right like he needs to know that one, you know, the chain will continue to be secure, the chain will continue to be there and two, the chain will actually provide the guarantees that you want and the chain will continue like actually have the scale that you want, right? Like uh what happened in like maybe 2015 to 2017 is like a lot of applications were launched on the basis of very low transaction fees that Ethereum had back in 2015 because nobody was using it. But then in 2017, people started using Ethereum like at at scale. And then in 2019, even more people used Ethereum at scale. And then suddenly a one transaction fees went up to like $50 and then like a lot of people were screwed, right?

And so we need there to be enough space like you don't just wants to, you know, like look at like the the fee charts today and then see, oh look, Ethereum is cheap and so like it's fine. like he wants to also be confident that fees will continue to be cheap in the future because there will continue to be space in the L1 for trans users and also space on the L2 for users in the going forward into the future, right? And like I think we have pro like quite quite solid road road maps for both. And so, you know, now is a a great time to start building and uh deploying applications that can actually go out and uh reach reach the world at scale. And so excited to see what you guys build over the next two days.

Thank you. Okay, fine. We have one talk for this weekend. Uh thank you so much. Uh one of the cool things about these events is that you as developers uh get to see an early look on what the next couple years are going to look like and you get to prototype them in a weekend and uh give people a heads up on what the next six months are going to look like.

So this is an amazing overview of what we think is going to be the next couple years and the message is clear. We have to scale the L1 and that's exactly how we're going to do it. So awesome to see that. And we're going to have a lot more updates on these in the future events, I'm sure. But before we kind of uh get have you all get back to hacking, want to quickly talk about the prizes.

All the information here is going to be up on the website because there's going to be specific things that each different partner will be requesting as a criteria to meet the expectations for qualifying. So there's over $200,000 in prizes are going to be given away. One inch and World are going to be giving 20,000 each in prizes. And then we have Cello, Circle, Citria, Curve Grid, ENS, Flow, Hashkkey Chain, Hyperlane, Noted, Polygon, Rootstock, and Self are giving up to $10,000 in prizes each. And then Metal, Saga, Uniswap Foundation, and Zorit are giving up to $5,000 in prizes each.

You can see all the specifics on what the the categories are, what the first, second, third place for those prizes going to win, what you need to do to qualify, the sample code you can just use to get started, the workshops, the videos, all those things are made available to you on the prize page and all that is available on the website. So, that's on the prize. If you're still looking for a team in workshop room 2, that's just behind you. We're gonna meet in 10 minutes, 15 minutes, and uh have you also meet other members who are looking for a team to and help you uh connect with a few more to uh get on an idea and get started for this weekend. And lastly, we want you to have fun here.

Pace yourself. Don't try to make this a marathon. Try to learn something that you did not know today and make that the goal for Sunday, which is did you end up learning something new? And in that process, can you show somebody something that you uh you made that they might actually find interesting or you find interesting more more importantly and see if you can continue that after Sunday too. So this is very much about experimentation especially for the 42% of you who are going to be building smart contracts for the first time.

Our goal here is to help you get excited about this space and we want to help you get there. So with that, happy hacking. We'll see you all on Sunday. [Music]