Scaling Ethereum: A Deep Dive into the L2 Landscape

okay so um we'll switch in English so hi everybody hi uh here at the live place and also on Zoom uh welcome to this uh Workshop um we'll talk about scaling ethereum a deep dive into the L2 landscape and uh hopefully by the end of this presentation you'll have more insights of what's happening on L2 but also in the world of startet I'm also excited to be here so thank you for the invitation thank you for having me here and um yeah let's get this started so initially I have a presentation prepared so let me just scroll back to the um start um so as I mentioned I'll be talking I'll be talking about the out to landscape um and then I'll take a deep dive into the differences and nuances between the RO s and finally we will take a look at um starkness specifically so for those who do not know me uh my name is Robert codra um I'm a developer Advocate at the Stark knet Foundation let me just get rid of this so I'm a developer Advocate at the start Foundation uh I've been working there for um roughly one one and a half like one and two months maybe um it's been quite a journey um I'm meeting a lot of developers onboarding them to the web three um and and um so far it's been quite a good experience um in terms of languages I know English Romanian Hungarian I also know German but it's a bit Rusty so I haven't added that there and if you want to reach out to me I have my Twitter telegram feel free to connect with me uh DM me and if you have any questions about web 3 or starnet or anything in general feel free to reach out okay good so the agenda for today um as I mentioned um we will initially take a look a little bit about um ethereum's history so we're going to check a little bit about like the most important things in terms of like um L2 importance as well um then we'll take a deep dive into L2 landscape uh we'll talk about why Stark net and finally I'll leave you with some information on how to kind of like get started uh with uh with stock okay so let's check ethereum's history so um I put here a timeline and um as you may know ethereum started around 2015 um vitalic and other co-founders um created ethereum or um cre ethereum in 2014 and I think officially they launched somewhere in summer in 2015 um vitalic has been uh true OG in the space um and there's um I think he's coming from um the BTC Community was writing like new uh newspapers and things like this and then he decided that we need a better um um blockchain that we are we are able to write um um smart contracts and this is why ethereum was launched you're uh together with solidity which is a smart contract language you're able to create these smart contracts uh that is on blockchain um ethereum is quite powerful um it's decentralized it's secure we've been seeing ethereum since uh 2015 um and uh it's it's uh it's it's very very uh powerful next up um in terms of like 2017 and 2018 uh there was a project called crypto Kitty so I'm curious if anyone knows about crypto kitties right that's good that's good so um initially crypto kitties was uh I think it's the first nft gaming um um project and uh first when they launched they were they had a massive massive influx of U users and players and basically just brought up the um like Network congestion right and uh that's when like ethereum noticed that hey this uh this is currently at this stage that the network is not scalable and what I see network congestion is that I think transaction fees were um several I don't know like $ 450 I don't even more so it was quite expensive to create a transaction and play the game um and in today's society like if you want to play something and to pay like $40 for a transaction is insane so it's quite high so um it kind of like brought to the attention that uh there's work to be done in terms of scaling and making it more efficient and what happened next is in 2019 and 2020 um there were um kind of like uh experimental projects or um that came um and I I call this the rise of roll because that's when the majority of these rollups such as starnet arbit optimism and Zing started developing these rollups um and uh yeah they they meant to try to scale ethereum and U have a higher trut in 2022 um as you may know the merge happened so um ethereum switched from proof of work to proof of stake um it was uh a difficult process a very difficult process because there were so many uh things to take care of and uh the developers and Engineers of uh ethereum and also like the community managed to do this uh it's it's quite complex work and they successfully managed to switch from proof of work to proof of stake and this basically changed um uh ethereum per se um at one point it was deflationary I think I've checked today and uh currently it's like 0.3% uh inflationary but uh yeah it reduced from like if it would be with proof of work I think it will be somewhere around 3.8% of um being inflationary compared to 0.3 um next up in 2022 and 2023 um we see some breakthroughs for layer twos um and some um maturation or maturity um we see that most of the Els um um are being launched so we have um ZK sync Stark not being launched optimism I believe they already launched with one year prior uh but what's interesting is that we noticed that there's significant growth in tvl total value locked and also users so uh this is kind of like the the the l2s start to um take um Effectiveness overall the network uh just for the public all the l2s majority of the l2s are build on top of ethereum uh we do share the same goal we are trying to um help ethereum scale ethereum and uh offer um users uh cheap fees and uh more features uh we will talk about those in just a sec and finally I think one important um history or like important upgrade for um l2s in general for rollups is the EIP 4844 also called Proto dang sharding um which basically introduced The Blob data for rollups and uh it made it uh made it more cheaper for uh rollups to um um have the data on on on um ethereum okay um I never mentioned regarding questions feel free to interrupt me at any point if you have any questions and we'll be able to take them so whenever you have something uh feel free to ask me okay going next we have the rise of layer twos so initially as I mentioned 2019 they started as experimental projects um it was very difficult to build these L uh mostly because um well some of them because some of them are evm equivalent and it's easier to develop but then some of them are non I mean evm compatible which there's some nuances and differences and it's more easy uh difficult to develop on them um today later Tuesday have reached key de decentralization Milestone so we noticed that I think last year and this year all the layer tws are focusing on decentralizing their stack especially their sequencer and their Network um current they are securing more than 43 billion dollar in tvl and they manag to scale ethereum by a factor of 14th uh so just to put this into perspective um I think overall like per day ethereum does like 1.2 million transactions uh and uh all the rollups um overall they do um 17 or something like that 70 million transactions per day um so we notice that uh they are managing to scale ethereum and this is what we want to see um the transaction fees are also cheaper um and uh yeah going next I do want to just touch upon the L2 landscape and specifically we'll look at um two of them so those are the optimistic rollups and Z rollups um does anyone know the like the differences between them or know about these these two rollups okay we see okay so just going to go through them a bit quickly then if everybody's familiar so I want to look at them at from six different point of views so the first one is security um and how they work so optimistic rups are based on fraud proof system so what it means is that um initially like in it comes from the name uh we optimistically um take everything for granted like all the transactions are um optimistically okay until proven otherwise so um the frut system works like um everything is so all the transactions are badged are are are posted on ethereum as an assertion and then for for a certain amount of time it's going to be in a pending State until um um something like a validator says that this this uh um proof or this this uh transaction is fraudulence so if it's fraudulence then they will reexecute the whole process again and then then uh the the the validator or the one that who mentioned this they're going to be rewarded and um this is how like optimistic rollup work compared to ZK rollups they work on Z zero knowledge proofs um I do like to call them validity proofs because not they're not technically Z zero knowledge so um but how they work is that they they have the execution of a transaction or a program or whatever you have um and then besides that you get the result but you also generate a proof out of out of it and with that proof that proof is being sent to ethereum and then the validators are able to verify through a solidity contract that proof and uh if the proof is valid then everything is okay if the proof is invalid then um U the proof is rejected and then the State updates is also rejected um in terms of computation um optimistic rollups they do have the advantage here because uh there's lower comp ation overhead you don't have to create or generate these proofs while with ZK rollups um there's higher computation because you have to generate these proofs um and then in terms of like develop development ux or developer ux um as I mentioned it's easier to develop on optimism so and we'll look at this in the next slide we'll see the difference between these two um mainly because they're evm equivalent so it's easier easier to um develop with evm equivalency while with ZK rollups they are evm compatible which means that um um they have some compatibility but they're not equivalent like not all the op codes are supported so there's other other things to take into consideration um networks like stet it's uh non evm compatible so it's purely has everything from different um even the uh spark contract language everything like the tech everything is completely different and I'll mention why exactly as well in just a bit in terms of finality this is where we see a huge difference between the two of them