Confessions of an MEV Searcher with Libevm

all right um hello um and good evening everyone uh so this is our epl june meetup uh tonight our guest is the one and only lib evm uh who is uh one of the most uh well-known uh mev searcher chad slash uh anime girl on crypto tutor and uh before we get into uh talking to him i will do uh some quick uh monthly round up and announcements uh so as usual our sponsor is escan if anyone is interested to join a malaysian team that's um very active in the crypto space uh feel free to reach out um and to start things off i am just going to get uh nick to share screen my sharing screen or i think i'll use your slides because i actually don't i didn't download when to click next all right thanks harriet and this is the 8pm news for everyone else for this month all right so very very hot off the press just yesterday is the robson merch so robson is one of the first public test nets to go through the merge where we test the the integration from e2 to the current interior as we know it it went successfully i think pretty much nothing very serious happened yesterday and we managed to mine the first um the first e2 blocks on both layers so if you can see the small picture on the right there it's the small eatstaker what do you call it uh what's the name for it watch party so um me and chung were in the watch party yesterday i'm not sure if anyone else joined um there were some cool people i think vitalik joined um patricia from poe atmos there and guest team they're all there so yeah after robson has merged they are next going to test if i'm not mistaken goalie and then we're going to test the poliah so then after supposedly it's going to be the real thing they're going to merge main net and we have finished the e2 upgrade so for developers you should not stay on robston actually because robson is going to be deprecated by the end of this year so if you can and if you're testing any if you have any applications that are being tested you should actually move to either goalie or sepolia next right next paypal announced that they allowed the transfer of crypto so i'm not sure how many of you still use paypal but at one point we could we were allowed to buy cryptocurrency using our cards so then just about last month i'm not sure of the exact date they announced that you will be able to transfer your currency from different paypal accounts which is not an achievement we can do that with metamask but the point is that it's integrated to a lot of the checkout options that we that we know so um it's probably a sign of adoption that you can start paying for crypto and moving it around like a wallet just that you don't have to manage your seed phrase with it next next is a very interesting case of pull together i'm not sure how many of you use pool together here it's the total of d5 so pull together is essentially for those who don't know is a i describe it they describe themselves as a no loss lottery so meaning that when you put it in when you put in it's like gambling just that you don't lose so essentially the fees that they earn through through what they call it through your d5 deposits get distributed to the winner but you would get your deposit back so the cool thing is that in the u.s someone figured out that this was not legal and deposited an amount of ten dollars into this protocol they eventually played around with it and decided that this was not something that aligns with the us regulations and decided that they're gonna launch a suit to pull together so to support pool together pull together launched this very very cute profile picture i assume it's a parrot it's a purple color parrot for supporting the lawsuit and whatever that all the legal costs that they're going to go into um defending pool together and if i'm not mistaken the starting price of mint is 0.1 ether then upwards of that you've got different levels of this um parrots i think the the most expensive parrot is i think 75 liter so yeah that is if you want to support pull together being sued by someone who deposited about ten dollars into the protocol all right next oh i remember there are more there's a local oh there you go you accidentally click too fast okay so then this month optimism also finally deployed the airdrop with a very very cool contract address if you looked at it and so those of you who got the airdrop uh feel free you're going to claim it earlier earlier last night also hope protocol also announced that you would be able to claim your eye drop tomorrow they've also deployed their contract address so you've got two eye drops this month but as the saying goes your ad dropped responsibility and not free money next going on to more local things in malaysia um cz the ceo of binance actually came to malaysia here i'm not sure the actual date and the actual event but um the cool thing is that he met up with strategic who gave him a very very beautiful buttock shirt on the right so yeah he was in malaysia and i think he met quite a lot of projects and he was speaking to some regulations here about crypto and all things e5 next and finally a follow-up from last from last month's roundup with kfc another adventure park called expire has started a campaign where they start where they start these rail utility nfps so by owning one of these expat nfts you would get that's a bit smaller tanks but i mean you would get two you're not mistaken okay maybe the tails are not too clear but you would get um something like discounts or free rides for owning this nft so this is cool this is one of the more this is one of the cases where you see nfts being adopted and used in asia for cases where it's memberships discounts and it's a follow-up from last month's effort which we saw from kfc malaysia all right that's the news for june everyone and back to you heart all right thanks nick um yeah so a monthly roundup and next is a sharing uh by chi chuan i am digging it yes yes uh can you guys can you see my hurry yes awesome uh yeah hi everyone my name is uh chichwan uh i i don't know uh probably some of you have seen me before and today um i'm here for a different reason so i'm here to let you know that um starland is coming to southeast asia and malaysia we are trying to uh stanley's trying to build a community here in malaysia and i'm here to let everyone know and so that everyone can join the party so for those who are not familiar with what starnet is there is a quick uh introduction i'm going to go to a quick production so what start net is basically a technology by a company called startware and it's a layer 2 scaling solution and we know that there are three forms of scaling solution out there one is sidechain a polygon is an example of sidechain you you may argue against it but i think it's a side thing uh optimistic roll-ups such as uh optimism and arbitrary and startnet is basically the third option that we are going through today here roll ups but to be technically correct it's kind of a validium roller and i'm not going to explain the differences but just let's assume that it's going to be a vk roll-up and unlike in optimism whereby it relies on crypto economics to ensure that executions are done correctly the third option the zqu lab relies on math to ensure the validity of the execution so so here is a quick picture of how how it works so basically there's proven a verifier the approval being startnet the verifier being an underlying l1 blockchain for example ethereum and you there's a huge bunch of execution being done by approval and the approval the start that you only need to send like a small proof more size proof to the uh layer one blockchain and the one blockchain is able to verify that hey that execution is being done correctly through this little proof and the proof it's what we call a start proof and hence the name startnet and startware and it stands for scalable transparent argument of knowledge and why scalable because the proof size increases exponentially slower to the computation size so for example i'm just giving like a rough very rough example probably not accurate if we have uh we are running a computation that runs for an hour and uh the time taken to validate the proof is probably only like a four three or four seconds so that's that's why it increases very very slow compared to the computation size so that's something that we want we are able to get like a very small proof and therefore uh after after sending the small proof to the layer one blockchain the guest fees that are required to verify it will be also very it's also very little and and it's also not a snap and some of you will probably have heard of smart uh because splux doesn't require a trusted setup and therefore it's called transparent and no trusted setups means that we do not have to go through the hassle of like a like a trusted um like a like a ceremony and we have to ensure that um this thing called the toxic toxic waste has been discarded uh responsibly by any of the participants and so on and so forth so that is a hassle and people don't like that and therefore starts has this transparent property that is very nice and we don't need that and finally uh the there's this uh nice very nice property quantum resistance uh that that smart doesn't have and there is uh people are saying that if quantum computer were to mature in the next probably 50 or 100 years uh they have to know they have to um have no choice but to rely on stocks ultimately because that stuff relies on like um hash functions and whereby a snuffs relies on discrete log problems which can be um can be cracked by a quantum computer yeah so enough of those like fancy