Oct. 13, 2021

Austin Griffith: building on Ethereum

This is a conversation with Austin Griffith about his continued journey as a builder/educator, and learnings from onboarding thousands of developers onto Ethereum.

This is a conversation with Austin Griffith about his continued journey as a builder/educator, and learnings from onboarding thousands of developers onto Ethereum.

Austin is a friend and someone I've had the good fortune of working with extensively in our shared time at the Ethereum foundation. In this conversation, we take a tour through the different projects he's worked on over the last two years — starting with ETH.build (the graphical interface for working with smart contracts), moving to scaffold-eth (his development starter kit that has really taken off and been used by thousands of developers), and then moving onto the BUIDL GUIDL and the Moonshot Collective (DAO-like collections of developers who work on prototyping interesting new ideas). We also go down a few fun tangents around Loot, the Doge NFT, DAO tooling, and we get to hear a walkthrough of Austin's curriculum for bringing a new developer into the space.

I consider Austin to be a gem of the ecosystem, and feel lucky that we have him here. I hope you enjoy the conversation.

Check out our newsletter for other episodes and writing: bytecode.substack.com
Twitter: twitter.com/sinahab


Sina [00:00:00] Hello, everyone. Welcome to another episode of Into the Bytecode. Today, I sat down with Austin Griffith. Austin's a friend and someone I've worked with pretty extensively in our shared time at the Ethereum Foundation. I think he's one of the best people around and has done as much as anyone for growing the ecosystem. In this conversation, we talk about Austin's journey over the last two years. We take a tour through the different projects he's worked on, starting with ETH.Build, the graphical interface for working with smart contracts. Moving to Scaffold.Eth, his development starter kit has taken off and been used by thousands and thousands of developers. And moving onto the BUIDL GUIDL, and then finally the Moon-shot collective. These last two are DAO-like collections of developers who work on prototyping interesting new ideas. And throughout this, we also go down a few fun tangents around Loot, PleasrDAO, DogeNFT, DAO tooling and we get to hear a walkthrough of Austin's curriculum for bringing a new developer into this space. I consider Austin to be a gem of the Ecosystem and I feel lucky that we have him here. And with that, I'll leave you to the conversation.

Sina [00:01:37]: It's been the last couple of years that there's been a whole bunch of crazy stuff with Scaffold-eth and BUIDL GUIDL and ETH.Build. What have you worked on, and what are the big kinds of turning points of the journey?

Austin [00:02:03]: Let's start with ETH.Build there probably, like ETH.build was this visceral, if you go to ETH.Build, there's a sandbox you can drop in a key pair. You can drop in a signature or you can sign a message, you can recover that message. Just a really visceral kind of poke at things and then like when Loot came out, I had a little quickly wired together with a little dashboard that would go talk to the Loot contract and tell me which ones weren't minted yet. And then let me set up a Metta mass transaction to mint. So it's like full web three, it's just clunky because it's so big, but it helps me talk through the concepts. Talk through key pairs, talk through hashes, and if you go to ETH.Build, there's a full curriculum there starting with hashes, key pairs, transactions, distributed ledgers., and byzantine general problems, which is like beyond me. I'm not like a co-sign researcher, I'm just building stuff on top of this and seeing what sticks, but then leading up to blockchain and smart contracts. So that was ETH.Build.

Sina [00:03:08]: Wait, hold on. So you said, when Loot came out, you use ETH.Build to interact with the smart contract, while the rest of us plebes were on ether scan, like reading and writing to the contract. Which I also thought was cool. I feel like that's almost its design pattern not having a UI just made it so much cooler. I feel like other builders have to think about the fact that they want their smart contracts to be the canonical thing and the front end that they're building is one amongst many. But you can just not build the front end and just have the smart contracts and let everyone else just kind of figure out how they're going to interact with it.

Austin [00:03:53]: It feels low effort at first, but then you realize it's kind of a statement, right? It's a statement on composability, it's a statement on openness. It's like just a composable piece and he's looking for everyone else to build the stuff around it. I did have that dashboard built, but I quickly shared it on my Twitter. So anybody who followed me could click the link and then they had the same dashboard ready to go. Sofor what it's worth, I didn't like hoarding it or anything.