so as I mentioned U like optimistic rops they take a certain period and this period is seven days so whenever you want to uh exit the layer to or like optimism to uh ethereum it takes about seven days until you reach finality on layer one finality means that on layer one means that it's uh once it's reach you're not able to change um the uh the the the proof for the outcome or whatever um they they need to work like this because that's how like the the the fraud Pro systems are um are um are working and uh this is a bit of a drawback because again if you withdraw some funds from optimism to ethereum it might take seven days and uh it's a bit uh tedious s to say uh with zobs you can reach near instant finality uh but also it depends on the protocol to kind of like set these um um finalities um um some protocols they have several hours or it really depends on how they they U execute this um in terms of data availability um for optimistic robs it's uh they need to post more data to enable fraud proofs so once something like like a transaction is fraudulence they need more data to actually uh reexecute the whole thing again um and then this in terms means more um um blob space they need uh in terms are like zobs um it can be com the data can be compressed more efficiently um and um they don't need um that much um um data uh scalability again here uh there's interesting difference with optimistic rops it has to be sequential because of the way fraud proof system works but with zq rops you basically can do paralyzation and then um you can do uh um like very efficient proofs like you paralyze the the creation of the proof generation and the whole process can be way faster um it's also like another um scalability for proofs is proof recursion so what you can do is for example you have two blocks that State uh that the output is a certain value you can create a proof from each and then from those two proof you can create another proof which basically you create a proof recursion so um and then you can repeat this uh for a certain amount of times and you can scale the whole process very efficiently um this is what stocket does um um or has been doing and um we'll talk about that in a bit question so far no everybody on the zoom is okay so basically the advantages of the are the fact that it costs more to compute more compute and the fact that I know I might have to switch my smart contract language or not necessarily maybe it's not going to be 100% portable right so the the zik Ops yeah definitely like with the evm compatibility it's it's uh uh trickier but on the long term in terms of scalability zq rops are more efficient than optimistic rollups why why is it trickier um mostly because uh they're not um fully even equivalent like not all the op codes are supported uh mainly because how the the um um the circuits work uh and the the op codes have to be treated differently for example for for Stark net it's uh fully non-compatible meaning that you have whole different smart contract language a whole different stack so you cannot reuse solidity you cannot reuse um um libraries such as stock stock Foundry sorry fundry um you have to create everything from scratch um there there definitely um good reasons why Stark went with full non evm compatibility but uh we we can we can talk about that in a bit Co do you think the op eventually to yeah um I think at one point I I know like uh there's a bunch of conferences and uh sooner or later um the Z roll will will be the dominant Force um they're easier with optimistic rollups it's easier to get started to develop um mainly because you have this advantages uh all these advantages like with the solidity you don't have to um create new tech um and you're able to develop just faster um that's why optimism was launched in 2021 while ZK sync and uh Stark net they took several years to perfect your uh Tech but once uh Tech is being perfected it's more scalable like for example I'll be talking a bit um like the the cool things that we have on starnet uh but the the I think in terms of like through throughput uh you can see like 10,000 transactions per second or even higher uh with the the the good uh the upgraded um Technologies one more question higher computation means higher fees um mainly higher Fe so higher computation means that you need um better um systems like um require system requirements like a computer uh uh you have to use uh because it's like it takes a much more time to generate this proof um so you have like you execute the what happens is that uh you execute the certain block like uh that certain block happens like I don't know like 100 transactions and you want to say uh see what's the uh the difference between the uh the starting point and the end point and then um from out of that you have to generate the proof so there's extra time that you have take to generate the proof extra resources and um this will technically add to a bit more fees what we'll see with rollups is that they can scale so efficiently that the more people use the rollups the cheaper it becomes um and uh you can make the the the the fees much cheaper with proof recursion um and so many other things how about as you said there different is effective like average Sol developer not going back to assembly right um so for example if you develop on ZK sync you it might be easier for you to develop but if you want to develop on Stark net uh you'll have to learn a program language called Koo mainly because um how ZK circuits work um and yeah K's technically tuned and optimized for ZK circuits essentially to to be optimized and Powerful for those okay cool moving to the um kind of like the landscape so this is where you also see the difference between optimistic and zop so you'll see that on the left side we have the um op main net we have arbitrum base blast and 35 more uh optim IC rollups while ZK rollups they're harder to develop the technology is much more complicated and complex and we have starnet ZK sync era scroll polygon zkm and 12 more so we can see the differences like you're able to develop uh easier on optimistic rollups mainly because it's the equivalency but uh there's some certain drawbacks as well like for example like the the um um um finality that you have to wait so you'll have to wait seven days until uh you finality on L1 um scalability you'll probably um um have some issues with scalability in the future while zobs are uh a better es scalability and uh overall like the technology is much more powerful but it's more complex develop okay so just in a nutshell how like the ZK Ops work there are three important phases so there's transaction batching uh which means that um it comes to the from the term roll up right we take a bunch of transactions we roll them up into a one uh one um um transaction or one block and then we process uh these and then we send these uh we'll send these results to the sequencer the sequencer will process this further um it will generate a trace uh and that that certain Trace goes to the approver uh and then the approver will generate the proof out of it so we have the transaction batching so we have the mle a bunch of transaction are coming there they're bundled up together they're sent to the sequencer the sequencer um executes those um proposes the new block goes to the approver the prover generates the proof out of those um transactions on the new proposed block and then once that proof is generated that proof is being sent to ethereum to L1 for a validation so uh there's probably a smart contract uh like a sidity smart contract for all the rollups on ethereum they verify that proof and if the uh proof is valid then the state update happens and everything uh is being updated so in essence just to recap there's two three important phases transaction batching proof generation and then submitting those proofs to L1 once the proofs are submitted to L1 um you have finality so if the what happens is like the op like node operators are picking those uh proofs up uh once the proof is verified you have reach finality and uh that's about it with optimistic rups it does take seven days they're in a pending State and it takes more um each rollup in particular has different ways and mechanism how they work with these so um for Stark net um um there's a sequencer there's various modules so there's a sequencer that takes all of these um um transactions they they batch them together they compute and execute these transactions and generate the trace out of it um this information is sent to the approver it's called sharp um and then the sharp basically creates the proof out of these uh transactions and then U this proof is sent to B one uh which is a solute this SM contract um now in terms of differences um the know the differences between starkk and snarks or raise of hands okay so um Stark is um uh scalable transparent argument of knowledge and snark is sixin non-interactive argument of knowledge so these are the abbreviations right um so the main differences um here between Stark and snark is that first verification for Stark is poly logarithmic time so it takes a little bit more time than snark snark we have um like constant verification and then the proof size for Stark it's much larger than a snark um proof so for Stark size for Stark the proof size is around 400 kilobyte and for snark it's around 288 bytes so technically if we're looking at this we say maybe snark is better because you have constant verification and the proof size is smaller right but there's three more important factors that makes the whole difference and these factors are um The Proving time so the proving time for start is actually a few seconds while on um snark it takes about 10 to 30 seconds so there's a difference of uh uh 10x in terms of scale um in terms of um so proving time is the generation of the book itself in terms of trusted setup so trusted setup it means that um if you want to so for example for a snark you need to have like these uh trusted set up like kzg uh ceremony or um for for zcash I think it's power of tow um you need to do this thrust setup in order to maintain um uh privacy for the proof generation and the thing is that um it consists of initially you have some initial parameters that are being randomly chosen uh you also have a toxic waste that's