explanations so basically what you need to know at the end of the day is that starts you just give the guests be a slap in the face just like how a friend of mafia is doing and yeah so that's why it's a layer 2 scaling solution and there's also something new something for startnet and uh basically they have created a language it's called cairo and um cairo basically is created because it has this yet it's constructed in such that the computations are able to um how can i say create proofs very uh efficiently in the start environment so basically they have to like rewrite everything uh rip off everything and create their own architecture so that they have this computation that is uh that is very able to generate stunt proof uh friendly but yeah that is uh something that is uh it's hard to learn and but it's very critical but it's pretty cool if you believe in stocks in the future and yeah so if those what i mentioned like starks and cairo and whatever pick your interest uh we have a startnet talk coming up next thursday at 8pm and anyone is free to join uh we are going to we are going to host henry and he's a deaf advocate from stuntman so you're free to ask him any questions that you want and i'm pretty sure he i'm very sure he's very happy to um to answer any of the questions so i'm being joined by carlos is in the call and also by the th uba it's a king hua university blockchain association so we are going to like uh have a one-on-one session with henry yeah so uh for those who want to join feel free to join i will share this uh i'll share this later in the chat uh yeah and probably the telegram group also so last but not least we have a we are planning to have an in-person uh meet up in kuala lumpur in kl we are going to try to build up a stuntman committee malaysia and it's tentatively going to be on the 2nd of july so here is a qr code to the link of the google form just register interest because that will makes planning easier and suffer that so so yeah um finally i have to end the end this with a meme so basically thanos after getting start net uh gains an infinite amount of power and yeah thank you i think that's all all right uh thanks to each one uh so yeah um thanks for waiting patiently everyone else uh for our one and only guest uh evm uh do you wanna try and share your screen to see if um it shows up properly okay yeah give me a second yeah and in the meantime i'm just gonna do a very quick introduction for uh for lib evm um so uh yeah his uh twitter account says he's a professional at alfalfa capital maybe you can share a little bit more about that um and uh yeah so he's been a very active uh me researcher uh one of the earliest ones and for those who don't know any researchers are people who find uh opportunities uh of mev to uh essentially uh be touched and make profit um yeah i don't want to [Music] say anything more in case it doxes things or not so i'll just let you go ahead and take it from there i think you just signed in with another device looks like oh okay uh looks like oh hold on let me see [Music] as always i am gonna ask you to a new okay can you hear me okay great um let me try and share my screen [Music] okay great all right so uh yes uh hello everyone uh thank you harriff and oh can you guys see it yes yes i could see it but just a very quick i think i forgot to mention uh just for your own reference as well uh so we did a very quick uh survey of uh attendees uh for tonight's meetup so roughly uh just to give you a sense of the kind of audience roughly two-thirds out of 40 said that they are noobs and almost all of the remaining said they are okay okay only and maybe like two people were brave enough to say that they are mev pros so just to give you a sense of the level of your audience oh great because i'm not a pro um that's great i'll try and keep things as as simple as i can but uh thank you harif and thank you caleb for uh for having me i actually went to the etherscan office in kl and it was a lovely place so thank you very much i was very intrigued by the whole setup and uh yeah thanks everyone for coming this is my talk this is a confessions but any researcher it's not technical it's more like a storytelling i would like to say a narrative on how i navigate navigate at the dark force and kind of how you can as well um so a little bit about my background um i'm just an average curious guy i'm also from malaysia uh zawan or whatever you want to say i'm just a very average curious guy i don't think i'm really anything that's special at all in fact if you would classify me i would probably be on the bottom like on the left curve on like this uh this iq curve i'm not even on the mid curve because my friends i can see them on the mid curve they think way too much before they execute but i mean i don't really think i just go and execute so i i want everyone to keep that in mind um when i'm giving this song because i i just execute i don't think too much uh and i've seen people on the right those guys they're machines and that is definitely known so that's just a little bit about myself so the point that i'm trying to make is that if i can do it like i'm pretty sure any one of you guys can do it too like it's nothing special it just requires a little bit of patience it just requires a little bit of elbow grease so it really is nothing special at all and i would like to begin this talk officially with a little bit of story time as we travel back in time to d5 summer 2020 where things were a little bit more optimistic than they are currently when pool 2 also known as eternal lands was just a thing and for those of those who weren't around that was the inception of yambo finance where if you deposited yam lp tokens paired with this other token called ycrv which at the time i don't know why it was called ycrv but it was basically curve finance which is a market maker where you put in like why unstable coins for example y and di or y and usdc y and usdc and pair with yam on the uni swap liquidity for i know there's a lot of words a lot of made up words right now basically if you put yam and some stable coin yield bearing stable coin together in unisop v2 port and put it into this uh pull two you earn yam so this is like one of the first time anyone has kind of done this this kind of very degenerate degenerate pool so things were a lot more optimistic it was it was the inception of the summer 2020 let's go and so basically for those who went around food yield farming is kicking in high gear so we had like yam finance you know finance sushi sauce pickle finance hot dog kimchi ramen you name if you can think of a food it probably happened and it probably had over 10 million dollars in it and it probably lost 10 million of it overnight but hey who's counting um and also something that i want to kind of emphasize at this point in time is that dye the decentralized stablecoin was preferred over usdc at this point in time and i suspect because the d5 narrative was just kicking into action so the d5 meme was kind of like oh it's decentralized compared to usbc so we should use style like that was the whole narrative at the time right now it's a bit different because everyone prefers stability it's like oh if i have usdc i could redeem it for die sorry i can redeem it for actual usd so we prefer stability over say decentralization and we can even see that on the price chart of dye um so for example uh this was the period in in bread in the red rectangle as everyone can see that was the period during the five summer and that was the price action of diet which was supposedly a stable coin and in fact we got so much demand that it often went over a dollar oh and for everyone around who isn't um into stable coins the stable coin is supposed to retain uh a stable value in this case it's supposed to mimic or have a stable property of the of one usd so if it's a bit um if it's uh if it has a market premium and it is being uh being sold on open market for over a dollar it's not really ideal for people who want to buy it's great for people who want to sell but there's no great people so that's the whole concept of like a decentralized dollar anyway yeah but decentralized stable pretty much died and so how die worked back then was that it had something called a cdp a collateralized network so back then um and and i was being run by this governed by this protocol also known as make it down which accepts some kind of collateral it's almost like a mortgage you like you put in some collateral and you take out a loan and in this case your collateral would be ethereum so you put in like ethereum and you can take out alone and die and so the theory is that if you over collateralize the collateral we can always redeem um dye for like a dollar with its collateral which is why you had to a maker had this over collateralized it was a dye was also known as the over collateralized stable so in this example if was probably in back then in naked eyes case in its cdp the collateralized that will have like 150 percent collateralization ratio what that mean meant was that in honor 10 in order to take a loan of 100 die which is like a hundred dollars you had to put in 150 dollars worth of ease and so let's just assume that one eve was worth a hundred dollars in order to take a loan of a hundred dollars you needed to supply 1.5 each and you can take a load of 100 die and then because dai was trading on the open market for roughly a dollar you could also change that 100 dime for one each and so notice the pan and now you have one each which can put into your collateral and now you can take out a loan of 66.67 die and now you can convert 66.