Sina [00:04:17]: No, I mean you and hoarding don't go together. So that's ETH.Build, so it's this graphical interface, it's pretty freaking elaborate. So you built that...

Austin [00:04:33]: I built that and I used it to create a curriculum and I think that's key. I used it to explain how these concepts work, like the fundamentals of blockchain and Ethereum. Then we moved on to Scaffold-eth, so ETH.Build it's a tribute to you and Albert that I probably would have stuck with ETH.Build and I would have focused on that.

And it wasn't as high leverage as moving to scaffold-eth and seeing that retrospectively I'm super glad that I made that change because with Scaffold-eth, it's having a ton of impact and it's the real impact I can see. ETH.Build this curriculum people get it every once in a while, someone will say something nice like "Yeah, ETH.Buildits fun." But with Scaffold-eth, I'm seeing apps every day, brand new apps getting deployed that I know came from that starter kit that probably wouldn't be here if it weren't for that starter kit.

So Scaffold-eth is an adapted template, the theory is that it takes a lot of work to build your stack in web three. Like you, you bring in a hard hat, you bring in others, and you probably have some kind of front end, getting all those things to talk, not having issues like who knows. The chain ID isn't right and you're talking with a Metamask, there are so many reasons why your stack will go wrong. And so with Scaffold-eth we thought let's put it all together and have it working out of the box. So you get right down to it, and this is Albert. You get right down to that moment where you're tinkering with solidity and you're seeing it update and you add a U-end and it shows up in the front end, you add a function to increment that U-end, and the new button shows up and you can hit increment. So you can get in this tight dev loop with solidity. So Scaffold-eth helps you start tinkering and get the syntax down, and get to be dangerous.

But then what we found with that adapted template is that people are building all these starter kits with this. We had an optimism starter kit, we have an NFT starter kit. We have like a D five leverage starter kit. All of a sudden it wasn't just the template, it was all of these new templates that were kind of extensions of the original template.

Sina [00:06:49]: So the idea was that the base Scaffold-eth template, you're trying to build a generic application on Ethereum here is kind of the basic stuff that you need. And then the NFT template would be, you want to build an NFT project so it comes with, I don't like IPFS and all of that.

Austin [00:07:08]: Our defy comes with maybe the Ava SDK. Or the gnosis-safe starter kits come with the gnosis-safe SDK, but then everything else is right where you need it to be in terms of, if you've built one thing with Scaffold-eth, you know where everything else is. There are burner wallets too, so you don't use the burner wallets, usually in production, we did with nifty ink, but most of the time you don't want someone landing on a site and just getting a wallet. You want them to connect in and use their wallet, but it's so much easier to test and build an app when you just have burner wallets and you just click the button and it signs a transaction and sends it and you don't get a Metta mask dialog and you don't get Metta mask as telling you, you need to switch networks and you don't get Metta to mask telling you you're on the wrong chain ID. You just hit the button and do what you want it to do.

That's so much more powerful for a developer or a builder to be able to focus on first, how do I learn solidity? Okay, I've got mappings, I've got payable functions. I've got contract-to-contract interaction. I'm getting this all down. Now, like where do I go from here? And that speed run is probably the next piece to Scaffold-eth. So I started doing lots of mentorship sessions, I don't know why I did this, but I tweeted. So this is a lot of the tangents here, but we kind of have this theory that there's a grip of developers kind of in orbit around Ethereum and every once in a while those orbits get close and that developer happens to pop in and start tinkering and thinking about things and like, how do we capture them on that close orbit? And so with the speed run, I use these mentorship sessions to help me figure out what I needed to teach. What was interesting, and what were the magic moments?  You and I talked about Cron jobs the other day and the Cron job analogy to how you do a Cron job in Ethereum, really opens your whole mind up to like how Ethereum works and it's like this massive multiplayer game and all these adversarial jerks can coordinate.

