The Power of the Partnership Playbook with Li Xixi, Nalin Mittal, Susie Batt, Temoc W., & Richard W.

great read or audible I'm Susie batt Steward on Spork Dell and that governs this fabulous festival and bufficorn Ventures and Biddle Brigade I am joined by my friends and panelists um Igor from Gateway FM nalin from Google Cloud and shishi from sender wallet uh so let's divide this half hour into two parts let's first talk about infrastructure needs and the innovation and then Partnerships our challenge in web 3 is to build wide widely accessible adapts for the masses we attempt to do this by attracting new Talent and Innovation and offering Builders comprehensive Dev tooling and open participation of running nodes would each of you please speak about your organization its mission to support web 3 infrastructure growth and devtooling hey everyone I'm Nolan from Google Cloud I lead web3 product there and uh our mission is really to be the best place for web3 developers right so today we're focusing on node infrastructure just specifically fully managed dedicated nodes as well as indexed blockchain data so indexing the data putting it into bigquery and making it available through SQL and we see you know the challenges are how do we make it easier for web 3 developers to build solving some of these foundational primitive needs like node infrastructure and data but then also how do we get more web 2 developers building on web3 how do we help them not worry about RPC nodes and requests and you know apis and provide the levels of abstraction that make it super easy for any new developer to enter the space Igor wanted to riff off of that because our PC nodes are your thing yeah I'm eager from Gateway we also run RPC notes distributed around the globe and we are also like a part of our mission is to make sure that's the spirit of decentralization of this space is alive so it's not only Google or Amazon where you can get your notes from but also like from independent data centers that's why we built like an extremely portable Cloud agnostic infrastructure on top of whatever that can build this thing and also we not only applying it for RPC nodes themselves but also for things like real thing like a node validation or like points of censorships when you can do because they're like typically two points of censorships on the nodes they used to be like an experiment in the old Bitcoin Community about uh if miners collude versus frpc providers collude on different softworks who wins and turns out that's both loose so so sort of both ways are important because let's say if your Miners collude and somebody runs their own nodes they will be on the fork of the miners but if the miners collude and let's say the RPC providers pull those who use RPC providers will be on their PC providers Fork so it's all very important to keep it all consistent and and decentralized and to have it open not only for the players who will use this service but also like who run it as well yeah definitely So currently we're building a wallet it's called sender wallet and it's basically a native near wallet and by using the strategy technology we um we find it a little bit um in like adequate or in terms of the users and we're going to use theorem and evm as well as layer twos So currently and Lancaster wallet is very very competitive and we are transferring our product or we are gonna come out with a cinder Pro which is basically an account obstruction and within that we are gonna need a lot of like you know um a lot of instructure a lot of proposals on eip4337 which is like very significant for the future of wallet and as well in the meantime I we find it very very crucial for the developers because it is really really hard to find you know really good smart contract developers and I think the developer like ecosystem is also very very essential for the infrastructure somebody just won a buffer corn congrats um okay life is good um this Festival East Denver is always focused on builders on you all um I'd like to ask the panelists what are you doing specifically to attract devs or improve Dev tooling um to your products okay so I can start so uh at Gateway we are a little bit in a unique space where we are both uh part of the core developers team on both ethereum and gnosis chain so we are also not only we are trying to like provide reliable tooling infrastructure documentation and some extra stuff for the developers but we're also trying to bridge this Gap that always was there in the community when there was one group of people who built snows and totally separate group of people who use and run them and that ends up with some funny stuff like your restart your server at the wrong time and you notice that forever so yeah so we are trying to kind of bridge this Gap to make sure that it's better not only for us but even for other infrastructure providers to run these things by participating actively in the core development of some chains Nolan yeah sure so you know I'll talk about the two Focus areas we have again so in terms of node infrastructure we're building a service that's already out in private preview for fully managed ethereum nodes so with a single API call or a single click in the UI you get a node it syncs with the blockchain we monitor it we make sure it doesn't go down restart it safely if need be upgraded for you uh you can choose whether you want mainnet or test Nets full nodes or archive nodes and because these are dedicated nodes you get a lot of configurability so if you want