67 die into 0.66e and kind of keep leveraging your uh your you can keep leveraging your your uh eve collateral if you're if you're long eve of course if you're the east price tanks you have to repay you have to either repay the loan so you're always above 150 characterization ratio otherwise your position gets liquidated but but that's not important but the key thing that i want to share is that maker thou back then only had cdps but as time went on make it out transitioned from cdp's to collateralize default into this thing known as the mcd or the mountain multiple collateral pools the mcdp is mcp and one of them basically um instead of only if maker now allows you a multitude of collaterals in fact we can't even check it on dyestats.com i don't know dyestacks.com i hope it loads and you can see the different types of collateral i really hope it loads i'm not sure yes you can see the different types of collateral it accepts this now if a if b if c you can even get wefts with all these random made up names you can even put in bat actually not anymore usdc you can get uh wbtc so you can mint die you can take out the loan of diet and wptc t usd k and c anyway the list goes on even like more funky stuff like uni if e2 lb uni v2 lp tokens i know that that is a huge mouthful to say but that's essentially it's multi collateral thing in essence instead to recap instead of just having if you can now collateralize the other assets and one of the assets that make it out allow you to collateralize this thing called usdc just a second i'm gonna take a take some water by the way and so usdc was a big controversy in this whole um in the whole governance system because the whole point of that was that it was a decentralized stable coin but now if it's being um collateralized largely by a centralized stable coin then it kind of defeats the purpose it's kind of like decent it's like a decentralized coin but only in me because usdc could potentially just blacklist the house address and make it the house maker that would immediately be under collaterized by like a large large portion but anyway make it out basically had inc basically over the um despite the and despite the strong opinions uh usdc was eventually added to a collateral meaning and and with a collateralization ratio of like 110 percent what that meant was that if you wanted to take out a loan of die like a hundred die you needed to supply 110 usdc um which is much lower than the requirement for eve and keep in mind um we're assuming that usdc is a dollar but in reality it kind of fluctuates between 99 cents and like a dollar and one cent frequently but you guys get the point uh just recapping a little bit die also during this period of time um the high was also dye had huge huge demand on the open market like it was it was actually being it was at such a huge demand that there was a premium uh premium for it like it was going above the dollar uh very frequently during the defunct summer oh and something i forgot to mention is that make a dial um how it tries to keep the eye stability die at a dollar is that it has this thing called interest rates so meaning that to take out a loan of die you're actually being charged interest you actually have to repay more die than your loan and during its initial phases make it down make it out have like an interest rate or 20 percent use die because the uh pretty sure like the uh the the incentive to mint uh wasn't it wasn't that high but in order to create incentives for people's mint it would lower the interest rate so more people would actually want to die but when device supplement was happening the interest rates were already at zero so make a dial didn't have any more mechanisms to kind of create any any incentive for people to make more die so what make a dial did was that it decided to change the collateralization ratio of usdc from 110 to 101. but i just want to run a pause for a moment and think about that for a few seconds like changing the categorization ratio from 110 to 101.

i will uh like i'm not sure should i check the chat let me let me open up the chat and see if anyone has any questions on the chat i wonder why die is only one technique i'll go to the view mode remote gallery gallery see everyone all right no one has any questions there's no usp joke here but there's no usd joke does anyone does anyone know can anyone think of anything that could happen by changing the colliderization ratio from 110 to 101 um you get liquidated no no no no okay i'll just i'll just give the answer all right so die is trading at the open market rate of between a dollar and a dollar and like three cents but in reality it was actually a lot higher but the minute you change the collateralization ratio from 110 to 101 what that effectively means is that in theory die should never go above a hundred dollar one on the open market but it still happened so now what that what that essentially means is that you can arbitrage the difference from mcdowell and the open market because if you have a hundred and one percent collateralization ratio what that means is that you can always supply one dollar and one usdc and get up one time and i'll make this point a little bit more concrete this is also known as a bespoken path and i'll make this point a little bit more concrete by providing you an example let's say if die is on the open market if that on the open market is trading for more than 1.01 usdc and let's say on curve.finance here as we can see it's trading for one dollar and two cents i know that doesn't sound like much but we're talking about hundreds of millions of dollars in liquidity here so this changes a lot of things what we could do is flash on a 100 die now remember die on the open market is 102 and we'll assume that the liquidity that is deep enough to handle like let's say up like five million dollars or a million dollars but for simplicity we'll just start with 100 die we'll flash on 100 die we'll swap it to 102 usdc on curve and we'll collateralize 101 usdc not 102 but 101 and will mint a hundred dime maker and it will repay the loan and it will profit one usdc so this is purely purely from a governance change on make it down now i know one usdc profit doesn't sound very appealing especially since you're like you have to pay for gas fees you have flag right up the contract you do this blah blah blah keep in mind the flash number 100 die on dydx back then was completely free if you knew how to do it like this loan then you did not have to use rv so it was completely free but um try and extrapolate this to millions of dollars try and extrapolate this to uh like five million dollars and just think about the profits that you could be you're making like one percent off like no like once then i'm like just woke up so math is not my best subject right now but just put in two more zeros uh you're flashing uh ten thousand die now you get a profit a hundred usdc you flash loan a hundred thousand die you get a thousand dollar profit you flash on a million died you get ten thousand dollars profit and this was certainly certainly very much so possible during d5 summit and keep in mind this was back in d5 summer before the aggressive memphis bots before the aggressive uh actually there were some front running parts but they were not sophisticated but this was very much possible two years ago and and i'm like sorry yes this is very much possible two years ago so you know you might ask like what's the point of telling you these um what i'm trying to say there's always always more opportunities the market is super super inefficient in fact when me and my friends did it we thought it was like impossible like we were actually hand-harving it we were like eyeballing the it we're eyeballing like a marketplace we didn't even have a robot because we were so afraid of participating in pgas otherwise known as like i was at uh priority gas options this was pre-flashbots we were really afraid of like okay what happens if you don't pay gas okay but the opportunity is asymmetric remember you're paying like maybe fifty dollars in gas but your opportunity is like ten thousand dollars like five thousand dollars seven thousand dollars could be five hundred dollars like the opportunity was that and in fact um i remember even after maker lowered the collaterization ratio down to 101 that window was not like immediate like the die didn't went down immediately to 101 usdc it took them like a few days like those few days were incredibly insanely profitable like the the market you would think the market is efficient but it is stupidly inefficient so i just want everyone to keep in mind there's always always more opportunities like like do not do not please like you have to enter this game with that mindset there is opportunities okay so how do you get started i guess uh well uh you got to pick a poison here you have two options long tail one short term um opportunities by the way uh that's what i'm trying to say also keep in mind this was back in a more optimistic time when everyone was happy d5 summer 2020 um very different right now in the bear market but that is exactly why you should do it right now because everyone's bearish everyone thinks there's no more opportunities left but in truth there is always opportunities you just have to be you just have to put in your elbow grease to kind of find uh to find find how to find one how to find it and who how to actually execute on it properly when i pick your poison uh long tail and shot two opportunities so let's go um they should say short tail versus long tail i know it's a bit confusing so edit it right now short tail versus long tail right so it's for everyone for for those who doesn't understand short tail opportunities are opportunities that have very high volume but very low profit margins long-term on the other hand opportunities are not frequent but have insanely high profit margins and one example of the short term opportunity would be univ two arbs so arbiting between say univ two and sushi swap or like sandwich attacks because these opportunities are like really really easy they're really easy to conceptualize and the most popular popular extraction like value extraction strategies they are out there you could easily like commit that's what people think when they think of like mavs like sandwich attacks or like back running so you're competing everyone's competing for that one opportunity so everyone like it's incredibly competitive your profit margins are going to be insanely low but the flip side for that is that if you can win the small to opportunity you'll realize you're good enough to participate with the big boys like it's always a mindset thing like these things are always a mindset thing so in that one example of a long-term strategy would