Sina [00:09:09]: So how do I go, the Cron job example?

Austin [00:09:12]:  I think it's enlightening, should we go through it just to talk through it? So a lot of the people that I talked to during mentorship sessions are web two developers. So I leaned into, okay, you've got a web server. People start using it. It falls over, you set up a second server for redundancy. You start scaling up horizontally or vertically, you handle that load by having redundancy. Well, the redundancy on Ethereum is like every single node has the entire chain and is executing everything. And when you deploy a smart contract, it's not like going to one of those nodes. When you deploy that smart contract, it goes to all of those nodes. And when you want to update a value, everyone on the network has to update their value. So already there's like this crazy amount of redundancy, but thinking through more things that are natural for a web two dev to think about, the Cron job is one of these tasks where it's like, okay, so I've got this smart contract that does compound interest every night. I need to poke this thing. How do I do it? First, you're like googling the Ethereum Cron job, you're wondering is this part of it? And then your second intuition is, well, I'll create a script and that script will check in at night and it will update the value. But super centralized and that script could fall over and it's not decentralized anymore and there's not that level of redundancy anymore when there's this single machine.

So the answer to this and like there are chain link keepers now but the answer to this in terms of just generic building on Ethereum is, you have to incentivize someone to poke the contract. So you set up a rule that says, this thing can be poked every night, right at midnight, or like once after midnight every night, you write that rule. The second rule is whoever pokes it gets paid 0.01 ETH, and you incentivize them to check into the contract. And all of a sudden you've created a Cron job, but on Ethereum and it's a different paradigm.

Sina [00:11:08]: That example clicks for me, it's such a good way of explaining what's happening here.

Austin [00:11:15]: And it's like when I say that Ethereum is the greatest massive and I'm not the one that came up with this Someone else, maybe came, I don't know who said this first on Twitter, but when they say Ethereum is the greatest, massively multiplayer game ever created, and you need to think of it like a massively multiplayer game. You need to think of these things like vending machines that help illustrate it a lot.

Sina [00:11:34]: Like when we talk about the Meta-verse, it's not a game that we're talking about. Like Ethereum itself and even more than Ethereum, the platform it's Ethereum, the community around it, and all of the interactions that these people are having with each other is the massively multiplayer game is the metaverse.

Austin [00:11:53]: I think I saw someone tweet about how part of the metaverse is arguing about what the metaverse is. That's like us arguing about it as part of what makes the metaverse what it is. I don't know. But thinking of these composable layers, the metaverse is like, okay, so I've got my loot over here, and then permission less someone creates this new thing that uses loot to do something else. And now all of a sudden we're playing this game from these illiquid SVGs and that to me is the Met-verse, like at first, I was like, oh its crypto voxels. It's me with my VR on, but it's not exactly that, it's more.

Sina [00:12:29]: It's the whole thing, it includes all of the sub games that are within it. Another thing that I was thinking recently is that it's another riff on this point which is that basically everything that's being built on Ethereum is one thing right? We talk about it as Defy and NFTs and Dows, and these are helpful conceptual constructs, but they're not separate things, right? It's like how in a university, you have a math department and a physics department and the chemistry department and a biology department, but physics, chemistry, and biology are just the same things at different layers of zooming in

Austin [00:13:10]: That's interesting.

Sina [00:13:11]: And it's the same thing here, right? Where you have an NFT, there's the most recent NFT that's kind of getting a lot of attention is the Doge NFT. I don't know if you've seen this one.

Austin [00:13:24]: I haven't seen it.

Sina [00:13:24]:  So Pleaser DAO bought the one real Doge NFT and has fractionalized it. And so you can go and buy a piece of this NFT on fractional but now all of a sudden, the NFT is non-fungible, but its fractionalized pieces are fungible. So you can put them into a uni-swap pool and trade them with each other. Or, you can build like perpetual off of them. And so all of a sudden the NFT and DeFi  got completely merged and then you want to govern Pleaser DAO as a DAO that holds the NFT. So they're not separate pieces is what's cool about it.

