Modular Execution & VM Landscape 2024

Transcript

welcome back to the rollup VMS VMS and more virtual machines if you've been following us FP for the modular thesis you know that we are pretty excited about what's going on at the virtual machine in the execution layer today's conversation with Robert from Multiverse X was strictly about execution layers we talked about the pros and cons of many designs in the space today we talked about parallelization and we talked about the feature of the modular stack with regards to execution and why it's so important that we scale execution this is going to be a good one you guys will enjoy thank you for being here let's Jump Right In after a word from our lovely lovely sponsors fuel achieves Higher throughput by utilizing a novel

utxo based approach that allows for parallelized execution this design significantly accelerates transaction processing which enables fuel to handle vastly more transactions than non-p parallelized systems thanks to advancements introduced by Fuel ethereum rollups are set to be more powerful than ever before and fuel is not just enhancing ethereum rollups it's redefining them making them faster more scalable and more adaptable to meet the needs of modern applications and developers go to fuel. network that's fuel. Network to learn how to build on the rollup OS welcome back to the rollup today a

conversation on something that we are particularly bullish for in the modular thesis for the coming months is this this idea of execution environments and particularly optimized execution environments for better user experience uh we think that there's been a lot of talk around data availability for the last uh three months rightfully so after Celestia's launch Avil a and others but we actually think that the more impactful changes to rollups and to General execution environments as is going to happen at the execution layer so we're here today with Robert who's a core engineer uh for project Multiverse X which has a super interesting VM design which we will touch on for sure

and of course with with Rob um welcome in Robert and uh it's good to have you on hello thanks for having me um I'm quite happy to see all those changes in the crypto space finally happening finally people figuring out that yeah there is an other alternative way of exec tion which can really scale and actually can make some better user experience better help on Safety and Security for both users and developers uh so it's quite actually a relief that uh things are moving forward and not everybody building on the same stuff which is actually not uh necessarily

good for our complex life yeah I think so do you agree with our sentiment like do you think that data availability is an unlock but the scaling at the execution layer in innovative ways is much bigger of an unlock or they equal or kind of like how do you view the comparison in terms of solving scalability bottlenecks uh from a data availability versus um execution environment scaling both of them are necessary because you cannot work if some if data ability doesn't exist or it's not handled well then you cannot scale because you cannot actually offer those kind of security Primitives for the users uh but execution it's where things are happening for front facing

for the user and then you had to make some changes and add the execution level in order to actually make awesome applications like let's be real all the applications which are web to based the new ones the AAA games or AAA applications whatever they are amazing to use Easy seamless inter interactions seamless onboarding everything is like fit for the user we are still really far away from that experience in terms of crypto space and uh some of those things which is happening at the in web to for the apps couldn't be handled using Legacy

technology and that's a fact there's no Denial on it so is that oh please yeah please finish and and so you need in order to reach billions of users you actually need those kind of better front-facing user focused products which resolve problems and that can be done only with let's from my perspective on new VMS which offer a better developer experience and better user experience and and that's a sentiment that we've heard uh quite a bit is that the reason these AAA games um and these high powerered applications

are not available today is because there's a lack of scale and so I'm curious does the lack of scale we can even look at like what what the current landscape is ethereum right now has somewhere between like I don't know 20 and 30 transactions per second you see you see like some of these other rollups that have they're getting up to like maybe a thousand transactions per second which is quite quite a bit more it's almost like 50x times that does the limitations of scale happen at the execution layer or do they happen at some other parts of the stat it happen on multiple levels it's not only like bite code or contract execution per se but actually access to data as well and

access to complex operations when we look at ethereum and how ethereum was designed it has only one big Patricia Merkel 3 in which you try to execute and as the the this Tre is growing and growing getting information and writing information it's getting slower and slower but if you can make these kind of trees smaller for example one three for one smart contract you actually resolve a set of problems then you can add caches then there are a lot of operations which are let's say common goods like big wind mathematics or big float mathematics or

cryptographic Primitives why would those things need to be executed in solidity or an interpreter just step into existing audited libraries for C cryptographic offer an OP code for the developer and the developer will use that op code and everybody's you don't have to reinvent the v in like and reimagine smart contracts and optimize more smart contracts for these kind of things because you have all those like let's say libraries better tested by NASA or better tested by serious applications and serious uh people yeah and so we kind of get to the state where