being used and these all parameters are used for creating that proof however um there's certain caveat so the the system is secure but it's not bullet proof what what I mean by this is that if for example the machine that these uh initial parameters are gen generated is compromised then the whole uh setup is compromised or if the toxic waste U that's used uh for these uh trusted setup is not disposed properly um again it's being compromised so there's very um a lot of things to take care of if it's not properly handled while with stars uh you don't need a trust setup uh that's that's a huge advantage and finally the quantum secure um so Stark is based on hash algorithms and they're Quantum secure while snarks they are not so probably in 10 20 I don't know when exactly will happen um at one point we'll Reach This Quantum issue um and um these these uh this will be a a problem for um blockchain in general but with Stark you have Quantum security so um we we we we can handle that makes the whole system qu SEC no so for Starks it's uh they are based on Hash Hash algorithm uh and it makes it more secure uh with Quantum secure so and because we don't use trusted setup we don't need these initial parameters and the whole whole U um process is safe I was I Wasing about SNS snars because the Tred set up makes the whole system of SNS on to the secur no no no so the TR start with something different that they have to take care um the quantum secure they're based on a different um algorithm um which makes it um not Quantum secure so the the hash um um there's basically the hash families and then Stark is based on those hash families and uh because it's using these hash algorithms uh they are proven to be Quantum secure [Music] towards yeah yeah for example sh 256 or Poseidon you these sessions as well okay so now the question comes like why stet um and starnet has a so as I mentioned it's totally non evm which means that initially like it was a very difficult betat like we we had to um create everything from scratch it was basically like we had to create the the the technology from scratch um the the language itself the community everything so it was quite a difficult bet to make but we uh overall we decided that we focusing on the right technology and make it scalable um and um I didn't want to mention like what what of the reasons why so the first thing we seen this with the crypto kitties so we have the L1 scaling problem so just to give you like an example we have a node operator that's being picked um um to produce a new block and let's say like they you know a bunch of transactions happening uh the the node operator um executes these uh transactions and uh states that the new Global state of this uh um block is 42 right so what happens in ethereum is that other node operators they don't believe you right so they they will reexecute everything from the beginning right and this is how the consensus algorithm work and it's a secure um secure way to do it however um it does impact the overall efficiency of the network so by re executing the whole process again it does take time to reach consensus and then this could uh definitely uh incure some um um um slow but overall like the consensus U on ethereum worked and nothing severe happened so far it's yeah yeah but on Stark knet what happens is that we have this phrase execute once and verify everywhere so uh you can notice even on the screen you have the the block itself U again we have the new state the new Global State 42 however you see those small dots you have an extra computation that happens there right so uh this is what happens on on stet there's extra computation to generate that proof but we uh from this from this new Global State we also have the proof so what happens here we send those proof to uh layer one to ethereum to a Smart contract um um soled smart contract and all the validators will pick that proof up and they will verify the proof itself so when they verify the proof itself if it it will only take a fraction of the computation of the actual execution so this helps to scale the whole network massively and allows allows um uh faster comput faster execution and um scalability in that sense um this is the execution versus verification just to illustrate so for example uh like the sequencer on stet takes U like x amount of time to execute those transaction however the validators the only the the certain uh validators they only have this fraction of the actual um um execution and this this number is poly logarithmic so it will take you uh log Square X to verify these uh proofs and just to illustrate as as well um what we have here is that um with improved recursion uh this can get better even so what happens is that you have as I mentioned you have you can have like let's say um eight different Global like blocks and then you st you create a proof out of those and then from those proofs you can create a recursive proof and then you can recurse again and so on like I think the depth of the recursion it's it's uh it could go a lot like U 60 something or something like that but um yeah it's it's quite efficient and it helps to um make the whole network cheaper uh and um more scalable now there's a pyramid there on the right but none of this wouldn't be possible with Cairo and Cairo is not a city um and also it's terrible for SEO because if you try to Google something on uh on on Google um you might book a trip uh but Kyro basically what it stands for is um so C stands for CPU um air stands for U um arithmetic intermediate representation and O is added as a flare sort to say because otherwise it will be care K so uh that's how K ended up to be become um again terrible for SEO hopefully this will improve in the in the next few months years but uh yeah uh it's not like everything that I mentioned wouldn't be possible with without Koo so what's Koo everyone heard about Koo maybe a raise of hands that's good that's good I'm excited to see that cool so there's a couple of things so first of all enables verific uh verifiable computation so what what does that mean um basically let's say you have a program right um it's a CARE program that executes I don't know like Fibonacci and you want to find out Fibonacci at 16 or whatever um it outputs the result U and you have the result I have no clue what's Fibonacci of 60 maybe 200 something U yeah but on top of that uh K generates a proof right so you have a proof generated with that output and what you can do with uh that is that you have the output but you also have the proof uh that you can send to anyone else and they can verify the proof and they can say indeed the output that you have is the correct output nothing has been changed in the execution nothing has been tempered so you can be sure that uh the result that you get is valid if if if it's invalid it would show in the mathematical proof so there's only like mathematical Pro proofs under the hood there's a bunch of polinomial that works together um and um you get this verifiable computation on top of it it's some kro runs on top of KM so K VM is uh the k v machine um we needed to create a whole new um architecture for for stocket mainly because it has to work with these ZK circuits and km has been specifically been optimized for these ZK circuits I have no clue how to do ZK circuits they're very complicated uh there's a bunch of uh things that you have to um consider but what's nice about kro is that everything is abstracted like as a developer you can code like in simple K language and um everything is being taken care of the compiler you don't have to take care of it the syntax is inspired by rust so maybe some of you developed in Rust okay that's cool so if you know a little bit of rust it will look quite similar um I'll show an example in a bit uh but yeah uh it's in inspired by Rost and it also um inherits the ownership model um and it's strongly tyed it has traits macros so the smart contract language is quite Rob and um to be honest um I believe at one point it started to become even better than solidity because um there's certain things that you can do with uh Koo and offers overall like a better uh developer ux or experience um now the nice thing about kro is that it can be used outside of startet so you don't have to necessarily you want to develop proofs or you want to create proofs um um you can use it anywhere you don't have to settle on starnet starnet is using that um is using kro and you can um use it for starnet but in case you want to create a um an example of a program that you need the the proof out of it and then you can verify this proof somewhere else you can use that so that's that's the ni thing about the KOB that you don't have to stick with stocket and finally as I mentioned you don't need to know about any zero knowledge or zero knowledge circuits everything is abstracted um um from from the developer so um yeah it's it's quite a fun language um I'll I'll show just a quick snippet in just a bit and hopefully um you'll think the same I know like new languages can be scary but uh yeah there's a whole history of Cairo because initially um so starcore is the one that actually created kro and uh just a small parenthesis they initially created KIRO internally um because they they needed to write these polinomial more efficiently they were writing them I think in by hand and to write them efficiently they created Cairo zero that we call it today uh it was a very lowlevel uh programming language um looked like python but you had to take care of a lot of things and it was very difficult as a developer documentation was a little bit hard as well um and what happened is that there was a transition phase to KIRO one which is currently now Kiro um and uh the language is way better uh it has uh this rust alike language and the overall experience improveed quite a lot and that was small parenthesis okay now in terms of finality um so we we can look at um so ethereum usually achieves finality in um six minutes which is quite fast for stocket um so this is we're talking about layer one finality right so uh layer one finality on etherum it's 6 minutes um for starkk net um the finality on layer one is around five hours so the maybe you you you your question is like why it's five hours mainly because there are these recursive