Austin [00:14:10]: So you hit on something there about the one real Doge NFT and this is fun, so as thanks to NFTs,

Sina [00:14:18]: That's a very strange sequence of words to say.

Austin [00:14:21]: There was only one true Doge, all Doge entered and one Doge left. No thinking about explaining blockchain to people. When I was talking about programmable money and I was talking about bonding curves, generally people would black out. But when talking about a thin little ownership layer, that's like a phone book of who owns what, like a really simple book that everyone keeps track of then it's a lot easier for people to conceptualize. So for me, NFTs are the thing that helped me explain it at dinner parties or, like, gallivanting around going to the dinner parties.

Sina [00:14:55]: So I tried right after Loot was happening, I was at a wedding and I tried explaining Loot to someone who didn't know anything about Crypto. Like total Newby, like basically didn't know what Ethereum was. And I found it to be very difficult, I failed. And so I'm curious, like how would you explain Loot without relying on previous knowledge?

Austin [00:15:22]: Yeah, we got backup to NFTs through. Let's get to Loot in a second and SVGs and NFTs, but I feel like just explaining the NFT and the ownership that's key.

Let's talk about the one real Doge. I can take a JPEG and I can deploy a smart contract and say, this is the JPEG and this is the person that owns it and anyone can do that. That's a permission less process, anyone, with ease, but the one real Doge is selected by the market and that's what's interesting. It's not me or anyone. It's the signal of the market saying this is the one real Doge and I will put money on that. And it's probably like where that money goes, probably to the original creators or something along those lines. But what makes it the one real Doge is not some hard and fast blockchain dry thing.

I mean, the storage is all there and you can see who owns what, but it's more about the market selecting that NFT as the one.

Sina [00:16:18]: Even the process of an NFT being the real one is decentralized.

Austin [00:16:23]: Yup, exactly and you can copy it and you can put it on another chain and it's not the same thing and collectors that are interested in buying those NFTs know that and see that.

Sina [00:16:31]:  I think this is what's clicking for me at another level these days is that I recently read sapiens and I feel like everyone who's in this space should read it and I'd skimmed it before, but it has this idea of imagined realities and interest subjective.

Austin [00:16:54]: It's like religion and stuff. And it's like a collective.

Sina [00:16:57]: A lot of things, so the idea of say countries that these borders are real. Like they don't exist other than inside of people's minds, but this belief, because it's shared by many people, becomes real. And then it starts influencing what happens at the physical layers. So, because people believe in countries, even though they're not technically real, you have checkpoints at these airports and immigration laws and like all of that. So there's these imagined realities and they kind of gain their power in the world because enough people believe in them. So when one of these new imagined realities is being bootstrapped off the ground, what you need to do is get enough people to believe in it. And then once it gets to this, it gradually grows that basically some sort of a network effect. And once it's in place any single person could stop believing in it, but it wouldn't affect the whole. But if enough people stopped believing in that, then it would stop being real.

I feel like being in Crypto gives you a better feel for this because that's happening over and over and over again as it happened with Bitcoin where it's these bits that were moving around and all of a sudden it becomes money, we literally like to create money. And then it happens with Ethereum and now I feel like it's happening all the time at the micro level. Every token, every NFT is a shared belief that has been bootstrapped and that's what's giving it its value. It's not the underlying utility.

I mean, there is potentially some utility, but the fact that people are coordinating around it as a selling point is what gives it its value.

Austin [00:18:51]: We could look at Loot for example, like basically, he minted these NFTs and he lifted them at like 7,000 or 8,000 and he had to just pay gas to get them. So maybe $50, maybe $100 and then there's the floor price, right? This is the belief of what people think its worth and that floor price starts at $300 and gets up to $100,000 and then goes back to $30,000 because that group of people, there's only 7,000 of these and people want them and that supply and demand happen there. The neat thing that happens with Loot specifically, is that we talked about Loot. We should probably get into it just because it's hip and cool, so basically an NFT that goes on the chain is usually just like the IPFS hash of some metadata. So what an NFT is, is usually just ownership of a Hash. This Hash is owned by this person and what that hash is, a manifest that's in IPFS and what's in that manifest is a name and description and also a link to some image and usually, that image is living in IPFS. So there's kind of like this storage layer, that's IPFS, and there's kind of this settlement layer or this ownership layer, this thin layer of who owns what, and that is what makes the NFT up.