I I personally think that we share this belief here where we don't really need too many uh more evm L2 general purpose rollups I think that parallelization at its core is like effectively a standard now um and and that these kind of the only way that um evm will will scale is through parallelization or um effectively kind of being a a like counterpart or a part of a of a system which offers other execution environments um so I generally want to get your take on this um is the evm going to be in a state of less usage over the next 10 years or will do do you

think it'll continue to develop in terms of the broader expansion of crypto and then also how how does that how does the impact of these alternative uh virtual machines and these um you know parallelized execution environments how do they intersect with with evm uh to kind of provide um both functionalities and and kind of plug into uh different types of applications there were a lot of questions here um but let's start with some of the things I think that evm and uh will drop on how many apps there will be developed on evm especially when you we look at uh Solana when we look at

uh move sui Aptos and we look at Multiverse 6 in all those L all those places you actually can write in rust or rust like environments and then you know these kind of projects are going out to University they are keeping hackaton they are uh teaching a lot of uh non web people to come to vbpr and develop those students or non nonb people they don't know solidity but they most effectively know rust or C++ and they will actually start developing the on those languages and then only let's

say the real the OG guys remain on solidity who were developing on solidity and they still remain on solidity and then the the crypto word will move on towards like in right now it seems like rust based DS or rust based languages uh you have a lot of uh functionalities better IDs better unit testing better audits more people out there who can actually review your code and see what's good and what's bad about it and um in some sense um while the liquidity is still on ethereum and still

because of this in evm people will write and actually writing interoperability modules in order to understand evm inside their chain or at least evm messages inside their chain in order to tap into the liquidity and we cannot deny this fact that everybody wants to tap into the liquidity of ethereum or let's say uh steal liquidity from there in order for their application but that's like the truth out of there so there will be uh different modules in which these kind of uh messaging protocols uh cross chain protocols will

work and evolve but the more development we see the more they will go outside of solidity uh that's what I see and with even with the new Bridges like exeler or layer zero or intangle uh they are moving to for example to using tender SDK and then developing in go or using a km wasm building in Rust uh so these will be the languages to be used further is there a uh is there a way to compose with the liquidity on ethereum and also use an alternative VM there are some types of doing that so

uh you can have a synchronous ways in which you passage messages and then you are actually executing the the user facing app is actually executing in other layer but in the you have a set of let's say relay or a service which transfers those kinds of sign messages into ethereum sign messages then you tap into uh one of the thingss we try to do it for for example with our space VM is to put our wasm based VM and evm one next to each other and offer Atomic composability between these two uh this would be a way in which you

could let's say could launch an ethereum L2 in which you have both of these things then on the ethereum L2 side you can tap into for example an optimism or an arbit room type of rollup and get all the information from ethereum as you would look like but you would on top of those applications you developed on your EV evm you would tap that directly uh into a one based contract which is user facing one of the ideas in this sense is for example if you do that and you offer like multi calls for users

uh then a user could do approve execute revoke approval or stuff like that inside one execution uh all is managed by let's say the space VM but you call the evm in this side and then you resolve a set of issues for the user that he has to sign a bunch of transaction give out a lot of approvals and just doesn't know what's happening in in the back um this way you can keep it uh let's say uh more safe one of one more example I would give is that U if evm doesn't want to change it so okay you put the space VM next to it in the space VM you have like other type of cryptographic proofs uh like the secp

2 256 R1 so you would be able to use that one um and you could create applications without seat phrases because you could use the P keys from your phone so what you imagine is that you create an U an account obstruction which is controlled via P keys from the phone the transaction is pushed by a relayer for example with a relayer and a guest service or a subscription service and then you have an an a completely great account abstraction layer which still communicates with evm underneath and the user is completely safe these kinds of

let's say Innovations or moves inside things uh can happen and then you are still directly connected to the liquidity and what's happening on the ethereum layer yeah and it's funny people are like ah EVMS dead and uh yeah all VMS are going to win out and nextg VMS are so much better than every everyone's like well we have to connect back to ethereum still guys it's like yeah okay guys it's funny stuff there butd um stupid question here I'm not a computer guy but I've heard wum is not meant for um blockchains it's more meant for like browser apps why like is this true and like why are you guys using wum or why is why is wum a good or a bad uh