to put these in regions of your choice you can do that if you want to put them behind a firewall so only your team or your partners can access that node you can do that so really just trying to provide the very basic foundational primitive for node infrastructures so that other providers whether they're platforms or end developers can quickly scale up their elastic you know set of nodes pool of nodes then on the data side there's there's you know we look at how do we make this data more accessible so by putting it in bigquery it becomes a lot easier for any developer to just query it with SQL like they have been their web 2 data and long term we want to do things like decode that data make it very easy to understand um do like token balances nft balances every developer doesn't have to do that same process over and over again so that's kind of how we look at it anything you want to add to that Sushi yeah absolutely so and as now we're building a wallet and since it's really near so and I think there are some developers over there but after we um you know go meditation and after we are going to develop like on ethereum and evm there will definitely be more developers involved and with as with more Integrations okay so the big elephant in the room is not really an elephant but Google and we talked about this um yesterday um you and I and how you know over the years um we've seen this sort of web 2 company creep but um as welcoming as we all are um we're a little curious like what's the play and um um what are you looking to the attraction to entrepreneurs is clear but I'd like for you to talk a little bit about it and kind of dispel any thoughts that maybe the audience might have sure so you know for us it's pretty simple for the last three years we've had a digital asset sales team in Cloud that's been helping web3 native companies and web 2 companies that are getting more into web3 build their workloads on gcp right so this is we've been doing this for a while if you look at you know what we've done with Dapper and help them launch NBA Top Shot and flow to being part of hedera's council running validators with them so we've just naturally been doing more and more with web3 workloads and it became very obvious that we should have a web 3 product offering in Cloud to serve these customers even better so it's it hasn't really been a um you know a huge internal cell or anything like that it's just it's a natural extension of the business we've been doing then you know there's other teams within Google that are looking more broadly how can we apply web3 to other parts of the business and those that's a separate team that we partner with along that spectrum is decentralized node infrastructure Igor you've been working on this for quite some time and your background comes or the Genesis is really uh turbogaf can you talk I know it's very OG a lot of anyone here remember turbogiath raise your hand yeah okay talk a little bit about that yeah all right so I think it was like 20 17 18 when the turbo DS was it it was a medium post by Alexi who was like okay yet is not really that performance at certain things we can try to do better and then for a year and a half he was going and trying to convince different people to change the data model for how because like going back a little bit uh how ethereum usually stores this data or store this data at first was basically sort of a carbon copy of how Bitcoin did it so basically you have like this miracle tree you have this miracle proof in the root and you store data in the tree structure and this three structure I'm getting very deep here you then need to store it in the database which is a flat key it's like kind of key value storage thing here right and then after that basically what's Bitcoin did and what ethereum at first did because basically yeah we just serialized it as it goes like in a very naive way which uh and the difference between Bitcoin and ethereum is obviously the smart contract so the amount of state of like Bitcoin is minuscule so it works for them you don't need to optimize anything but one you have like 300 million active smart contracts on ethereum Main net ish in like maybe a couple of years a year ago I think I checked it uh then it's totally different numbers so then at some point when uh we gave up trying to convince people to to change the data model convince Insurance house is hard so we decided to just build it so I joined the team in 2019 I think and uh yeah we've been basically started with just guess throwing away the data storage replacing it then it eventually we replace the scene model as well and then we replaced at some point because we also found this issue that with the databases that usually are used let's say maybe lmdb no sorry level DB is the default one that Geth was using then rog's DB I think is what nethermind is using so this database that they don't have transactions that means that if you try to do multiple operations and let's say I don't know updating try applying a transaction to your state is multiple operation and you restart your note at the wrong time it corrupts the data because it doesn't have transaction to roll back so we switch to a database called yeah lmdb that was transactional but when we try to put ethereum State there which an archival state which was at that point like 1.2 terabytes I think in Oregon which I think in gift it was around 10 terabytes or something like this and we found some bugs then the developer told us that no you shouldn't put 1.2 