What's cool is what we're seeing now. I had it on my Twitter, but I can't remember it right now. Someone told me actually, Simon DLR told me who did the first SVG somewhere on my Twitter. Someone OJ deserves credit for this, but you in the solidity and your smart contract you craft that JSON file, and you craft that SVG file. So you're drying the picture in solidity. Like this is an ellipse and it's this big and it's this stroke and this fill and then once you have the image, it's this JSON, and this is the image data, and this is the name and this is the description. And so all of that comes straight out of the smart contract and there's no IPFS layer there at all. You're querying the smart contract and it's drawing the thing and then putting it into a manifest and serving it to you. So to me, that's super cool and I did Looties where you have these ellipses and they're like these kinds of circles that get fatter or wider and there's a colour to them. And so what I'm going to do is set up like breeding of these things and the breeding will all be in solidity, right? You'll mix two Looties and get a new one and it'll be a mix of colour and a mix of width. And it's basically like, I don't know how the actual breeding will go, whether it'll be like an averaging of the two things or something like that, but you can do that in solidity with these. And so that's why they're super legit to me.

Sina [00:21:38]: So why is that cool? Is it that it's contained within this shared layer that it doesn't need to call out to anything else?

Austin [00:21:47]: So there are no dependencies there. It's hard to do, right? Like writing solidity to write JSON to write SVG is tricky. But then I think something we haven't talked about yet is just the composability, the fact that I can make those Looties, and then someone else can make some other breeding thing that breeds them. It doesn't have to be me that makes them anyone else can make any other layer and that stuff is just all on a chain and you own you’reLootie and you can do whatever you want with it. So back to Loot for a second, all he did was an SVG with white text, this is what's in your hand, and this is what's on your head. This is the robe you're wearing. These are the shoes you're wearing.

Just mix them up and give each player a card that shows what Loot you have, but he did it in SVG format. So it's like the smart contract itself is rendering these things. You can query the smart contract to see what kind of boots you have. There's no IPFS there. And then that composability and that permission less ecosystem that explodes around that is what's exciting is that all the other projects that are derivatives, we saw Loot unbundling, right? Where you could just get your gold ring out as one, and then you could put them all back together.

You talked about fractionalization right. You can fractionalize one of those NFTs, since it's a standard NFT, it can do all these other things. We can throw it into defy and it can get weird over there. So there are games and all sorts of layers that come out of having that, just a composable permission less layer. And then it becomes like a shelling point too, like when it's worth a hundred thousand dollars and there's only seven, 7,000 of them. Like people want those and people want to play with them and then just having those things have value, causes all these other things to spring up around it and that's exciting.

Sina [00:23:36]: Totally, having value is key to making them a schelling point. So maybe let's roll back a little bit. So we were talking about Scaffold-eth and starter kits and the NFT starter kit. And I feel like the last thing you were talking about was people who are in these orbits and catching that. Maybe just to talk about this orbital idea because I think it's pretty cool. It's something we've talked about together in the past. So the idea is that when you think about a developer for the first time, coming into the Ethereum ecosystem if you apply a traditional kind of developer growth-like perspective on that. It's a funnel but it's not a funnel. It's maybe better to think about it as a series of concentric circles and the outermost circle is like, someone who's heard about Ethereum just literally has heard the word Ethereum, the circle inside of that would be someone who has spent an hour, like Googling around and reading a couple of things. The next circle is someone who spent two hours hacking on something.

Austin [00:24:49]: But that's it those hours, like an hour or two gets you one more circle deep.

