Welcome to Kernel Talk, this is Rose from Kernel Ventures. We are a crypto VC fund focusing on the growth of dev communities and backing the most ingenious founders in the world. Today we are at the StarkWare Sessions in Tel Aviv and we have the Ecosystem Lead of StarkWare - Louis here with us sharing about the conception of Layer 3, the powerful and provable language Cairo as well as its future upgrade plans.
1. Does StarkWare have any plans for further implementation of Layer 3 and if so in what direction? What kind of applications would be a good fit for Layer 3?
Okay, that's a very good question. So StarkWare and Layer 3, to put it simple, we kind of invented Layer 3. I mean, it's a big word, but the first sort of conceptualisation and public discussion about Layer 3 started with a blog post by Eli from StarkWare. So yes, we have very clear plans around L3s. And those plans are still in discussion of how they will incarnate, how they will look like, what they will do. Many companies and projects are coming to us right now about L3s, things like Kakarot, Rádius, Slush or various other projects are interested. Now the use cases are rather plenty, you have to learn L3s is just the best way for you to spin out your app chain. So today, app chain already exists here at StartEx, Sorare, ImmutableX, DiversiFi, dYdX. Apex, all of those are very dedicated app chains. And the idea is that, at some point for them to move to StarkNet Layer 3s. Why? because it's gonna be cheaper. And so the whole idea here is - any application that requires guaranteed throughput and doesn't require composability because it's a payment system, a game, a privacy solution would be eligible.
2. What is the future upgrade plan for Cairo and what new values will it bring to the StarkWare Ecosystem? What are your expectations for the complete version of Cairo?
Which verticals will benefit and grow from it? First of all, those are very good questions and thank you very much. People usually ask me like, rather shallow questions, (we did our research) so you did, it's very impressive already. So okay, the future plans. StarkNet right now is undergoing three major updates. The transition from Cairo 0 to Cairo 1, basically what now we call casm - Cairo Assembly, and to Cairo, which is like the newer version. This is the first step, this change enable us to add something called Sierra. Sierra is the mandatory middle step that is a safe intermediary representation layer. And it is something that is fundamental for the behaviour of StarkNet. Sierra enables us to do things like proving fair transactions, which makes us really resilient to DDoS attack on the network. So there is Cairo 1 the language, Sierra the tool, which is fundamental to Cairo, and enables a network to be decentralised. Sierra also gives you the ability to separate logically what proving mean from what execution means. And so today, what people don't understand when it comes to StarkNet - the biggest blocker is not the prover. Proving is easy. Proving is trivial. What's complicated is execution. What is trivial for Solana or Ethereum, is harder for us. Why? Because our VM is slower when you execute things - it's made for proving not executing. And so Sierra gives you this intermediate step that enables you to both target a nice environment whenever you are proving and also when you want to execute. So just understand for StarkNet, there is not one network of validators, sequencers & provers. It's a network of sequencers and an external network of provers. Sometimes they overlap, and sometimes they don't, and you will learn more about this tomorrow when we talk about the protocol. So Cairo 1 is made in nowadays, and if you look at Cairo 0 today, Cairo 0 could look a bit like C, it's very raw. Everything is linking in a sense that you need to understand what's going on underneath to write a program that is optimised and work. With Cairo 1, it looks like rust, with all the benefit of using a high level language. And so the idea that Eli was talking about before is true today - Cairo is the best language for proving and will be the best language because of these nice dev experience in all the new features we're bringing with Cairo. And finally, Sierra - these building blocks is also going to be integrated into our network in order to do something called Regenesis, which enables us to benefit from all the advantages Cairo 1 brings. So to summarise, there are three things that is coming with Cairo, Cairo the language is going to be rust like, and I believe will be the best possible language, and then you have Sierra that makes us resistant to DDoS attacks and will be pushed into regenesis. And Sierra will also enable us to solve the sequencer, which is our biggest bottleneck now and give us all the throughput we want.