proofs that happen um and um to to make it more efficient and more uh like cost effective um there's a bunch of recursiveness happening and then a big big big proof has been sent to ethereum and then that's why it takes so much uh the finality on layer to is 2 seconds so if I send a transaction in the next two seconds reaches finality for layer two and then for optimistic rollups the finality for L1 is roughly a week because they they're this um um pending State until proven otherwise or the the U one we passes isation do uh that's a good question I think um that's a good question so when the centralized Iz ation will come into place uh this might change but um I I think the number will still remain the same um so for example with decentralization a lot of l2s are working on decentralization uh for stockness specifically it will be tender Min and what happens is that the whole stack will be decentralized and uh the finality I think it will be roughly the same or uh yeah I don't think it will change much for for the layer one finality or yeah which one um I don't know that's a good question I haven't checked for other l2s but initially like even like for example starnet could technically have layer one finality in one hour if they want but then the cost of the transaction will be higher so there's some trade-offs that you have to do like you have to like when whenever you're building these systems there's some um architecture choices that you have to pick and then this was would be the most optimal one um there's a like if you want to have finality every let's say block then you're going to have to post uh proof every let's say five seconds but it's going to cost you more in terms of like transaction fees prob Sorry probably yeah yeah you can check on L2 be there a whole section about finality and yeah l2b has a bunch of information about like activity tvl uh TPS stuff like this and yeah okay now regarding TPS so I don't really take TPS as an actual metric because I think it's a really bad metric uh but the whole network likes TPS um so in terms of like Stark Net's performance um I think in October 29 there was a game called flippy flop and um it was a fun game it kind of like through the um like the whole ux with account obstruction um and then an interactive game with um it was I think it was called like players versus bots so the players would have to flip a TI and then the the bot like the evil bot would flip it back so there was a constant battle between the the uh the communities and the actual Bad Evil B but anyway um during that time uh the network reached around 800 57 TPS was a major Improvement for starnet um and I think the the um average uh TPS throughout the day was 110 or something like that so it shows on l2b as well um you can check it by yourself on Voyager and Stark scan these are the two block explorers that we use um and uh yeah uh the the network itself is um scalable um we do have the a bunch of upgrades coming up which will even make this um um metric higher now why I said TPS is not really like I don't like TPS as a good metric is that a transaction for startet could be a multi call so what the multic call is is that I can bundle multiple calls towards a contract and I can make it as a one transaction so it doesn't really show the actual um how do you say metric behind it but we almost reached 900 TPS on mainnet and um in the future upgrades this will be improved even more uh transactions are cheap so I think this was taken in the 2nd of February so roughly a transaction would cost you like 0.

007 um as a user this is really good uh like if you try to whenever like ethereum is congested like I don't know trumpcoin comes in everybody wants to buy it and you want to trade on unit swap it takes you like 20 40 $100 to actually pay for like a swap or something like that um if we want to scale and make it like accessible to everyone we need to reduce these transaction fees and um that's the aim of Stark other rops to make the the whole process um way way cheaper and one thing that I do want to mention also uh it's smart wallet so anyone heard about account obstruction or the concept of smart wallets that's perfect wow I'll let you talk about it or right okay uh so smart wallet so I think so overall like so the issue with ethereum is that um it's the uh traditional e eoc's or externally owned accounts and um what happens is that the the signer and the the the account they're coupled together so with with smart wallets these are decoupled so the signer and the account are decoupled and they're interesting things that you can do so the account can act as a the smart contract itself and then the signer would just use some certain functions from the uh smart contract um to verify the signature or do things like this so what happens here you can create better ux for the users right um what does this mean you can have multi calls so uh if you go on uh Unis swab you have to if you if you're trading the first coin um you have to approve it first so that's one click and then you have to swap it that's another click um so there's like two two step into something that should be just one step right with multicol you can just bundle these together and then the user would just submit one one call then there's Hardware signer which kind of like adds extra security so you can have a hardware signer it could be your phone it could be your laptop you could be anything else um you can have social recovery so so again this is all possible because these are decoupled and then you can create this uh program uh programmable uh Logic on the account and um you can have social recovery so for example Argent has something called like guardian and then if you create an account you lose access to your account you forgot your seat phrase even seat phrase are terrible but U if you forgot your seat phrase and you don't have access to it um you can have your guardian recover the account itself which uh in um like traditional wallet is just terrible if you lose your seat phas you basically lost access to all your funds session Keys uh session keys I'm a big fan of session keys because um I've been a gamer um and um I I don't know if you've seen some of the games uh that every time you do an action you have to sign a transaction and it's so annoying because it's like it just cuts out the fun so you're like super focused you have to sign a transaction you want to kill something you have to send it so session Keys what happens is that it gives you a limited amount of um um scope and permission towards the uh contract or application or da uh and they you create this session and they will be able to sign uh and without interruption you'll be able to play the game so it's quite quite good it could be applied to any different that doesn't have to be games there's pay Master pay Master again it's a cool feature so as a um application I can incur all the fees from the uh from the users the users don't have to take care of like oh I I I don't know let's see if I do a um I don't know like a shop or whatever I don't have to take care of the fees as a user I just want to browse I want to buy and then I don't need to know about like okay this is this amount of gas fees or should I wait maybe a couple of hours to maybe go down so it's quite a nice feature and all of this is enabled by account obstruction and what's what's nice about stock is that this is currently natively supported from Day Zero from Day Zero there was account obstruction and there's a lot of cool features um that we have so for example there's um I think the cartridge wallet that basically it operates on um a pass key so you just create an account a pass key is saved and then um you'll be able to play the game it has pay masters Incorporated has session Keys Incorporated there's a bunch of cool things that you have even achievements so your basically account have like many achievements that you can uh have there's arjin braavos Kepler as well so they all are good uh Wallets on stock um yes all of this is basically so there's a specific interface for uh all of these accounts that have to follow so it's called snip 6 I won't get into um details but you can think of snip as like eips on ethereum um and all of the accounts have to follow this specific interface and uh yeah there there's supports for uh signing transaction for different elliptic curves so um and then on stocket everything is a smart wallet there's no eoas so yeah any questions here cool sorry the session please you mean or invting start investigating how we Al you mean the El c yeah yeah yeah there's also like the forgot the the basically allows you to have a sign like on mobile as well which allows you natively to use them as as wallets uh but yeah you have the the uh standard litic curve you have the the star cytic curve and then you have the uh SEC 256 something like that that's basically for uh Native mobile wallets and what happens is that I as a assigner I can use that and also have it in my um what's it called um secured Enclave for iOS I think I can use that and I can uh sign transactions and process this okay now some exciting updates for this year um there's a bunch of them but uh I just mention some of them so K native um maybe this was all new to you but K native basically will allow us to uh completely bypass the km and execute native CPU op codes uh this will improve like the overall uh performance of the network uh I think the the the me or at least by 2x then there's this um so stew Pro it comes from Stone two um and it became two um what what's currently in production is called Stone one so Stone Pro and that's been in production since like three years now it has been securing uh over three billion tvl of transactions happening um and then other like U um companies are using it um um and the the Stone Pro basically has been in production and quite successful now compared uh to Stew um stew is basically the NextGen um open source prover uh it's based on something called Circle Starks um it's a different field it's M31 field uh it's merman field and they're more efficient than the the field that Stone Pro use Stone Pro uses a Fel uh Fel 250 252 and um just to give you a comparison um the metrics show that Stover is at least 100 uh 1,000 times faster than the uh Stone Pro