be um like a few months ago like the eighth the eight coin uh the eight point uh incident where someone basically took our loan of all the products from a liquidity pool claim all the eight tokens and refunded um refunded the liquidity pool with of 2.5 each but they got out with about 100 leaves paid the miner like 50 so they profited like 50 plus 40 e for 50 plus eve within like 40 seconds i think i think not too sure but they made a huge chunk of cash that's all i knew um so but that only happens once and it's like it's very these opportunities are incredibly rare the other the other one was like the me bits i think it was like the me bits incident also nfts but these opportunities don't come very frequently but they're very very very lucrative because these require expertise in the space they require you to be constantly up to date with like the latest obscure protocol like for example ball apes which i was not uh and like with like the most obscure protocols because remember you know as as was what's his name uh one of china's great leaders one said you know i don't really care what color the black the cat is black or white as long as it catches mice it's fine for me i don't really care what chain what protocol you know however i get my money as long as it gets money you know it can be a shitty protocol you know something that's going to extract value from it that's what matters so yeah long tail short tail um yes so a lot of people will ask okay how do i get started so i i'll ask you back the same question like what's your strengths what's your weaknesses is strength infrastructure is your strength coefficiency is your strength recon a refund um i guess reconnaissance like all right how do i discover opportunities how to like find how do i find out like what's what's trending or like uh like what's the next thing that people don't really know about because keep in mind uh theorem is a public database and within this public database every single opportunity that's being executed is public even sex to decks arbitrages even sex index arbitrage you can find these weird patterns you can compare them to binance order books or ftx autobooks um i'm sure you'll see what's going on you just have to have you just need to put enough elbow face into it but if if you're good at infrastructure try playing the l2 game like a polygon uh polygon uh phantom avec and maybe not avex because it requires huge amounts of capital but try playing try playing that two games if you're good at code efficiency and gas golfing like uh ethereum l1 uh sorry ethereum with flashbots is probably your best bet simply because um you are kind of playing against people uh the the whole infrastructure game has been leveled up by the flashbacks team in fact i would actually the the con is that getting started on ethereum is a lot more expensive like uh capital intensive because of the deploy contracts it costs you like a couple hundred dollars usb which might not translate well to myr so if you get someone to give you some if i would say that ethereum l1 like sorry ethereum is actually a lot easier to get started with but infrastructure wise if you're good with infrastructure wise like l2s l2s would be a lot easier yeah recall yes my favorite quote from my favorite person um good artists borrow great artist steel just in sun 2020 during the d5 summer phrase uh but essentially what then i would like to extrapolate from that is that you know again as i mentioned ethereum or a lot of these lock spaces uh a public database so everything that happens is public information so no does anyone know how to extract any information from here well you gotta thank the etherscan team because now you can easily see function signatures decoded or not on a very easy to use interface so for example right here i'm just extrapolating i don't really know how true this is but if you see a a random uh four byte signature going to a random address like a basically unlabeled address chances are that that is a someone extracting value from the blockchain is extremely high in fact we can check it out right now let's go to bse scan the most popular explorer from each scan and we'll just go to a random block 201 transactions let's go and let's see if we can find one with a random yes we found someone after charging by the way 41 seconds ago this transaction i assume banking library invalid path the contract yes someone was arbitraging but someone was also failing very hard um but the the point i'm trying to make is that you can find people yes they are in fact arbitraging um this is in i'm assuming in with 46 out with 73 in fact he's taking a flash loan from sorry like a a loan from panking swap so i'm not too sure what the direction is but i'm assuming he's making money otherwise this whole arbitrage process is not working but you can find people doing arbitrages simply by going on etherscan's website like finding some random four byte signature that is not labeled in fact sometimes people might label it as uh sometimes it might be labeled but you gotta kind of understand the nuances like if you see like an unlabeled contract if an unlabeled fallback signature it is very likely that that is a quote like contract and you can kind of dissect it kind of see what it's doing like this is incredibly important because a lot of people you know they think that oh it's a it's a and in fact i want everyone woohoo someone's doing it i i can't beat it but keep in mind that this market is incredibly inefficient i i'm sure having two competitors having two competitors does not necessarily translate to you not being a prophet um so yeah so you could do all these elbow like all these elbow grease you know try and steal people's uh still people's alpha like try and try and find out what are people doing like scan the blockchain so you could do that or you could just join a bunch of obscure protocols on twitter and discord keep like keep up to date with that twitter and discord like how i did like with make it down which was not necessarily a obscure protocol but it was it was like a governance process right lowering the collateralization from 110 to 100 and like one you need to see the opportunity i actually in all frankness i didn't actually see my opportunity i had a clone friend saw the option right away uh i i i didn't even realize the implications of that my quan friend is like oh so yeah so die would never go below 101 usbc it's like a why but now i'm looking back with hindsight because that's that was very obvious but it didn't make sense to me at the time so you got to be familiar with the mechanics of the market with the mechanics of these things and so i guess tldr is like join obscure protocol discord like join oil finance or back then it's like i don't know like uh like what primitive finance like even though it's not popular you might think all right the tv i was only like 10 mil but you know if you can get like a thousand dollars each time to to do some random ass uh i guess maintenance thing or like some unit testing for them or like some kind of ad advanced like esd i don't know if everyone's around for esd like but you call you call function advance the certain time talk and you get like 100 so like it doesn't matter if the cat is black or white as long as you can extract value from it that's all it matters like it doesn't matter how big the protocol is you can get a hundred hundred dollars or make it now it's just as good as a hundred dollars really there's a protocol that only has only has like one male tv alright it doesn't matter it's still money well in my book it's still money say join obscure protocols you know join the discord let's keep twitter if you if you really want to put in the time that is like keep up to date with like governance like say sushi so oh we're gonna have this uh what governance we're gonna merge with wire and and so maybe sushi will come i don't know if we're gonna we're gonna merge with what was it again again abu kadabra or spell okay so spell my palm or sushi might come i don't know there's also little little tidbits of alpha like this by keeping up to date with all these random protocols oh yeah so in short uh this is like the summary of the whole talk um as a confession i am not the best in a lot of these things in fact i'm just an executor so uh realize that the market is inefficient please i personally think i am on the left curve of the iq curve avoid imposter syndrome like i think the mindset thing is really important you need the confidence you need to realize hey i can do that too market is inefficient uh pick your poison you know when you want to do a long tail shot i mean you could do both um also yeah you could do both it just depends on how much time and effort you would like to expense into this and uh last but not least tell your strengths find a teammate team size of two to four is like ideal like because you guys can complement each other's weaknesses because i was good at the contracts i was good at executing on the infrastructure contract side my friend was good at finding opportunities so we worked extremely well in teams yeah that's kind of it from the topic thank you for listening to me rambo for 20 minutes if anyone has any questions we'll open the chat right now and see if anyone has any questions uh okay so oh yes um maybe um before we get to uh audience questions um generally i'd like to start with a kind of a quick kind of backstory for you um as much as you'd be okay to share like uh would you uh be okay to share with us like your story uh before you went into ethereum or crypto uh and then when you got into uh mev uh on a more personal side other than what you shared earlier oh yeah sure so i was just very curious about this funny money space actually i started mining deutsche bank in high school because it's like it's a doggy coin and then i guess i didn't really take it seriously until like i met someone who told me hey you can program on the blockchain and that was the most thing i've heard you know a program on the blockchain but uh yes here we are actually programming on the blockchain um um yeah and then i slowly looked into it and said oh this ethereum thing looks kind of cool and there was no real like i think i think the important thing to note here was that there wasn't any like money incentive when i jumped into this whole thing it was genuine technical curiosity all right how did the bitcoin protocol what are they how did the p2p system how did you do proof of work how did you actually how did the consensus theory work and how did like all these cryptographic things like how did it all collapse in place so i was really interested in technicals and then ethereum was like the next level up because it had like a virtual machine blah blah blah you know it