virtual machine for blockchain um that might have been through six or seven years ago but since then there was like an insane evolution of wasm and in ins insane evolution of compilers of compiling rust or other languages to wasm uh like two months ago I was in one month ago I was in Barcelona at the wasm summit and there was like insane things happening there and the thing is that people are running like complete services on wasm uh serverless services or on demand servers which can spin up in milliseconds for you for your

microservices with thousands of machines and like a bunch of companies like Sony is using it for iot and for for like the whole stack and uh wasam actually was developed by Mozilla then Google Facebook came into it a lot of people put it put a lot of effort then llvm the compiler was actually updated several times and it has you know for example if you look at python python is used to test out smart contracts or it's python is used to do all kinds of interactions if you you take the python code compil it it in wasm and you run it the wasm

for the same interactions it's three times faster and you can do so much more amazing things uh and was it's really deterministic executes fast you can do it with compilation or you can do it with interpretation with compilation it's even faster we do some you can do some amazing things like you compile once then you keep the serialized object it's called ahead of time compilation and next time you don't not compile only this serialize or another way of doing is is that you compile and you keep the compiled object and next time you simply just run it and with our benchmarks you can get 1.3

times only 1.3 times slower than running on the same code directly on Barbor machines so that's kind of uh speed you can run reach and um you know with wasm what you can do is actually make not even smart contracts but if you go above in the stack you can make microservices and front end safer what you could do do actually is to run the micros service and to run the front end as a wasm by code directly on the user's phone it runs in a sandbox you can put the binary for the front end uh directly on blockchain for the

phone to read from there to have the authentication from there for example and as an inscription you inscribe your front end to the blockchain you run it the user have the complete Security that he's running the software which is supposed to run there is like the front end was not taken over it's just one verification and actually everything is run faster U this is one of the the things we want to go for and to generate all these steps from Smart contracts microservices and front ends from rust or typescript JavaScript and to have the whole stack for the user [Music]

cartey is your home for app specific rollups at the Forefront of developing new tools and platforms that enhance the capabilities of blockchain Technology ctei enables applications that are flexible and built to scale build on a full Linux environment and use traditional software libraries but in blockchain development go to cy. that's cy. iio to learn more about building your own app chain today can you give a like a concrete example of what one of these verifiable microservices could look like so a microservice is actually reads a bunch of data from the blockchain then

makes some computation and forwards it a front to the front end in order to have it for the user or like specifies what kind of transaction the user have to sign for in our example we've write these kind of things in typescript now you can take the whole typescript code compile it inscribe it on a chain on chain or put it on what directly on the tree then next time when the user checks that he wants to interact with app X then you know the wallet or let's say a micros service runner on an app checks the blockchain downloads the code runs it

executes it on the users on the user and then it directly uh the user can after that sign the message and do it and actually it's it's you know it's much uh it's much more scalable than running your own microservices in a lot of times we had a bunch of issues with our microservices not not the blockchain the blockchain was running fast but like having the indexer the microser is getting too many API calls but when you have these things running on the user user faing it runs his own on his own machine or what you can do more with it you can actually run transaction

simulation on the user's machine so the user would see the result before signing it but directly like you know you cannot fake it yeah I I think it it brings back like Satoshi Vibes because like sat he was all about verify don't trust uh to run something on a local machine is much more decentralized much more self oriented than uh than something that's running on like kind of a like a cloud Network or a centralized server um and so I think the infrastructure makes a lot of sense to distribute this infrastructure across these these net this network of local machines what are could you give some examples of like

what microservices are unlocked via this new infrastructure we don't have this live because you know we went to the wasm weekend uh in Barcelona and then we saw a lot of people launching using this what we saw that there are actually uh some banks uses this kind of wasm stuff in order to run multi party computation before uh everybody's running the same micros service signing them signing them then aggregating the signature and putting that data out between some banks and to create trust in this there was some keychain application there in this sense uh we actually saw a lot of games running like this uh and actually

accessing via this the backend data and you know the wasm it's like you can run it on any browser and it runs deterministically plus it's like a the uh plus it's you know it's uh it's very iable you deploy that or you post that wasm code and you know what's happening there it's it it's somewhat readable as well you actually can decompose it and put it in another language and like put like verified uh tags on that that everything is is fine uh and spinning up wasm it's milliseconds so it's the user doesn't see it it's not like you have to

spin up the whole complete system gotcha and I think that that's like the uh wasm is the microservices are chains so each individual app chain could be like its own microservice I I've even seen like rollups are the new microservices or vice versa which for some reason that word has a bad context but not exactly sure why um out of wasm VM you're clearly bullish wasm VM um what about move VM uh is is move VM exciting to you uh move brought some a lot of innovation in terms of security inside the space uh like linear storages are one of the great things like actually needing because you know I think people from