that we have so uh once this it's already open source you can already create proofs um um there's actually uh recently there was a Workshop given by starware um and it's super fast so there was like uh some some um testing done and uh they were like proving some hashes and the speed of it was quite quite amazing um so this will definitely improve the network overall and we will see this scale the whole process way way faster and then there's other couple of upgrades or updates so mol currently there's no mempool on startet so everything is first in first out but at one point mol will be added a fee Market will be added as well and then um uh this will will basically benefit uh the users over like the experience if you want to process your transaction faster as possible or not yeah so um and also you have two seconds blocks I think currently the the uh block creation is or how how long how much does it take to create a block is roughly 20 to 30 seconds uh which is good but this will be further reduced down to two seconds it will benefit a bunch of apps as well and games as well to toh have this Improvement then there's the decentralization so um we're currently in the decentralization there's four step phase decentralization for starnet the first one was introducing the the um permission staking and delegation uh the next phase I think it will be in Q2 uh which will basically uh mimic um uh validators and then at the end uh in phase four uh the the node operators and the validators will have full responsibility of the whole overall work as I mentioned we will go with the tender me the decentralization process is very complex and um the way it does it it will try to decentralize every bits of the the uh stack itself and um I think for myself client side moving is quite an interesting uh upgrade so client side moving will unlock a lot of interesting things such as the case of I can run my whole program on my machine um I can create privacy it I mean you still can create privacy with this but there's better privacy like that you can create the proof um on your machine you send this proof to the network and then the everything is processed so for privacy you can create interesting applications for games as well you can start unlocking games that are have high FPS or like shooters or something like that because currently if you develop a game on like any blockchain you still wait for this uh uh block confirmation and it's like every two seconds it's very very slow let's say say right so you can in terms of games you can think of like you play the whole let's say game as a level and once you finish that level you send this proof to the network uh if the proof is valid if you haven't cheated the game or whatever then the rewards unlocked so um definitely a bunch of new interesting things are coming the uh is the client required have I know some minimal system requirements to run gr complex mean because one of the disadvantages of uh SCS is that you have to consume lots of compute yeah okay so but if if some of the computation is done locally on the client and it's probably pushed and just aggregated by the sequencer and so on still comp computed on the client side so it means that I should have some computation power yeah so which Ste actually this is uh like improv like it's Ste Pro is being optimized for like creating like generating these proves and um it uh like I I think like they're able to like create like proof um I don't know how much was in in terms of metrix but a lot of like um um anyway I'm forgetting the metrix but overall like what what matters is that um the Stover and circular Starks are way more efficient it's like 1,000 times better than the Stone Pro that we have and you'll be able to like create these proof efficieny so I think like even like a normal machine of like normal requirements would help I don't know exactly what what would be but the aim is to be able to generate them on the mobile as well like or mobile phone or something like smaller devices as well um it will be quite powerful like it's it's quite powerful and um yeah I have I have somewhere a slide with how many Blake hashes have improve I think there's like 600 something thousand Blake hashes uh in one second which is quite fast and it was done on like an M2 apple or something like that um but yeah it's quite interesting it's like the client side proving will unlock a lot of things and um as I mentioned there's a workshop you can actually test it out it's quite simple to do like there's a package called Scar and then you can create your uh program you can create your proof and you can verify the proof it so it's quite we tried to initially with stone one it was a bit difficult because first it was in C++ I think um and then the second wasn't necessarily straightforward how to use these and with Stone Pro two two it will be much simpler as a developer to to generate and create these all the or some of send right so um so basically like the finality for transactions is like two seconds so whenever you create like this um proes on the client side what happens is that you will send it to um starnet to the mle um and um the M who receives it it knows that it arrived um then uh the transaction is being processed by the the sequencer um and then you will reach finality of that transaction like roughly two seconds might be improved in the future I don't know exactly I have to ask the people from starkware but um um the finality on L2 will be reached in just one or two seconds and then the finality on layer one will be reached in five hours or six hours this right so that's up to the like developer itself how they uh do these uh client side proving so for example you can send like every time you um it depends on the developer so I think how it works like if you develop the game uh you can send it per levels you send a transaction to the once you finish the level you send the proof um and then on the the stet self these are aggregated and have the recursive proof but on the client side it's up to you to decide how you want to um Define these like you can send the proof every minute every level it's it's up to the like d developer how how they want to create this architecture it's still like state-of-the-art um I know like the people from starware and cartridge are working on this um and um I think we'll try to they already like sent some proofs um tested some proofs and we'll see maybe in some of the uh upcoming games um to be tested out to see their power okay and just to show like overall like the ecosystem so there's a bunch of applications projects being built um it's quite initially like my my journey into Stark net is I came for the Cool Tech I've heard about many interesting applications like the carrier language um the the technology that stock and darket um has been developing but then um for me the ecosystem kind of like um made me wholesome because like the EOS is so friendly helpful and uh um a bunch of folks like people are comparing it to I don't know how it was for early ethereum uh Community but people would compare it to early ethereum um so there's a bunch of um projects being born I think this is kind of like an outdated uh ecosystem it's from November from there like we have a bunch buch of others uh but you can see also like the gaming um e ecosystem where like the projects are growing um there's a bunch of things that you can do on the gaming side mainly because there's a engine called Dojo engine and it offers developers to create these uh interesting games and more like heavy compute games um yeah so I do invite you guys to um explore it to see it for yourself uh it's quite an interesting ecosystem overall okay so just to give you a quick summary on why Stark net so again as I mentioned it combines the power of approval execution with the transparency of an open Ledger uh there's the KIRO it's a powerful um programming language it's um um which helps us to create these uh ZK U like proofs under the ZKA circuits um all of this is optimized for the ZKA Technologies and what's nice about this is that it's secured by the star proof system but also by ethereum itself which inherits its decentralization transparency and security so in the end we settle on ethereum uh it comes with Native account obstruction so there's a bunch of new features that you can uh create um the I think the important point is that once we try to unboard the web to users we need to make it as simple as possible um I know like in today's society like if you're web to user you probably have like seven eight wallet extensions installed um we need to make this process way simpler um there's terms called like invisible wallets users don't even have to pay attention of like wallets you can create all sorts of things and then finally um what's important is that you execute once and verify everywhere as I mentioned you have this um uh execution you you create a proof out that execution and then you can send it uh to ethereum to verify it everywhere that's kind of like the summary of startet um and um do we how are we doing on time 44 yeah I just wanted to show you like just a quick example of like a smart contract um and like to show you like how how friendly it is so um this is uh let me just quickly get out of here so I do have here like a complete like fresh VM so um just for the sake of it and I'm updating something that's not good have some questions until then yeah let's let's have some questions any questions from Zoom or from uh from the [Music] audience um so in terms of developer experience I believe like um K has come a long way um obviously it's a completely new language so it's more difficult for example if you want to uh develop on optimism it's going to be solity again it's going to be easy for a solity developer to go there but in terms of like um experience personally um I think the the CH language is more robust more secure uh it's more um um Dynamic so you you can create certain specialized macros that uh if you want to um that helps you to uh create these abstractions um and overall I think um it's it's uh it's better than solidity um it it keeps improving compared to solidity um but uh I can I can probably try to show you in just a second so um let me see if I what do I need to do here um all right so there was