was like it was like a playground very experimental very raw and i personally back at the time was doing machine learning and i don't know if everyone was there it was around for andrew andrews days like machine learning on coursera i did the course uh i really like machine learning you know cnn's back then there's like a hot thing but trying to go into the field was extremely painful as someone who's not very academic like for every conference i go and we went to a couple like people won't even look at me because i had no like ivy league university degree i was just from this randomized local australian university i was i grew up in australia born in malaysia a little bit more context but yeah this random australian is the non-ib league um and you know i didn't really get any recognition also i did have a phd i had a bachelor's and i barely finished my bachelor's i was like still in the process of it and so i was like you know what it um if you know i'll just go into this crypto thing since i could not really i could not see like a career progression into machine learning unless one i go into iv legal universities or two i somehow get like a phd which was not on my roadmap and i guess i initially got into like zero knowledge stuff but as d5 summer came and came along i was like yo i should probably have a better grasp of finance that was really it like because i was really into tech but i also wanted to you know understand the finance law and you know bit by bit like interactive these protocols i actually really understood a lot more about finance um but that's that's kind of how i got into this mev space and like opportunities are kind of kind of started coming out around and yeah you just execute i really want to emphasize like a very left iq the fact that i'm not an ivy big universe i'm not i'm also not from chima university so yeah very left curve please all right um yeah so i think uh so i listed on some questions from the chat earlier so chiranji asked can you explain pga a little more and maybe if you can also expand like other methods of um doing mev uh so pga is a priority gas option what that means is that this was pre-flashback stage and there was no like bribing happening behind the scenes so every miner was quote-unquote honest you know like as long as you follow the rules of um the ethereum protocol which was like okay i guess you pay the more uh and the more you the higher guys you create the faster your transaction gets mine so what that meant was that when you saw a transaction happening that was doing trade on say the univ on uni swap at let's say 50 what and you want to do a background in ideally what you do is immediately on all your um all your bots or your on your infrastructure you will send out a transaction saying okay i want to send a transaction 49.99 recurring um uh to do this arbitrage right after this swap because after this swap the univ2 pulls are inbounds and what that means is that the opportunity only happens right after this trade um so you want basically your your gas your gas you you're using gas option gas option system as a way to make sure that your your transaction is being included right after that particular transaction which creates the opportunity um this is very hard this is actually very hard and in fact you lose a lot of money and you need to like go into like stat hubs this is harder back then compared to let's say flashbacks because you basically it's it's riskless with flash but it's pretty much almost restless unless you get an uncle block it's almost riskless because like either your bundle gets included or it doesn't get included but for priority guest options there's a high likelihood that 10 people gunning for the same opportunity nine of them will fail in fact uh there's a in fact if you checked the etherscan link back then i'm not sure if i'm still presenting there's a reason why there's so many fails on like the bse scan show before just because pga is still happening on the bsc scan as let's say there's one opportunity but it's like 10 participants nine of them will fill only one of them or succeed so basically tld i guess sorry i'm going to pga is by you using a gas option system to ensure that your bundle or your transaction gets included in a very specific section of the block i think that is the definition that's my definition i know some people might disagree with that but yeah yeah and i think chiranjit mentioned earlier like uh before 1559 um making transactions was hard and i think uh when before flashbox came and everyone was doing pga it was like even harder because essentially uh people kept spamming the um the chain right to make all these uh transactions um so yeah but uh having said that um there's a question here from search which is uh related um uh since uh flashbots doesn't cover every block uh does pga is it still around on ethereum yeah so that's the other alpha so so let's say you see like a like a 50 000 opportunity you there you could either go to flashbots or you can try and include it in a normal block if it's a 15 000 opportunity i would definitely go flashback and try and include it into the into the normal block with like the normal pga method simply because the opposition cost is so high yes the transaction might fail and i'm losing 500 bucks but because the opposition cost is so high like over over like let's say a long enough period of time i know that i will make my money back because uh like flashback i'm pretty sure like flashbacks has like any 40 or 30 of the total cash power this is this has been a while i haven't really checked it recently so like if you could you could send two transactions out and yes you could try and get the transaction included in the normal block so you could do pj on a non non non uh non private block it's a quote quote um you just need to make sure well you don't need to make sure that you make money you could donate money to the system yeah but ideally you want you want to make you want to extract value right but yeah there's some tricks on these kinds of uh these methodologies to kind of kind of utilize yes statistical methods to kind of make sure that hey um over a long enough period of time you know i'll come up positive yes so yes to your question yes um all right then so another pga is using relays uh such as flashports can you explain a little bit more about the different roles uh that are involved in this game like you have the searchers and and a bunch of other people so so you have the protocols which create the opportunities or create a value where you can extract you have like the the miners which uh participate in flashbots so uh that part is a bit opaque because if you knew how the system worked you could game the system so mine is in flashbots i would consider like the infrastructure side of things or they'll be under one uni unit that part i'm not too familiar with um and in fact if i didn't know i'd probably be under an nba to not say anything but i'm not working for them so i don't really know but yeah so you have the protocols you have the infrastructure the new searches and the users the users which uh make the market incredibly inefficient like as we saw from the ust debacle someone fat fingered on i think it was arbitrary and then withdrew one of the lps they basically were willing to take like a nine million dollar hit on um on the ust pool taking out like only usdc instead of like an equal proportion of everything so they took like nine million dollar hit i believe it was it was like seven figures seven because it fit seven figure hit and someone opted immediately um by the other market please realize the market is incredibly inefficient like everyone is dumb like including me sometimes i fat thing fat finger happens like every day so there are opportunities happening every day that's like it's like a mindset thing um yeah so anyway protocols infrastructure users and searches so i would say that's like the stack um yeah all right then um so related to that uh cook's question earlier was uh what do you think about prospects for protocols like cow swap that attempt to use mev by direct approaches um well i'll let you explain a little bit more about household and then what your thoughts are about them oh cloud swap is basically you sign like this it's really well set up it's like i think it started out with like some kind of mvv protection service but essentially what it does is that instead of sending a trade to a the public mentor what you do is like sign this transaction saying i want this trade to happen at this price point and then you send that transaction over like that signature to counsel that means a central service now so it's a trade-off right give or take personally i don't really care as long as i get the best grades but some people might disagree um but with that signature and then that signature will be kind of used in the contract saying okay i want to be guaranteed these rates if these rates don't happen then then my signature will not be uh would not be a valid according to the contract so what counts up does behind the scenes that it tries to pair you up with other people who are one either doing the same trade or two trying to find background opportunities after your trade and by doing so they kind of basically let's say after your trade they're able to extract fifty dollars worth of value and the transaction was free for them and they'll do like batches of these transactions at one goal to save on gas and essentially find the most optimal route so that's a little brief history on kaoswa what do i think about it i think it's great i think i think it's incredibly smart but that also means that such searches would i mean the more people use it then um i think kaosok has like a there's an opportunity for you to be like a counselor searcher meaning that you solely perform uh root searching which is like a short tail uh short tail opportunity on cap on count might be more profitable i don't know i haven't really tried it simply because like there's a lot of users there which submit through the the website so those trades might not those trades will not appear in the public mempos so i don't know it could be more profitable could not be not from from my perspective but the question was attempt to reduce mvp by reducing the mvp by measuring a lot of trades each block off chain yeah i think it's great you know but uh you know i always think there should be competition um so hopefully someone else also comes up with a better solution that someone else is probably not not me um i'm too i'm not that big of a brain to set up a whole new protocol and the balancer team is really next level i just want to say so and the counselor team i think that from the same uh they're the same people i presume but no i think