3. When can Cairo be used to write provable programs, such as writing a Layer 3 app?
So tomorrow you're going to hear about how yearn finance is using Cairo to make provable rebalancing of their vote. So you can already do it today or almost today, you can do it with StarkNet, you can do it using recursion using SNARK, people are working on plenty of solutions.
4. How do you aim to build a thriving zkRollup Ecosystem? What makes StarkWare a strong ecosystem?
I mean, your reader won't see this, but you've been here. And you talk to people, and this guy flew from Paris to come. And you can also assume and and, and I think we already got there. So the question is a bit different , the question is, how and why StarkNet have a thriving ecosystem? Because we do at this point. And the reason is, because we thought from the get-go, that ecosystem building or adoption is not something that money can solve, it's a question about making real connections, providing blood and sweat. And show care and love, and show that you want to help the people building to be successful. So if you come here at StarkWare, where most of the vast majority of the company, the users right now are fresh entrepreneurs, one thing they share is a curious mind that is willing to try new things and not just copying Uniswap and make weird shit. So there's a StarkNet slogan which kind of came along randomly - Keep StarkNet Strangest. It's a bit of a cliché, but just do useless cool stuff and be surprised, most of the time useless stuff turn out to be useful.
5. Is there any focus on use cases or verticals that the Starknet Ecosystem is particularly interested in building? For example DeFi, gaming, or social? What support can developers receive from the ecosystem?
Great question. No, there is no vertical people do random stuff, they do strange things and that's actually why the gaming ecosystem is thriving on StarkNet because it's fucking weird. And there is something beautiful about trying new things. And exactly you have organic growth from bottom up, so the support we give is tiny. The only thing I can tell them is you can reach out to me anytime of the day. I will help you. I will do my fucking best to connect you to the right people, to invite you to the event where they can inform you. I'm your friend.
6. How many developers are there currently in the StarkWare ecosystem？And how will StarkWare attract more developers to build?
So you need to think in comparison, according to the Electric Capital Developer report (https://www.developerreport.com), StarkNet has twice as many full time developers than Aptos and Sui, and probably three times more part time developers. We have probably around, from my perspective, developers active in the community, we have 123 people coming to the building, the hacker house, which is by the way, not organised by StarkWare, but by Only Dust a company that does open source funding. And so I think at the committee right now would be 300–400 active devs but you can use the data from the report (376 on the report). And we have the fastest growth (up 214% since 1 year ago and up 2220% since 2 years ago on the report), if you look at the growth, we are growing almost as fast as BNB Chain.
7. There are many native projects on the StarkWare ecosystem, how do you aim to attract projects from other blockchains or networks to expand or migrate to StarkWare?
A very good question. So we weren't ready to do it before. And now we're starting to do it. We did a couple of times with Maker and Aave. Now we're are going to push more. Because Cairo 0 was so rough. You only need people who could eat glass and have no issue doing that. And by the way, no kidding, this is not a joke. I literally said that in Denver last year you know, StarkNet dev needs to like to chew glass. They know the truth and they loved it. If it was easy, it wouldn't be there. Interesting things happen when things are hard, because everyone has the energy and the passion to do it. And that's what makes them stay exactly. It's hard. It's tough. It's rough.
Thank you for sharing all this with us today Louis, what a cool experience to witness first-hand the thriving ecosystem and fast growing community of StarkWare, keep up the great work! And if you like our content, click the subscribe button and stay tuned for more exciting interviews!
Kernel Ventures is a research & dev community driven crypto VC fund with more than 70 early stage investments, focusing on infrastructure, middleware, dApps, especially ZK, Rollup, Modular Blockchain, and verticals that will onboard the next billion of users in crypto such as Account Abstraction, Data Availability, Scalability and etc. For the past seven years, we have committed ourselves to supporting the growth of core dev communities and University Blockchain Associations across the world.