actually just give me one second um software mention right so software mention just created like um like one line code um just let me see if I can quickly see it somewhere yes codes SC Stark up yes so this is one so basically there's a couple of libraries that you have to install so one of them is called uh scarb so scarb is something called like like a package manager it's not this one though SC it's a package manager that is similar to um cargo yeah exactly and basically allows you to install other libraries um dependencies things like this then the uh other thing that you would need is um SN Forge um so or star net Foundry um starnet let's put like this stocket Foundry similar thing as a Foundry on ethereum it basically it's a Tool uh for testing so it allows you to test your contract but also it has a CLI to um deploy and uh interact with your smart contracts and uh besides that oh we have something in the chat any questions okay got it and then besides that maybe you'll have to install like um devet if you want to so there's Ross devet Stark net Ross defet but essentially you just need um Scar and fondre to get started so uh I'm just going to come here U so there's a simple Cur command that you can use and then install all the necessary dependencies uh not good hope the internet is fast uh let me see if I can make this okay let's see start yes okay I do have to run a new terminal so I can exit this so now like we install all the necessary uh dependencies and then besides this you'll probably need an IDE let's give it a couple of seconds I can take any questions let me want like something for solidity like I mean this is the only reason my smart contract it's a right so for the audience in the zoom the question was uh so it's a bit difficult to move to uh stet because it's a whole new language and uh if there's any plans to do something similar as uh the arbit stylus if I'm just to like translate the code like design I still need a new audit that cost me a lot to yeah so um Cairo like initially KIRO um it will be KIRO won't be like an extra Library like stylus as as for rust um it does come similar as rust so if you already know the rust picking up Kyo would probably take you one or two days uh the same Concepts apply um now in terms of like um there was a project called Kakarot which they uh developed this um where you could convert basically solid code wrong solid Cod run solidity code in kakar and um they basically would just transpile the bite code to kro I believe they have um um um paused that process for now because the network won't be able to um um um hold the the amount of like um TPS and things like this like the activity but once the KIRO native and the St proofer comes into play uh they might uh use that again so basic basically kakar you will be able to use solidity codes into Koo without even writing K but uh yeah I don't think there will be plans to actually create like a library like rust it's it's purely um car um Koo which is similar to Russ and um um yeah I I think as a developer so for a solidity developer I think nethermind mentioned this it would take roughly two to three weeks for a complete solid developer to switch to um um Kum but if you know rust as well it will probably take less than that um developing in um yeah I I'll show you just a couple of examples so just give me any other questions so I have here A scarb version Let me see so we see it's 2.9.2 can you guys see or is it enough and then um s SN Forge version so yeah so basically there was just the on line command that I had to run initially um you would have to install this so what what what we use is ASDF it's kind of like a package that allows us to switch versions between different packages it's super simple and then we have scarb and Stark net Foundry so initially um um this was it um I can start creating um like on my side um on this terminal I have different shortcuts like if I go to Linux I have completely different shortcuts but um just to Showcase here so I'll create um let's say um demo two thinking Romanian English just not so we say scarve new and then here we say the uh package name so we can say includ and then we can choose like which type of like test Runner we want to have and we have The stet Foundry or you have the native kot test so we'll pick a stark Foundry for this case um we will download everything so it just takes a little bit and once we have that we can notice the uh file itself we can just go inside it and then um open the the project so let's see so what happens is that we have a couple of things so we have the SRC which is the source folder we have scarp tummel which is the cargo tummel uh and then if you take a look here it's just basic things like you have the package name version and addition um you have the dependency so if you want to write smart contracts you have to add the uh Stark net as a dependency otherwise you would just write vanilla Ciro um and then here we have the development dependencies these will only run when you um um devel or like test so we have stet Foundry and then some Ser micos um and there's other things here that um like Sierra Sierra Sierra is something called um safe intermate representation um so what happens is that you compile kro to Sierra and then the Sierra sent to the kro VM and then the C the km will compile it down to native kte code that's what happens initially uh it was just KO bite go directly to the EV to to the km uh however this had a big issue um in case of um the execution would be um invalid like for example the transaction will fail the sequencer wouldn't be be able to be paid uh because um the transaction would be rejected and then the sequencer wouldn't be able to process that so uh it would open like a dos attack of deny service and you could basically uh like the how to say um um like the sequencer wouldn't be paid for the dos and uh it could incur like a financial attack so this is why Sierra was coming into place um CR is able to handle these rejected uh transactions and those rejected transactions are are able to be processed on the network and included in the network itself parenthesis anyway um so this is your standard um code I think this a bit too big so what happens here is like you have the interface or the ABI um these are the macros that I mentioned so the macros are kind of like abstracts a bunch of code for you and it tells to the compiler hey treat this specific uh function as an interface or uh treat the specific functions as a contract so these are the macros how we communicate to the compiler um this is the public interface I hello Stark this is like just boiler code uh borderer play code and we have two functions increase and get balance now maybe a question for the audience do you guys know what's the difference between the two so you notice there's a ref self T contract State and then there's a self at Contract State one transaction one sorry one want okay yeah yeah okay any other guesses yeah so with one similar thing like one with one whenever you um read what happens is that you just quing the um um no uh like the uh full noes right so you're quing the result you don't act really interact with the startet network you don't modify the state of the network but the other one it's a right method um because I change the state of my contract so for example um maybe in the contract itself in the storage I have something called balance where I keep a number and whenever I want to change this I have to um change the global state of the contract right so for that you'll also create a transaction because whenever you want to change the global State you'll have to pay a fee for it um and then this is the difference so the increase balance whenever you have the ref self you will have access to the contract State and you'll be able to modify it you can have like these right functions but whenever you have the uh at is just a copy of that contract State and you won't be able to do anything to it so you'll be uh sure that if if I tried to do here like a right method uh the compiler would give me errors I won't be able to compile my code and this is the type safety that I mentioned so it has Type safety involved dat for example I write lot of data you have a liation gas know uh you have a limitation to right so U it's called something like um just give me a second it's called koeps which is basically like the amount of compute and um where is it resources not found for topics to economics so there's there's actually some uh so for let me see if I can chain info maybe they recently changed the [Music] um um so here you have so basically like the uh transaction limit so you have a maximum transaction size and uh which allows you to write 10,000 10 million K steps um both on cipia and M and this is how you actually measure like the um like the amount of um uh compute that you need to do so you would be able to write this and then the same applies for the read Because if you cannot write more than that you won't be able to um any other questions um yeah so this is how simple contract looks like this is how where we denote the uh like contract itself he we have the storage so the storage each contract has its own storage where you can keep like variables and all the logic for your contract um then this is the um implementation of my interface that I have defined here so whenever you create an interface you can say like okay I'll have like two functions I'll have increase balance and get balance right I Define them here at the top these are public functions that other contracts or users can interact with and now like in my implementation I need to define the code logic so how do these functions work so for the increase balance we'll notice that there's a right method here so this will basically read the balance and then we'll get the amount and then for the get balance is just a simple read even if I try here to say self balance uh right and then say I don't know like 10 let me put like this you will see that the compiler would um um when I move this let's say like this it will will compile because it's like uh I'm not able to um Implement a right method on a read read function so um you have the comp you always have this error at the compile time do you have custom errors um custom errors yeah I think you can