terabytes in my database and the declines all support so then we switch to another database which was a fork of this one but with the developer who actually wanted to work with us on this so we ended up with this kind of interesting project so the reason why I bring this up for young devs who are here is to understand that where you start is in a project or is not where you end and that there is an evolution like Igor and you know all of us here through the years um of the product and of the infrastructure I'd like to move now into the talk of Partnerships because as Igor's um talked about um being not new to this ecosystem you cannot grow without Partnerships Google who is new to web3 also relies on Partnerships and shishi being a wallet you also rely on many Partnerships whether it's wallet connect or multi-chain so could you speak about first some lessons learned of Partnerships that you have manifested through this product iteration hello yeah so yeah sure I think you know for us we want to really focus on what Google does well so data compute uh strengths and so we can apply a lot of expertise to the space there and where we don't have expertise we're looking to partner so you know we've done a bunch of Partnerships over the last few years coinbase was a big one where we helped them run their blockchain workloads and they're helping us accept crypto to pay for cloud right so if you want to pay for cloud with usdc that's a partnership we do with them also holding crypto on our balance sheets is something that would have been very difficult for us to do without a partner like coinbase Solana is another one that you know we partner with so we we've helped them from quite early days with holding their historical data on things like bigtable and managing that and now we're you know helping their developer Community with uh helping helping run nodes making that data more accessible so you know I think a lot of people in the space talk about Partnerships in terms of Di or sales and that's fine that's like a form of partnership but we really look for more bi-directional Partnerships where we can help our customers and and they can help us actually move the ecosystem forward so whether it's New Market or new tech stack yes exactly so you know for for us we're we're we've obviously been doing web 2 Cloud for quite a long time and by partnering with other folks in the space I think we can kind of um start to extend our leadership in that web three space and then likewise I think we're seeing right now a trend where you know there's a little bit of slowdown in activity I think I'm very energized by this conference and how much is going on but you are seeing like a lot of web 2 or traditional Enterprise companies get into web3 and that's where we can help like be that bridge and bring web 2 customers to the ecosystem please add so and I actually just started my um my experience in terms of like reading a product I was nxvc so my Partnerships experience and the Lessons Learned I basically all from VCS and I think the least meaningful Partnerships are basically those just co-marketing like those co-marketing stuff on Twitter I just returned and it's like nobody knows like what is what do you do and what exactly are you working on and I think one example that I Really Wanna I think that's the most meaningful partnership which is like the superhero Sam from Paradigm he found out like the bags that saved 350 million dollars on Sushi strap and then he found another bag on dydx and I think that kind of like partnership they invested in them and then they are like helping them fundamentally is the true partnership and I think think that's that which build up security and that actually are helping the industry is very very meaningful and as a wallet I just learned that we have so many Partnerships to establish and collaborations I never knew as a VC but now I know that like we could not only you know like co-marketing but we could also integrate their their like functions within our UI ux and that we could have improved liquidity on both ends as well as you know Advanced adoption and innovation uh yeah and for me the Partnerships are the core that we do at Gateway for instance we again two of the more successful ones are one with gnosis chain and another with Eric on the notes so in in the notes it's again as I told we're just Bridging the Gap a bridge in our experience of running the nodes and issues that we find while we run them at scale because it's difficult to test this kind of tricky protocols and things as scale right and with everyone we've also been helping like to going through the merge process the merge of ethereum Now with uh withdrawals and stuff like this fagnosis chain for other like more because it knows this chain is quite an OG I would say it's like upcoming protocol but still when you're launching your protocol on ul1 you you need things like devnets test Nets reliable rpcs some data indexing and things like this and if you're building an like a protocol and you're doing like super cool consensus algorithm that's not your core experience to do like devops stuff in general and uh so that's where we help them and of course the better their note Tech is and the more popular their network is is better for us so it's been very very fruitful again with noses chain and of the Oregon yeah and some other players so looking forward um I'd like to get your insights on what we could look forward to at Devcon vitalik talked about possibly in the future and it's he