Facebook back then with DM they saw that okay there have been a bunch of re-entrancy attacks why don't we stop it like at the basis you know sometimes when you stop some of these attacks at the most basic fundamental level that's the most scalable or you resolve some of these issues and then yeah it's pretty easy to write code and it's uh like having those kind of Primitives built in to to the whole tooling system it makes uh people uh more confident and one other thing it's like all the audits are cheaper like auditing is solidity costs one and then auditing in wasm or auditing in our

space VM it cost like five times less at least or stuff like that for the same code because they don't have to verify all those attack vectors they have in solidity uh that's one of the funny things and yeah these kind of things are needed and when we see outside of cryptospace we see that a bunch of companies which are working with uh highly secure models or environments where you need high security code they add everything inside formal verification chaos testing and a lot of lot of things uh some of those things we want to introduce uh direct uh our VM is actually formally

verified then on top of that your smart contract can be for formally verified there is a tool we are working on into releasing towards the builders and then for example auditing companies can actually use that as well in order to write their own specifications and verify that the contract is actually to their spec to that specifications there are like some some languages are little bit exotic but for auditing companies it's it makes sense it's about a lot of mathematics and when you have like formally verified contracts and formally verified VM you can all you can have like proof Generations mathematical proof Generations easier as well in order to

demonstrate the correctness of of whatever is happening back then uh it is quite nice to see that uh people like took this approach as well we did this as well on our space VM and especially on our assets our assets on Space VM or multiv 6 are are native every asset is native like from fible non-fungible semi fungible Financial tokens then all the operations are uh at protocol level or VM level so there is no re-entrance see and there is you cannot actually lose some

of those tokens pretty easily and you know when you have these kinds of Primitives built into deeply on the layer even composability it's easier the one of the things it's like when you have like native assets is that um you can add multiple op codes into the VM in order to enhance composability for the developers uh one such example is for example in our case is back transfers and then you can have integrating any D app with only one line of code and any defy smart contract you define as an interface you call that interface and then you have all the transfers that

smart contract sent to you as an object and all the output as an object which is actually typ safely verified if any verification fails or the transaction fails so you are safe then you just iterate on the transfers verify whatever that def app sent you is that okay if I'm okay with all the assets I was received then okay otherwise I can fail the transaction nothing happened everything is reverted and then I can move on so we see like derivative contracts building on top of liquidity pools or routers with one line of code or like nft Marketplace buying

you can buy nfts with any token which is listed in in any kind on that nft then it's like he has one line of code to integrate a router which does a bunch of stuff in the behind and it just works yeah that's awesome I think those types of improvements have broader implications for developers users of course and just the expansion of a given other given Network so you're bullish on the properties of mve VM especially regarding the security properties which things like formal verification and realtime smart contract auditing um what about the idea of utxos versus the account based model what are your

thoughts on DMS which use utxos uh instead of the account base model I didn't really necessarily uh get into details of utxo based VMS I see that there is some great Security in implications there because as a user I can just send a few utxos combined together which and I make an execution on that utxo not on my whole account uh which for example compared to evm it makes a lot of sense because you can lose a lot of uh money through hex and uh it is pretty easy to fall into

scams uh but I think it's pretty hard to develop for developers and to switch into a mindset of utxo develop development and I think that if you have native assets at the protocol level you can reach the same conclusion uh for example in our case you can have multi transfer of multiple assets and execution toward the smart contract and then the smart contract doesn't have any access to any other assets of from you only what you signed and what you said that okay I want to send uh one e $100 and one Bitcoin to do I don't know to to do a

derivative and then he can only operate on those kind of assets and those kind of transfers so that's how we resolved it and it's like easier to um it's easier to like kind of go back and backtrack as to where things came from less so than in the account based model um which I think is interesting cool all right um what about the spvm the salon of virtual machine are you are you bullish on the salon of virtual machine or have you uh looked too much into it I looked a little bit to see how their tests and how rust is used there there are some good properties so Hardware acceleration and executing some of those things through Hardware makes a lot of