Sina [00:24:54]: And the cool thing is that this ecosystem we're talking about, it's not a singular product. So that people are kind of going through this circle all across the world. It's happening globally. It's happening via all sorts of channels. There are different academies, different YouTube channels, and hackathons like conferences. So it's very kind of multi-layered and what you've been focused on, is like grabbing these people and bringing them deeper in. How do you kind of conceive of what you do in terms of these layers and what it is you're trying to do?

Austin [00:25:37]: So I think I talked about it at first. I talked about just making a call out on Twitter and being like, hey, if you're a web two dev that wants to get in, just get into my DMs and we're going to have a session and we're going to talk through it. It was very raw at first and I was trying to do my best to mansplain things but as we go along, there are things like the Cron-job talk and there are these other things that I've learned that I can easily deliver these concepts just in kind of a session of working through things. So I wanted to scale that, you don't have to get on a Zoom call with Austin to get into Ethereum right? And obviously, there are a lot of other people too, but you don't want anyone to be blocked on having to go get a 30-minute mansplain session to understand things. So I started kind of synthesizing the learnings from those mentorship sessions into this Eth Dev speed run. And if you Google Eth Dev speed run, you'll get this medium article and it's basically like, step one is tinkering with solidity, get scaffolds pulled down, be able to put in a mapping. Tinker with how mapping works, make a function payable and have the front end change and be able to send in money to that.

Get through enough syntax that you're ready for the first challenge. The first challenge is staking and I think the staking challenge helps illustrate Ethereum's superpower of being able to get people to cooperate. You're building a contract that says, from a bunch of people we need to collect X amount of Eth, and if by some period, we haven't collected that Eth, then you need to put that state machine into a withdrawn state and people can pull their money out. And that goes into why there is a withdrawal pattern and there's a lot of stuff there. If you get to a successful state, then you execute something else and it puts it into another contract, that composable piece is done and ready and it kind of hands it off to another contract. So that's the challenge one it teaches you payable functions and date that timestamp.

Sina [00:27:47]: Interestingly, I think this makes me think of one blog post that Vitalik has, that's called a Functional Escape Velocity for layer one block chains.

Austin [00:27:57]: That sounds like a metallic post.

Sina [00:27:59]: Yeah, it's really good but the core point that it's making is that you want your layer one to be as simple as possible to limit attack surfaces and ways that could go wrong. But expressive enough that anything else can be built on top of it and layer twos. State fullness, like the ability for layer one to contain a complex state, is one of these requirements for other things to be built on top of, and it's one of the things that Bitcoin doesn't have, for example. You can't do the thing that you described on Bitcoin and so I think that's just interesting.

Austin [00:28:41]: The simple EVM is interesting and it's something that on a mentorship session, there's this moment where you go to stake into the contract with the front end and you need to put in .01 ether or something like that. You put in .01 ether, and you have to hit this annoying button that takes it times 10 to the 18 and then you get this giant number and you're like, what is this? Why is this happening? And it gives me a chance to mansplain, but hopefully, we have more scalable methods, but the difference between ETH and WETH is why we need everything to be a whole number because we don't want the EVM to have to F around a floating point, math. So we want this thing to be simple, it shortens your orbit just a little bit. It gets you closer to having the mental model down.

So challenge two is a token vendor. So you deploy an ERC 20, you put it into a vendor contract and then you sell the ERC 20, but you also set it up so the vendor can buy it back. So you create a vendor that will buy and sell the tokens that teach you ERC twenties, it teaches you contract-to-contract, and then it teaches you that stinking approved pattern. The bad UX and understanding that if I want to send a token in, I can't just send it in and have something trigger. I have to go to the token contract, I have to approve the vendor contract. Then in a second transaction, I go to the vendor contract and have it go get the token, and then in the same atomic transaction, I do a bunch of things. And atomic is another thing to dive into in a mentorship session. How things either all work or all roll back, how flash loans work like that's all mental models.

Sina [00:30:24]: That's definitely a unique thing to Ethereum.