it's great i just don't think i'm smart enough to do that i think there are similar protocols as well like there's the alchemist um and maybe one or two other similar like exchanges that you can trade oh yeah you're right there is alchemist um there's alchemist but i don't think i've used it but yeah well how sword seems to be the more popular one um all right uh flipping on to a different uh topic so there's uh at least alex and chiranjit uh asked this question so i think i should ask you this um so the question is uh how did you personally go around assembling your team so me it was actually a friend of a friend and then we kind of just hit off really well for me it was it wasn't it was by chance that we found the team in fact i was really just arbiting back and i was like loving's evil i shouldn't do that but ah you know we're back running we don't do sandwiches back running is good you know we're creating we're making information more symmetrical but i know that a lot of other my other friends found their teammates was that it was actually through the flashbacks of discord like i know it sounds so weird but like yeah like trusting this anime pfp behind this chord with a huge amount of time and money but you know it seems to have worked out like i think you can kind of gauge uh you can gauge personality or like how trustworthy someone is simply by interacting with long-term okay and even if you don't trust them you can encode it in a contract like the profits are split immediately after after it happens so even if you don't trust them you can encode that into the contract but that of course means that um your gas your quote is less efficient so it's a trade-off but yeah discord on flashbots or you know it could even be on the etherscan zoom chat right now you could find your next teammate who knows how you can find your your next wife who knows [Music] so is this is the flashbacks discord where the whole anime girl meme started i don't know the the backstory to that i actually don't know either i thought it was funny so i just [Laughter] joined um so um when you mention flashbacks discord right so like i i kind of always imagine it like how if anyone's played that werewolves of or mafia game but you just played online because like there's a lot of uh people who potentially can uh front run you in that channel right like if you were to accidentally say something that uh leaked some alpha so like what's the communication like between like the searchers and the other kind of entities within that discord like um do you notice if it's any different than how people communicate on twitter for example oh i just noticed okay so this is one thing that i think flashbacks didn't do well is that they had a white hat channel and then every time i go this is people saying hey i lost my tokens can you help me so that is like probably like the most prominent thing on on the flashbacks this quote other than that is there are people discussing in office but they're not really dripping it's like ooh i made a big profit today like check out football on chain so like they don't really tell you the the secret sauce you have to actually go and find it yourself so that's like maybe 10 seconds of time one percent of the time is people actually leaking alpha and deleting and someone would be please do not leak alpha please delete it so one percent of time that's real alpha and people will be asking other people to delete it so what back back then what happened was someone someone actually made a discord board or something to basically keep track of all the messages so even if you delete it exactly if you delete it you'll be notified which lead i don't know there's a lot of like mind games to this and i would say the other forty percent is me harassing but or like someone from the flashback team saying hey why isn't your api working what's the api down um but yeah there's no real like structure but this was a while ago like i haven't really checked it in a while i want to add and and uh we've also seen like um on chain messages between uh different mev bot runners like um do you know if that's a a real thing or it might just be like a one or two like one-off um kind of scenario no no it's it's happened many times with me and my pals actually more like my powers but what basically if there's only two competitors to one opportunity there has been times where and telegram link has dropped and you would go into negotiations with the other party because what happens is with pgas and all these like bribing systems is that if there's only two of you then either one one like that basically the opportunity cost goes to zero right or like maybe like a tiny portion so why don't you just make an agreement with the other guy if there's only two of you all right so i'll keep fifty percent of the profits you keep fifty percent of the profits that's a lot more that is way more money than us competing each other to the ground so oftentimes like that actually does happen um i'm very happy that bud has not posted any of those links those links that involve me but it does happen it really doesn't happen like and it and we we don't even know who the guy is from it's just a kind of ukrainian guy that's all i know um then uh so you hand asked um a good question here like um if you were to be a researcher like what percentage of profit uh would should you um give towards uh bribing or tipping the minus um and assuming this is something in the long tail oh i mean long tail i would say on on just just rough gate just like roughly skimming what other people are doing they're giving like 50 you could definitely go 51 or you could even try 10 um but that would also mean that you're assuming that no one else is going for the same opportunity okay like so i guess like the point of bribing is okay how much are you willing to pay to obtain the opportunity if it's a hundred thousand dollar cost are you willing to pay fifty thousand dollars to get fifty thousand dollars do you it now it becomes like a psychological game okay or do you think other people it's almost like poker do you think other people are going to pay 50 000 for the 50 000 opportunity whatever like the probability that someone is going to pay 51 000 because if they if they're willing to pay fifty one thousand dollars for that one opportunity you lose out on fifty thousand dollars but if you were willing to pay for one thousand dollars and five hundred dollars you and you get the opportunity of course this is all depending on like code efficiency uh and and like uh you know gas prices and a bunch of like other other factors but one i guess it all boils down to one how much are you willing to pay and how much for the opportunity how much are you willing to quote unquote like uh lose uh how much are you how much are you yeah willing to pay for the opportunity and how much do you think other people are willing to pay and you're trying to find that nice nice intersection in those two uh two to function the two probabilistic functions that you have in your head um yeah it there's no real right answer it's all like it's like poker right it's like what you think your opponent thinks so then it goes a bit meta yeah very high stakes too um so search asked uh basically the opposite question then like uh for the short field um how do we decrease the minor bribes uh which is at 98 to 99 right now oh it's very easy all you have to do is ask everyone who is also doing short tail to say all right guys we're not going to drive any more than 95 that actually happens by the way not a discord um they're like guys we're driving too much can everyone decrease it decrease your driving you know sometimes people do sometimes people don't because if everyone agrees i think this was a lot more easier when there was only a couple searches but now that institutions have moved into space for short tail it might be a bit harder um but you know institutions they don't really care if they earn like one cent from like a hundred thousand dollar trade you know it's still one cent profit so it doesn't matter to them but as a soul very important for such a independent searcher as i would like to say like back then people would actually kind of collude and say yo 0x abcd123 like you're driving too high please message me and we can come to an agreement on what how much percentage you should write i think one time like back then it was possible for short tail but now it's probably very very very hard to kind of coordinate and collude to get to like a priming percentage it might be easier with long tail right now i think it's still possible with long tail as long as there's only a few people but for short till i don't think you can you can you can collude anymore so it's very interesting part about centralization of power i think um i want to ask a related question uh when you talk about how people do it post uh ethereum emerge uh but um before that i think this one topic that maybe some people who are not familiar with uh mev might uh be find very useful to to hear about from you is um this uh idea of sandwiches like can you kind of explain what a sandwich is um so you dropped out a little bit yeah like i just just heard the end sandwich and that was it yeah just like can you elaborate a bit about what a sandwich is oh uh so sandwiches like this very predatory attack where it's almost like rob so it's just like robin like if the thing is free you know you're the product and so how it works is that because before you trade you basically need to tell the world hey i want to buy like ten thousand dollars worth of eve of usdc and so you broadcast that out to the public that trade being out of the public is now information that can be utilized against you because uh it's not on an order book meaning you're just saying all right i'm willing to accept the flight okay actually let's use shiba because cuba has high slippage because by high slippages i mean that because it's so volatile you might execute the trade at say like a dollar but it actually might get executed like a dollar fifty so a lot of times when people trade shiba on these decentralized exchanges the slippages is like ramped up to the max it's like 50 slippages what that means is that i'm willing to take a 50 hit on this position and the 10 000 trade is 5000 uh gone but it might only translate to like maybe 500 or a thousand dollars in profit uh to the surgery but essentially what happens is that say let's say shiba is trading for like a dollar in open market and you want to buy steve and you're willing to buy up to dollar fifty but it's only trading at a dollar you broadcast the information what happens is that someone listens on the on for people broadcasting information let's see all right this guy is willing to take a hit up to 1.