you can Implement custom errors as well so um um I believe so yeah events events Works similar to um one more time so about the of events yeah yeah it works similar for example because I'm asking for deps that require the gra for example to process the data events yeah yeah they work similarly as ethereum I mean I can even show you like on the Stark scan to see like um let's see an event here we're just going to grab a transaction ction hopefully they mitted an event and then you have the events here and then you can U query these like you can use indexers and then you can uh think you use the graph as well and then you can uh get these events and then you see like the approval event has been triggered and then inside this event you'll have like whatever um you decided so you have the the main difference is the syntax yeah yeah I mean the logic of a smart contct is a bit the same although here you have other nuances so everything is uh immutable by default um for for star for Koo um and um um it inherence like the um ownership model so once you like pass the ownership to another variable then you lose the ownership of that variable like R for example you have um let a take this ownership um and then I cannot use any more like like let's say let's put it this way way let me put here so let let a equal to 10 right um let b equal to a and let c equal to a now the thing is like I cannot use let c equal to a because I already passed the ownership here uh towards it so you have to make the mutable and like this is like the ownership AO like um once you pass it to the other one um inherently you lose the ownership uh it mostly comes from rust um um and uh it's it's more safe to do it like that um and um yeah the nice thing about KO as well is like you have these macros so you can also like create your own custom macros um so if you if you for example look at uh Dojo engine the dojo engine it's um a tool to help game developers to develop their own game they have consist of quite a lot of cool things and we have this Dojo starter um let's see so you can see now like you can see okay it's not stet contract it's a dojo contract why because they Define this macro differently and um um how Doo works is they have like this ECS architecture for games that's based on entity component system and they could like remodel the whole thing according to their specifics um and there's a bunch of things you can Define do events and so on and so forth so it's like highly customizable and uh quite powerful as well um but yeah um this is mostly that I want to show you I mean we can deploy it as well on the the um network but I think uh we can skip that um I'll probably invite you to uh one of our base cams I'll talk to that in a second but maybe if you have any questions about KIRO feel free to let me know um how do you guys see see that the first site is it have standard like PFC yes yes yes yeah so like you have um from open Zeppelin um we have the standards um and um the nice thing about this is that they work as a component um so components are kind of like blueprints uh and I can take this blueprint and reuse it throughout my contract as many as I want to like um for example um let's say we have you want to create a brc2 token um and then this is the implementation that you have to like you you basically create the component here I mean this is the whole code to create a token uh that's basically it um this is like the component uh structure and then you uh you inherit this blueprint you initiate this blueprint in your contract and then you'll be able to uh create it so blueprint the storage I mean the storage of the smart contract for example when you import libraries onm right yeah so components so components basically they become part of the main contract become like uh like when you compile it down to Sierra it just becomes part of the code itself it depends on you what type of components you want to integrate you can create your own components as well like for example let's say you want to create game is super simple to use as an example you want to create a game that has this uh component of creating a character and you want to reuse this across your code you can create that or you can use open Zepp in their standard um um um contracts you have USC 721 you have the security stuff so you have everything here um it's quite customizable but if you want to use like specific calls like libraries so this is a bit more advanced there's something called addition to to 20247 this is kind of like the Prelude that it comes with only the essential libraries um and then as you go and develop your contract you import those libraries and those become part of your contract you can use another addition it's called Uh 2023 this comes with all the libraries but technically it will be your contract will be a bit bigger or in terms of like yeah oh you have like the yeah yeah so you have the upgrade here and what basically becomes is like you replace the hash of the uh class like class hash you replace the class hash you um initiate this um um function in your contract and whenever you deploy a new contract you replace the class hash and then we'll point to that new contract like axy yeah yeah like a proxy yeah yeah quite ster you have this and then if anything changes to your smart contract you maybe decided to change the function or whatever the logic you'll deploy it you receive a new class hash that you deployed and then you'll use that in your all contract you'll upgrade it and that's it a symol no worri um any other questions question so uh could you explain on St that and right these are always confusing to me like um I don't think I can properly explain it um so sequencer failure there's no yeah I I might be able to check this out and then come back to you um the sequence so the IDE is that there is this to centralize it yes and now uh how would it look in the final architecture basically in the end you still end up with a kind of network of nodes yeah that have to have some kind of consensus algorithm and uh they all as a network that that communicate through consensus algorithm agree to State and then push the state to L1 so B this is the way that I say I don't know just spe yeah um it's a good question I'm not that familiar with the whole decentralization topic but what from what I can think of is that all the not operators will be able to like act as a sequencer um they will be picked for example and then they will produce like the new block and whatever and I believe um like the rest of them they won't reexecute the whole thing they would just verify the proof that was generated um and then that's how the consensus reached but I I I have a good uh documentation on and I can send it to you regarding but the whole process is quite complex because the centralizing rollups is not straightforward there's multiple components so for example you have the sequencer uh there's a um Carri Caro ORS that there is like a Koo contract that operates that needs to be um also integrated um there's the shared prover sharp that creates the proof so it's there's multiple um mod that need to be decentralized um and um it's a complex um um situation but I have a good documentation all send it over so like what's the questions like do you have like a messaging system inside the yeah so it's one L2 messaging and L2 L1 messaging as well also l21 yeah but that's right yeah and quite fast I don't know exactly I think so yeah it should be fast quite fast but there's like the like if you want to start start bridge I think um start gate sorry this like the they also updated the whole like bridging from from L1 and L2 and L2 to L1 and then it's it's quite fast there's other like rhi and all others that you can use and what about start the it's basically a gas token just like in or so um yeah good question so the star token is basically like will act as um so you'll be able to use it as a gas gas token I think at one point um So currently as a as a gas fee ethereum and stocket is supported on L2 but at one point will it will be only a St token and uh it does have like other capabilities for decentralization you can use it as a voting U mechanism so you can have your star to able to vote on um different um proposals um you can also delegate the star tokens um yeah okay then um I do want to just finalize on a few things so we've seen this um code hopefully it was not that uh scary uh but um now the question is where to start so there's a lot of places to start and whenever you're coming into a new ecosystem uh it might be confusing because there's a lot of places to go so we do have this um developer Journey that we created um and then it consists of of two six modules and I would like to Define them like by levels so the first three modules is level one and the last three modules is level two the first three modules is learn where you basically start to hear about Stark net you start hearing about Cyro and here you have the opportunity to uh check Base Cam so base camp um I'll get to that in a second but it's kind of like a free 4 to6 week course that you'll be able to learn about startet kirro also get hands-on experience I started myself with Base Cam cohort number two now we're at 12 and that's how I kind of like got into starnet I've learned about starnet we've learned KIRO zero was terrible it was a very difficult language but the whole process was very nice I've learned quite a lot and uh since then I kind of like always wanted to give back to the community and I got myself involved and um yeah it's it's just an amazing um Journey then there's starkings um anyway with rust background it's rustling similar to rustling it's kind of like an interactive uh way to learn how to code then there's no Guardians scaffold star and webinars then there's the next stage is connect so this is where you already learned something and you want to connect and um you know make connection the ecosystem the community so there's various ways to do this there's only dust platform which is kind of like allows you to connect with uh developers and also like other people um ambassador program meet meet ups such as this or even like start meetups or conferences um anytime you go to like an Etc conference or uh Defcon con conference make sure to check Stark net we