sense from my perspective it was launched too fast and there are still a lot of things which should be resolved and can be um improved they have copied some of the things from evm and that's why users can be drained on Solana for because of this they somewhat copied the erc20 token standards they made their own Factory contracts through which you can launch your own version but still they didn't have resolved the issues re uh regarding re-entrancy or the issues regarding user facing when your wallet get gets drained um these

kind of things let's say should be enhanced hopefully they will do it because there is a lot of movement around Solana BM but they did a pretty pretty awesome job about parallelization it seems that they nailed it because people can actually write contracts executing in parallel with uh with uh with pretty much ease and then they opened up the let's say that everybody is putting parallel execution this is fine and again using rust is is really fine I see a lot of improvements to be done there it's their job to do it and uh to make make the place uh safe and I think that Solana

it's actually made a case for scaling uh scalable chains that's the first time like like you know everybody is using Solana and it's not like the Legacy chains but somebody something which is really scaling and it it it's a good good thing because it opens up for more Innovation and more development in the in the broader crypto space and uh and the bit VM what do you think about the vit the Bitcoin VM navigating multiple blockchains can be complex and cumbersome let's say you want to deposit into a lending Market on polygon because they have the best rates but you only have funds on optimism so you bridge your funds to polygon only to realize that you need madic to pay for

gas maybe use a centralized exchange or you ask a friend to pay for the gas but only then can you finish your transaction in a chain abstracted future powered by connect this is not the situation you can connect your wallet to any dap from any chain with all the complexities of swapping bridging and depositing all completely abstracted behind the scenes head over to connect. network that's connect. Network to learn more about connect chain abstraction toolkit I come on that's like I don't want to necessarily say but that's like

like really come on you cannot develop anything there like let's be serious that's not a during complete stuff whatever you build build there it's like 100 or 1,000 times worse than you would build on ethereum or 10,000 times worse than you would build on Solana or move or space why that's a fact why is that because there are no op codes there is no support for almost anything um you have to make a lot of uh like uh mathematical assumptions and a lot of operations for nothing everything

will cost huge amount of huge amount of tokens and actually huge amount of money to do that and that is not like execution you know that's not like somebody will say that okay this execution is correct or not and it's no compossibility there are a lot of things adding one to the other yeah it's it's a narrative on the Bitcoin you're not a big you're not a big Bitcoin scaling rollups believer are you I don't see a very good way to do it right now on bitcoin if they changes some things yeah it might happen but you know I want to see scaling or or scaling or l2s like to be real like no

centralized sequencers let's have everybody to be able to run their own nodes to to do stuff to verify stuff um I don't want to get into a place where there are a lot of custodian and I can I am sending rep Bitcoin instead of Bitcoin because I am using custodians which are let's say GP Morgan is the custodian for that and then I can send SATs for nothing because uh they are running the server come on that's that's so far from the Cyber fun vision of of uh crypto and the space let's be real you know the thing is that when you have multiple nodes hundreds thousands of nodes running your system you have

sensor Shi resistance you have a network which can actually uh even if countries are attacking that system it still works it still works 247 but when you have a centralized sequencer or where you have like three five addresses which can do a state actor can simply destroy that and the protocol stops all your transactions are sens and then you can do kyc and I recently read that two biggest problems are token issuance which is some some problem that I I I read about but it was a little bit tough to understand with regards to the

inflation and deflationary mechanisms of EIP 1559 but that the other one was lack of censorship resistance and do you think that ethereum switching from proof of work to proof of stake was a mistake I don't think so so proof of stake is great but proof of stake when uh you have a bunch of nodes and you have decentralization and we see that ethereum has lot of actors there everybody is putting their stake and everybody is running their uh their set of validators so if my transaction is not uh put by coinbase because I would need kyc for that then somebody else will take it up and will put my transaction so I have

sensorship resistance that's fine what's not fine when you have like five 10 running in a system and then those those entities who can run those 510 nodes they can be stopped when you have thousands you cannot stop that's that's the beauty of the chain or or blockchain and these kinds of uh the vision of Satoshi that blockchain could never be stopped yeah I want to I want to come back to the Bitcoin thing because I feel like this is going against the status quo quo a bit at the moment and so there is like this idea of you know kind of like expanding the functionality of Bitcoin and and what I what I gathered was that the the main limitation of the