50 let me buy shiba up to a dollar fifty sorry and then you can buy a dollar fifty uh cheaper a dollar fifty pushing the price up even higher and i can then sell it and then basically profit off um off of the the price differences so that's essentially a sandwich attack you have like a front slice the your victim's transaction and the back slice it's like a three-step process but it's extremely predatory because it's it's you're basically you're basically giving people a bad rate is the tmdl just like robin hood you know before you send your transaction you know it gives that information to a bunch of institutional traders who can bump up the price a little bit higher so you get a little bit better a little bit worse rate not noticeable but when you when this is done in volume it's quite profitable for for robin but that's a sandwich trade yes and like in a blockchain that doesn't have uh protocols such as uh flashbots like can you do this uh let's say using pga oh you could but it's just a lot more riskier right you could so if if if you could even yeah you could it's just very risky without and it's not probably not worth them really the answer is you can't but it's not worth the risk this is the point yeah uh so nicholas uh asked uh this is a fun question for those who might not be familiar like um so how common are um searchers um attacking each other using uh tokens like uh salmonella uh i'm very common extremely common uh i mean back then when extremely like i think that if there's an opportunity to extract you it will be extracted in fact you know if your competitor is if you're like okay if you are a bot operator and you find a more efficient bot and you realize that hey he's only more efficient because he's doing less security checks i would be mad you're stealing my opportunity because you're doing less security checks you know it's not my fault that i deploy a token and you want to buy it and suddenly all your stuff is gone right just putting it out there you initialize the transaction but anyway to answer nicholas question yes it's quite common and in fact i think these days on eve it's a lot um it's a lot less it used to be common sorry it's not common now because all the weak bots have been weeded out short-term short-term wise anyway but it does happen there's like evolution of it there's salmonella there's like i think it was like grub or something i'm not too sure i forgot the next iteration that was extremely like tailored as well to like a specific part like this guy reverse engineered this parts uh code and realized he was using tx not origin and because of using tx origin he could basically do a almost like a reentrancy attack at the bottom and take out all the money but yeah it's still common i'm sure if you have a large enough reserve and the incentive is today i'm sure it will happen as we've seen with a lot of these protocol hacks yeah this would be long-tail blackhead i would say not that not that i've done anything before nor have i got any cryptocurrencies of course all right then um so last question on sunday just like so um people have talked about uh multi-block uh sandwiches before um i don't know if that's actually been executed in the wild but could you explain what what that is and whether uh you've seen anything like that multi-block sandwiches i don't think i've seen i think i think multi-block sandwiches are just people like decoupling i i assume like this is probably not my this is not my domain expertise anymore maybe i've seen it with all these like random s transactions but it sounds incredibly um uh incredibly risky uh because what happens if the block gets unclear like sandwiches usually require a lot of capital for it to be profitable you're buying like six figures seven figures and coin and coins to immediately be selling the coins so but but i think i don't have the expertise to answer this question because um i personally haven't really done much on multi-block sandwiches maybe multi-block trades yeah where you use obscure tricks to reduce your gas efficiency for a very highly lucrative positions yes multi-block transactions multi-block opportunities yes but maybe not multi-block sandwiches yeah all right so yeah so i see a bunch of questions related to uh ethereum uh so maybe before that i'll just ask you again uh feel free to answer to your level of comfort uh since it's not even out yet but once uh the merge happens and ethereum is on proof of stake like um how would the mev game change uh change no that is something that i have not looked up i know so i probably am not qualified to talk about that uh yes uh there's some on my to-do list there's too much stuff to do okay then uh maybe uh so i'll ask you some of the merge related questions uh but uh so turanji just asked um uh for if there are faster blocks does that make it harder or easier for mev to happen well it definitely makes it harder because now you need to pass information and kind of uh you need to pass information find opportunities uh with a shorter timeframe actually that's why that's why i would actually recommend people who are not uh great at infrastructure that start with l ones with ethereum specifically because the block time is 14 seconds it's like an eternity in hsp right you can calculate a lot of you can compute a lot of opportunities to uh you can compute out a lot of opportunities in 14 seconds is what i'm trying to say but usually you need to get it done by seven seconds because that's when the actual crunch happens but that's like eternally you know like programming languages um so yeah it does make it harder for example like polygon or like phantom like these one second block time uh blockchain you gotta you is you gotta get things done in the milliseconds so you can't use like you can't use javascript for for a lot of your operations it is you basically need to be more performant but yeah so it does change things so is it true that uh some people have said about like to be competitive in the mvv game these days like you have to go even like deeper to like assembly code does that happen uh okay that depends on what uh you're doing right like what poison you're picking if you're going short tail where you're fighting for the institutional traders and we're fighting with all the big boys established players yes you might need assembly i i have written assembly you need to juice out everything in code efficiency for ethereum might not translate to l2s um because we are going through the flashboard routes flashpots make people weak by the way it's a lot of searches like the same but you know whatever i don't really care um but yeah no like um i forgot the original question but yeah you do write assembly for yeah you do right assembly yeah that was the question especially um um so when you said flashbots make people weak um can you elaborate a bit more okay because now the infrastructure side is handled by like a third party so you don't actually go through the pain of handing like pga auctions you don't actually want you don't write your infrastructure too you know basically a lot of the seat things are being extracted it's like i don't know if there's any python developers here but it's like you're in machine learning or like statistics r guys or whatever you guys are it's like using python and importing gravity you're like importing sklen and boom all the algorithms you need is right there so you don't actually need to dive into the nitty gritty details and you can actually do your thing so uh the hardcore searches would say that flashbots make people weak because you never had to you never you didn't need to go to pain yeah it sounds like when the older generation tells you uh tell us that you know technology makes us like more spoiled um when my mom tells me that she'd have to walk for a couple kilometers to go to the street yes walk through the mountains yes oh sorry maybe not assembly but you was also a good alternative i think from circles to mentioning yes yo is really good for performance it's very small and readable and it's quite lightweight for anyone like assembly is more so good for the libraries or like in interacting with long-tail products oh yeah i remember what i was gonna say short tail yes for code efficiency but long till probably not long tail would probably require less code efficiency but more on like finding that opportunity would be the hardest bit all right cool uh just a quick note we're almost at 9 30. um generally we'll end it uh 9 30 pm religion time i just want to do a quick check um lib if you don't mind extending a little bit more or if the people in the uh call want to extend because i think there's a bunch more questions i'd love to ask personally but yeah i just want to check your time oh i could probably go for another 10 minutes okay yes all right so let's see uh and you know if anyone has like some dying questions you want to ask like please ask it now because otherwise we might not have time uh to ask so um i see alex's question here so i'm just gonna uh ask for him uh what languages are most performant or recommended for mev i mean that's that depends on a lot of a lot of factors so i guess if you're gunning for performance or i mean just use whatever you're most comfortable with like people always think you have to be the most performant um you have to use the best language with the best tools to get the best results but just get it good and like what are you most comfortable javascript you know javascript can be faster than c in some cases uh simply because of all the optimizations done for idiots like me um you know if you want performance like rusk okay if you say russ woohoo russ but you know if it's good enough and it works in my opinion that's fine but if you're unauthors okay maybe not javascript you might need to use rust or in