always have some kind of like a big conference happening there and it's it's it's super nice it's it's quite fun um I know like last time in Defcon we did have um so the gaming folks are a bit U Crazy and they had like a actually like a boxing ring as their main conference and they were like just doing a uh kind of like a podcast here on a boxing ring and it was like very thematic so it was quite fun um then there's the build phase where you learn all of this tools and skills and uh also connections and you take you know like you can do be a solo developer or you have a team and then you can participate in hackathons hacker houses and also like ideation program um we all have this planned for you U it all starts with base cams and then from there it's just going with the flow this is where you level up so you level up from level one and you go to level two in level two we have um the Kickstart where you we you receive support from the foundation and from stocket U there's cants and rapper programs I'll talk to that in a second but basically you receive um a certain amount of um stock tokens in Grand uh and you'll be able to develop your project business on on stocket then iterate you you go through you can go through accelerator program to help you even um um build faster and uh build your business faster and you also get VC relations and finally the last uh module is grow uh where you basically reach the like maturity of the project of the business but also you get better support there this is like the journey that uh we have and um I I'll start inviting you with stet Base Cam so again it's this one is a four-week program um with cam with the bonus session so in the first four sessions you have fundamentals smart contracts testing and front end and then the bonus session is build games with Dojo so if you if you like like building games feel free to check that out there's a link there I can share the presentation later um and again it's it's it's also the way I started uh it's quite a good program and um um it's quite fun when is the next yeah when does start April so it starts April the 5th maybe I think so something like that yeah so and I think the time is quite good like it's European Focus time because um because there's like some big events for example for E Denver we were focusing more on the US region I was like quite late but now like these ones are I think around like 5 to 600 p.m or something like so it fits really good then you have like test your skills so you already learned something it's time to like show off you have Stark Ling which is kind of like the interactive tutorial not Guardians is quite fun because it's like it's um it's a very like gamified platform and you have like a character that um you can solve like ctfs like you solve various like uh small challenges and if you sol them you unlock items right and then these items you can equip your character with it and then uh you can just show off to your friends or I don't know if you can put it in a CV it's going to be quite fun and there's like exorcism that's kind of like an open source platform you can also solve exercises there so there's a bunch of ways to do and then time for building it's as I mentioned the C grants it's a program designed to offer um uh grants up to $25,000 in Star tokens and basically we are trying to support teams um to build um Stark net and uh develop their business ideas um other other platforms that I do invite you is only do so only do is quite an amazing Community overall it's an open source Community there's open source Believers and basically you're able to work on um like state-ofthe-art projects um and the uh the nice thing about it is that you're able to develop your skills you're able to connect with people but also you're rewarded for your contribution so if this something uh that's uh interesting for you you can explore only dust and uh that's it I think uh that's my uh presentation uh feel free to scan the QR code if you want to uh you'll have access to the presentation and if there are any questions that I can help you answer uh I'm happy to help when when's the next heon so you said you organized heon recently yes how are those you organized different what's the experience in there right so uh so the next one is going to be in May um so usually how we have is like whenever you see this um funnel we always have like um these pipeline so we start with the Base Cam then from Base Cam you have the hackathon and then we have a hacker house so once you start a base cam there's like cool things that you can do like you can uh start learning you go to um hackathon you test everything you build a PC and then with that PC um proof of concept you can go further to The Hacker House hacker houses are nice because everything is sponsored like the the um staying the venue the the food everything you just have to show up uh with your team and develop the product so it's quite nice and it's mention me so it's usually around of conferences it's ored Side Event yeah yeah it's usually conferences Big E conferences correct correct it's usually like we tend to make it before those happening because I know like um people are focused on the uh conferences and then you know like they go to the conferences like HCC or deathcon uh it's usually like with it starts with five days before and then it ends with one day before the main event but usually like we have like now we have at e themer one we're going to have at e cans uh or sorry cans not e cans ECC cans um and then death connect I heard is going to be in Buenos Aris yeah this is like Alpha I don't know if it's Alpha anyway but um yeah usually we'll have like around these uh Main Events and uh yeah it's quite fun um you can participate with the hacker house and then you can join the the conference later uh the nice thing about The Hacker House you also get access to like at the Demonite so um um you present in front of the public like the top projects present in front of the public and you have VCS judges there and then uh it's quite nice opportunity um hackaton itself they're quite like stet focused um and um the last second that we did we did in January we had about 800 registrations with like 110 submissions in total project submissions um and then it had like five different tracks so we had defi ai ai agents um um telegram gaming and uh Community um and then for example if you're interested in particular track you can develop but also you could apply for multiple tracks so for example you can develop like a telegram defi app right that that's also POS possible um their fun experience um I think the the the one that we organized prior to that was quite successful and the next one will be in May so um it's again it's kind of like takes part of this bigger picture to funnel people like keep the the journey for uh through through this dark ecosystem any other questions why do you think it's missing on the EV side because I think [Music] good question um I think um consumer apps so we need you know two one or two big consumer apps like um so first of all it's people are scared of like switching from something that's familiar right like from evm to to St not because it's like there's a new technology a new programming language so definitely there's some challenges um I think uh we so we have the uh Network performance uh we are trying Stark net Foundation uh starware starware as well we're actively trying to uh onboard companies um and projects and then like consumer apps and I think what we were looking for is like the next one or two big consumer apps that will change everything so it's like if you have the the right thing everything will snowball um yeah there's a bunch of them I don't know like if it's over here yeah so so there's like U you have like well these are the centralized exchanges the D5 protocols um M name ZK land is there uh but yeah dexes so have like my I mean I I don't want to pick favorites but one one of my fav it is Avenue and uh ubo these are like good uh good um dexes um ubo they they also like just released like um I don't know if I can find it privacy tools so they just released this thing um I think is this one is it is also yeah so kuo like the f d Kubo is um X Unis swap um and then they decided to create something better um and then um equ Bo and they recently just released this like privacy pools and U just leverage circum uh circuits with starnet and grow 16 um yeah it's a bunch of nice things and Innovation um and uh yeah you can explore like the whole ecosystem but there's a bunch of other um texes as well and I guess all the concept are appear yeah yeah even even better because you can create a lot of of different things as Landings yeah yeah so vzu you can use Vu as well for Lending and uh borrowing there's um Autos as well cool any other quote Yeah interrup between uh rollups um there are some challenges yes um but um yeah it's difficult to um it's not straightforward so definitely there's challenges between other rollups um especially being um non evm completely like we are I think if you know like there's type four types type one it's like full equivalent type two you have some equivalency type three and we are type four like we we're not like we're far away um but it's it's a it's a big betat that we took uh mainly because we we try to focus on the technology on the scaling side uh we came a long way like for example like the community in 20120 there weren't so many people but uh now now like in terms of developers and Community is growing um and uh yeah it's challenging there do there are some um hiccups here and there but um they're solvable they're solvable and there's a lot of cool things that you can do with um like um um stocket as as well you can have like circum on stet and like use circum for proof generational privacy you can have Noir as well so there's uh there's a cool Library called Gaga um that they do all of this um interesting Innovations and then uh basically enables you to do like uh growth 16s um and then you can also use Noir U and a bunch of other things yeah okay well thank you again for U inviting me here and listening you and thank you it's a fun language like U initially like the whole transition U from K zero was like this pythonic thing and you also had to care uh take care of