Austin [00:30:24]: And so challenge one and challenge two just get you like, "okay I've got a pretty good grip on how all of this stuff works." Then it starts to open up, I do an NFT challenge where you build an NFT and then you add a gallery to it then you have a price function. Then you add like a price curve, basically super simple you don't have to do weird math to do this. There's no like integrals, when you're buying an NFT it's like a discreet purchase. A single purchase of an NFT, so you can multiply the price by just a little bit. So you have a price as one variable, and then you have a multiplier as another. This is another insight into how Ethereum works, your multiplier is two numbers because you need a numerator and denominator, right. And that teaches you about fractions within solidity and teaches you how to build a price curve by adding one line of code. It's so fun to do, like have a group of 15 of us on a call and I have my NFT and I'm about to deploy it to the main net and everyone's got the jitters and I just like, "Alright, I'm going to add a price function, I'm going to have the money go to the BUIDL GUIDL and then I'm going to make the price go up by 3% each time one is purchased."

Three lines of code, compiled and deployed to the main net, that ability to be able to innovate and try and prototype so quickly is definitely a super power of Ethereum. It's cool to show off.

Sina [00:31:55]: That's one of the vibes I get from your videos. One of the things that I like the best is that you just like to ship stuff to the main net. I feel like I've personally seen you spend like a few thousand dollars on your YouTube videos on gas fees.

Austin [00:32:16]: I didn't use to do that, I would say post buying and selling a couple of Loots, I have a little bit more ETH to be able to pay gas. So we've got challenging one, we've got challenge two. We've got the NFTs, we've got the price curve for the NFTs. Then, the next level is a deck, learning how a decentralized exchange works. Learning why you want to decentralize exchange, learning how those reserves work, learning how liquidity tokens work. But zooming in on the price function, showing how the math works with the reserves and how it takes a little bit, leaves a little bit behind in fees. And if you rattle it back and forth trading, you slowly build up the value with the reserves. So just getting someone over that hump is like another side quest, you have to do the Dex side quests and you have to understand what's going on there and you have to understand why. And then we look at randomness, randomness is super hard on a deterministic public blockchain, there are ways to do it and they each have trade-offs. We could go with the previous block hash, and there are reasons why you don't want to do that because it's kind of predictable.

There is commit reveal, where it's a lot more random and a lot safer, but there's this bad UX of having to commit something and then remember it and then reveal it later and have that hashed with a future block hash or something like that. Then there's Oracles and VRF and then it's like, what's an Oracle?. And it's like, oh, well I can show you the world, let me get into this and you learn how you don't want just like an Oracle feed coming from a web server because it's the same thing as you running your cron-job in one server, right? Like it can't fall over, it has to be decentralized. It has to be as secure as your smart contracts or as close to that as you can get. So randomness on the chain, Oracles, and sign messages are another thing in the speed run. Just show them how you can do a signed message.

You can recover that signed message and that's cool you can do that off the chain. You can do a state channel with that, but having it settle onto a contract and being able to do an easy recovery and get the signer, opens up all this like weird delegated execution kind of stuff, where I can sign a message and someone else can submit it to the contract and I own the contract and it checks that it's my signature, even though it's some other dudes submitting it, paying the gas. And so signed messages are a fun one and I think it ends with you building a signature-based multi-sig. So you build a multi-sig with multiple signers, it's an M-of-N multi-sig and you don't store the transactions on the chain. You have one execute the function and it takes in the two and the value and the data, and then an array of signatures. And you have to make sure you have to loop through those signatures and make sure that you have enough valid ones and then you can execute it.

And there are some gotchas in there and I don't want to give it away to any developers that are listening, but this speed run is like level one of your journey on Ethereum. You learn challenge one, challenge two, NFTs, curves, the decks, sign messages, random numbers, and multi-sig. And then it's like, where do you want to go work? There's a handful of places where if you have that base knowledge, or do you want to become a smart contract auditor and go to level two and go talk with a lot bigger brains than Austin. I can get you through level one and then I can send you off to someone on level two or do you want to build a product with Scaffold-eth? We have all these starter kits. We're going to get you dangerously close to having a prototype, so you can see if you can get product market fit without the months of headache that it takes to build a decentralized app because we just have all these starter kits.