fact in fact fun fact like the most performant med searcher i know on these opportunities actually just uses python like in fact python with like c bindings that you don't need to go because it's easy on the math right there's many factors these things like bus is nice you know it's fast there's libraries for it but python says there's strengths and weaknesses to these things like you have to take into account the factors involved and not like like shoehorn rust or gold into everything like go is great because like the the um the get protocol is built onto it so there's a lot of libraries pre-existing libraries russell was a bit more painful back then but i think like paradigm they're writing the ethos rs library so it's a lot easier right now but back then i was just using typescript that was like in all honesty that worked really really well it doesn't really matter if you can make it you can make it performant enough in your use case yeah all right so um you mentioned a couple times other changes before uh do you think you'd be able to share any of like the current sort of um environment in some of the different chains out there like what the what the space is like oh for the other trains i've just been doing short term stuff but i remember there was like a game called cab kind of brow or something some crap came on avalanche i think someone else i think bert posted a thread on that cupboard covered yeah yeah but essentially it's like this game where you know you have like uh crabs or something some nfp thing and then you you you mind something or something i don't really and you get like 500 like 50 or something and this guy claimed to make 500k over six months with two people so it's like 250k profit each i think it's extremely extremely possible it's not really out of the ordinary intact um yeah there's like other changes there's ecosystems in other chains like um like binance and pancake swap and a lot of like obscure stuff like in fact the more obscure there's like a this is like a known secret i would say in searches is that the chains that you don't use actually provide you the most profit and that was true for me as well like ethereum is probably the least profit chain for me and ironically simply because it's like so much competitive competition but the minute i went to like eco chain or vfc chain phantom chain avalanche moon rifle moon beam there was no competition like fine like i'm i much rather be a big fish in a small pond than be a small fish in the ocean right so like being the big fish in that small pond i was able to dominate and grow really fast like coding on ethereum was great because it gave me the security features that i needed to keep in mind gave me the confidence the confidence is the key thing as well like i hear someone ask like what was your first mevtx it was actually a sandwich tx because we knew that and we only profit like on one cent like maybe a fraction of it but the minute that we did it i say we because it was a team effort it was not being when we did it we realized that hey we're just as good as the big players you know we can do this so we can also it's good to have a reality check by the way like if you can land a sandwich transaction the odds of you being able to land a long-term transaction or any other short-term transactions is also very high simply because that is the most competitive uh most competitive uh area in ethereum l1 flashport specifically so if you can land that that means you're just as good as everyone else all right uh yes um so yeah i'll just do a couple last questions um from me and i'll ask this couple from uh the chat as well um so you mentioned earlier um avalanche is harder and kravita is on avalanche so uh could elaborate a bit more to the audience oh um so avalanche is a bit harder because to access the mempool you need to have a validator i think uh one of i think chain analytics or something recently gave you access to the menu but it's like a few seconds slower so your opportu you they might not be the best but essentially you basically back then when avalanche was like i don't you have to stick i'm not too sure how much i think like a few a few hundred thousand dollars worth of avalanches this is when avalanche was like a hundred dollars so it might not translate right um but you had to basically stake a few hundred thousand dollars worth of avalanche in order to have access to the mempool so that really like we didn't want to we didn't want to put in that much uh then what we didn't want that much exposure into avalanche so we didn't do that and and that's kind of why it's harder for avalanche on some chains there's like commissioned member access so the which makes things a bit more harder and depending on your risk tolerance and how much you're willing how much how much is the opportunity cost that you may or may not want to participate in that chain yeah all right so uh before i forget so research asked um about any recommended uh tools for guest profiling and debugging uh oh yes i think you know my personal favorite as a haskell fret fan is that tools like it's incredibly lightweight now in fact that's not tools it's that tools that's it um i think the more popular opinion these days is forge forward ethereum yeah from the foundry team um yeah foundry like forge uh yeah it's really rust a lot more faster but you know back then i was a haskell maxie so yeah i used that tools very frequently uh for gas profiling remix is good but the debugger on like dab tools and i think forge these days is way better than remix like a way more informative than remix remix to me it just feels like your browser can crash at any moment and i can't really save my progress so i can't really save my progress but i can at least feel better because it's in the term it makes me feel like i'm a hacker if you will yeah uh we just had a remix uh for the last uh meetup so apologies rob if you're listening to this one oh sorry your remix is great you know contracts on remix that requires to be very fun i love remix thank you so much just not for the mev searching maybe not for debugging yeah it's very painful okay um so uh yeah couple of questions unless anyone has any super last minute questions in the chat um does it uh is it necessary for that to be like a good explorer for these other chains before you would consider doing mev on them oh yeah yes yes actually and ironically yes eat the skin thank you guys so much so much because every time i go to the channel like there's no eat there's no like ecoskin equipment like it takes me like 10 times more time to find these opportunities i remember the last ones like clayton's plating the korean chain oh my god that was like so painful i don't think i end up doing anything with that because i i kind of started going to conferences but but yes like either scan is incredibly you know like the i guess like the interface and all the nice little features uh and in fact like the transaction debugging oh it is extremely useful um so thank you guys so much it's very very like if a good explorer makes a difference between taking like a couple hours to find an opportunity or like a couple days oh well that's assuming that i don't have my own node but yeah it does make a lot of difference okay maybe i should uh talk to you separately from uh talk about any feedback um yeah okay then uh just the last question for me is um before we end like do you have any funny anecdotes from all the mev stuff that you've done or any of your friends have done uh that would be safe to share and it was i mean i guess the funniest one was when uh the ukrainian well the ukrainian guy was like hey you want to split the prophets 50 50 and not kill each other okay and so the guys we just went on agreement with this random guy from that eta scan message he was like and then we just put the profits 50 50. uh please don't please don't try and find it uh i hope you guys can't find it you know i'm sure it's out there it's gonna take a while but it's it's there somewhere all right cool um sorry i missed uh question that's a final question a real final question um where do we start um now pick your poison right what do you want to do long tail short tail long tail fine like an obscure protocol you need to have skin in the game so that's the thing you need to be willing to lose some money i know i know like it's kind of scary but it's an asymmetric upside like you can't is that you can't really participate in the ecosystem without putting in some cash like find some strategy that you're confident will work locally and you know test and prod that's the only way really like you have to test and prompt as much as i as much as i hate to say it you have to test it all right um so yeah thanks thanks a lot for sharing your time with us even going past the timeline uh yeah everyone else please feel free to thank him uh as well um a quick note um so uh nicholas uh has told me that there is some interest within the community uh to create like a mev channel uh so we have a discord server and if yeah if people want we can just create like an mev channel for the mev aspiring uh to find their squad um so yeah if uh you'd like to do that um maybe you can let nick know maybe first in telegram and then we can get the whole thing started um yeah and maybe live at that point uh if you want to do some uh custom workshops uh for the ekl people that that would be great too i'll definitely think about it but thank you harif and thank you caleb thank you guys so much for showing me around the etherscan offices i really really appreciate it uh yeah thank you also thank you everyone for listening to me rambo i know it's been i look at the time now it's been like an hour and ten minutes i was not expecting that all right yeah so thanks everyone and i will end the recording now