always has these Oracle talks right and then they manifest but um he envisions running nodes on your mobile um what what do you what do you foresee like there are incremental steps to that um what would you like to see well I can come and direct him on that because trying to put nodes on mobile was my kind of specialty from 2017 until 2019-ish and uh we are not there yet uh and especially even if we can put the notes like let's say a light notes there's the whole light node research that was revived in the ethereum foundation not so long ago but they still relied on a semi-trusted third party so at that point like you cannot run them in a purely decentralized Manner and the technical issue for that is mostly the battery in your phone because it will like completely be drained in like I don't know three hours of usage if you keep the node running because what you want to do is you want your notes to always be at the head of the chain if you open a wallet you want to see your balance right now or not three months ago but if you're actually saving the battery you need to stop your no and then it needs to catch up and if you want to catch up quickly you need to trust somebody that you get the data correctly and so far maybe there is somebody who is building like a good verifiable protocol for that that does a decentralized industrialized manner but technically what it needs to do it needs to find other nodes and not just trust one URL let's say from Google cloud or something that's that's that it trusts and then it needs to find if they're fine to sing from them then get data and verify this data somehow and this is the major curl bit like you can otherwise technically our phones are I mean most of my our phones right now are faster that our laptops like five years ago so you're definitely with everything else you can run a note on your phone no problem if you have an infinite battery essentially so that's so battery capacity and of course internet yeah we see here like Ryan that's we are kind of doing the P2P things and then you cannot pay for coffee with tokens because internet is down right so it's not really decentralized as much as we would want it to be especially at The Last Mile and this kind of age Computing well I just mentioned internet because like there's always lack of here at our conferences and today's just I know it's terrible um Nolan yeah so and I think the you know latest architecture for ethereum lets you split up validators from consensus layer right so I think one vision is you have really lightweight validators maybe like ethereum Foundation believes you can put them even on Raspberry pies like very lightweight machines because yeah I believe you can put it on your fridge yeah there you go and then you can connect them to a node in the cloud that's doing more of the heavy lifting on the consensus side so I think that's one way to achieve this decentralization where before you needed you know a much heavier weight validator now we're moving to a place where that's not needed and that decentralization can actually happen um and to your point where like not trusting one URL in the cloud for Google cloud or Amazon you know we would love to support multi-cloud uh node deployments in the future so while you know we don't want to be your only node infrastructure we hope to be your best node infrastructure we understand like in the web 3 world you're going to have other redundancy options you're going to have a multi-cloud architecture and we want to play we want to be the place that helps you get there yeah and uh by the way what you're talking about the validator clients we're basically already are doing that at Gateway where it's not on the Raspberry Pi's but in sometimes we need to put validators themselves in some places in the world where data centers are extremely expensive so that's where let's say you have your validator in one country but then you have the actual consensus and execution layer nodes in another country and then you need basically like a super tiny machine with like five gigs of storage and it's like one gig of RAM to serve like validators no problem and it's completely non-custodial because we don't see any keys from anything and you don't need exit transactions and stuff like this so that's already happening right now okay we're at uh two minutes just wanted to add what do you read you read all that since that's inferring that you'd like to share and where can people find you uh okay so the last book that's sort of uh related to that was built by Tony Fadel so uh build by Tony Fadel that's an exceptional book I really recommend it and you can find me at Twitter like mandrigan is my last name so just search for that so that's the main kind of source uh for me I think last book was Network State bibology and you can find me at nullen a at n-a-l-i-n on Twitter so we are basically building a future that everyone will say hey remember back in the 20 early 2020s we still needs deliveries memory and to log into your non-custodial wallet that's gonna be like the future is gonna be account abstraction and that's where I'm actually building right now so you could find us send our wallet on Twitter and the the internet address is sender wallet.org and you could just install browser extensions and mobile apps on iOS as well as Google Play and it could find me and crypto and mega Miga great and again I'm Susie um thanks for coming and please stay our panelists will be around for any questions afterwards thank you everybody [Music]