Bitcoin Network right now is the the lack of op codes and so there's this thing that's been floating around it's called opcat it just got assigned like a Bitcoin Improvement protocol like a bit uh bip 420 are you familiar at all with uh with opcat like what what is the purpose of that op code and and why is this such a uh unique thing when it comes to bitcoin that doesn't traditionally have op codes actually opcat is you can move two times of two types of factors together that's what I got into what basically means and uh actually today I was reading the beep 400 uh four uh

420 in this sense and I heard in E Denver about some people speaking about Bitcoin L2 and a starnet event and it was funny of how they explained some of the Met how you cannot do um you cannot do a multiplication using the op codes but you can do multiple uh multiple uh ads or some other types of mathematics in order to resolve some of your problems I think that it's a start but it's just to that's not enough that's quietly not enough how you will do the cryptographic

proofs how you will do account abstractions with pys for example uh there are a lot of things which the users from from the users St side it will be not good and then you know as it's not gas but you have to pay for every bit of takes data you add into Bitcoin it will cost a lot like a lot lot yeah miners will be happy about it but that's not a scalable way of running yeah this is this is fascinating and I just have one more qu one last question on this which is that you mentioned that you know it's not satoshi's Vision to to have just a set of nodes like five or

six that that can be censored and right now Bitcoin has like this very vast set of nodes um and miners I'm not sure exactly about the centralization of the minor but but are are you suggesting that uh taking like opcat doing these Bitcoin l2s and and bit VM reduces the set of nodes so that it goes from however many let's say there's a thousand uh a thousand uh full nodes on bitcoin's network right now and are you saying that on bit VM it Con it uh it constricts this set of nodes to a smaller set no I'm not saying that I'm saying that Bitcoin at2 as they were envisioned

by some of the teams will be run on centralized sequencers they will be run like you have you will give some kind of custody for your Bitcoin not necessarily custody but you will send to an address which maybe have a multi party computation for that utxo to move outside then you will actually not interact with the Bitcoin Network you will interact with an L2 which is controlled by I don't know that multi party computation which is is fairly limited then if somebody attacks that multi party computation which is like five or 10 nodes or one node then everything is gone you may have an exit tree but I don't know how they will

build it for example on l2s on ethereum you have the L exitor meaning that even if I have my uh tokens in an L2 and the L2 stops for whatever reason for example a network State attack I can go on and get my funds back uh I don't know if that's will be possible with Bitcoin L2 with this kind of op codes and this you know the op cat it's like only the conc concatenation of two values nothing else it's like right it's not multiplication it's not like any of the more advanced types of computation okay so do you think like at is it is it ever possible to achieve the state of like what we have on

ethereum as rollups right now like everything everything Bitcoin you're mentioning is more like a side chain it has its own security uh uh principles and Trust assumptions do you think there's a way that we could scale Bitcoin to where we don't have any any uh lack of the security principles that are inherent to the L1 but we can still get some some greater throughput and and scale yeah definitely if they put space VM or other VM spe especially for for rollups to run like to have for that verification on bitcoin coming soon but like directly at the minor you know they're never I mean

they're not gonna enshrine silly to do something right you have to do that otherwise it's like I don't know otherwise it's like but even if there's bit VM right like it's still it and there's still like uh there's still trust assumtions that are baked in and you're trusting a different set of you're not trusting the miners you're trusting the validators of this L2 yeah actually yeah so you are whatever you write it's on the L2 it's like hardly it is hardly getting the security Primitives of Bitcoin yeah it's it's run by other just one last question man how did you come up with the name space VM

multivers 6 it's like Multiverse at the power of X so we have a lot of space regard or like um like uh names in our uh Vision like ex portal X fabric stuff like that so using uh and the at the basis of everything like let's say in terms of physics it's space so all the running is done on space so space VM and the like the smart contract framework we are using which is a d on top of rust in which you can do integration unit tests interactors and uh formal verification is coming as well that's named

spacecraft SDK as you are building your spacecraft to move towards space with your smart contract and actually spacecraft comes with from S as well so it has the same big Astron energ thank you Robert for coming on and sharing some insights man really appreciate you and your uh your execution layer expertise thanks for having me it was a good chat