Sina [00:36:12]: I've had a lot of developers recently ask me like people who are getting into this space, where do I go to learn about building on Ethereum? I mean, this is a really solid curriculum to take people through.

Austin [00:36:30]: It's a good starter, it's you're licensed to start learning and building in the space. So this is a cool insight, if you want to build a decentralized and you're not a programmer, you think I need to go find a web three programmer. And the thing is all the web three programmers are all yellowing some garbage SVG, NFT to main that right now, and trying to make a bunch of money. They're not going to want to do your cool project no matter how cool it is. So the key insight here is that if you want to build a decentralized app, what you'd need to do is find a web two developer that's interested in Ethereum. Find one of those close orbits, give them the speed run, give them some of this content, and give them some of these challenges. They'll learn asynchronously, I hand this out and most people will go through it with a couple of questions to me, and then they're ready. You've given them the knowledge and they're building your thing at the same time. So you kind of like to bring them into Ethereum by giving them, they'll build your project first and then they'll pivot and go build their crappy SVG NFT project. But first, you'd get them for a little bit as you educate them.

Sina [00:37:39]: And how long does it take someone to go through this?

Austin [00:37:42]: It depends on the builders. So I did the Ether DAO is a good example where this is a kind of a small tangent. Ollie reached out to us and said, Hey, I want to do this ETHER DAO thing and I was like, oh, heck yeah, I'll be a mentor, and the way it's set up now is there's this big pool of developers that are looking to be mentored and then some Giga brain comes in and says, I'll mentor one or two of you. And really what they're doing is trying to pick the cream of the crop and mentor them and have them come into their company. If you're Joe from sushi, you're trying to pull someone to come work at sushi with you. If you're Ollie at synthetics, you're giving someone the knowledge, but you're trying to get them into synthetics. My approach was different, mine was I have this curriculum, I'm not looking for one or two. I'm going to send a message to every single person that sends a message to me. And so I said, I'm here to mentor, hit me up and then it was just like 40 or 50 developers. I'm a CS student at Berkeley and I'm working on this here's my stuff, I would love to be your mentee. Then I sent him a message back like awesome, you're in, here we go. Here's the speed run, get Scaffold-ETH installed, tinker with solidity, take on challenge one, and take on challenge two. The curriculum is there and the way I knew it worked, and I didn't know it was going to work until this moment, but the way I knew it worked is I sent that to all 40 people.

And instead of it being busy work, instead of it being too hard, instead of me having to hold their hands through it, I still continuously keep getting these messages from these folks that like, okay, I'm on two now. Okay, I'm on three now. Holy cow, I get it, I learned that thing. It's very asynchronous. So going all the way back around to answer your question of how long it takes depends on the builder. I think in a concentrated eight hours, probably even two hours if you're quick, we wanted it to be two hours but like there's some setup and there's some time that it takes. You can probably get through challenge one and challenge two in two hours, but to go through the whole speed run and understand it somewhere between eight hours and eight weeks, depending on how effective the developer is.

Sina [00:40:02]: So one thing we've talked about in the past is this idea of the tour of duty when a person first enters the space, they could learn how to ride Solidity. But it takes some time for them to grog what sorts of things to build in the first place, like the thing you're talking about with the Cron-job like that, but applied everywhere. What sorts of games to build, and what sorts of products and protocols make sense? And I think one interesting way that manifests, is you look at the people leading the projects of today and most of them have been in this space for a long time. There are the OGs who've been in these waters and they're building something new, what we were thinking through before thinking of this as a tour of duty. You have to go through your tutor tour of duty before you can build something really interesting and then the question is, how can we shorten this tour of duty? So someone doing the eight hours, do they get to the point because they do have a couple of aha moments through that, but what happens after that?

Austin [00:41:19]: So there's still a little bit of a zoom session that still happens where we go through and I just like poke at them a little bit to figure out do you understand how the gas price mechanics work and how this is a bid and how the block is limited and then minors are greedy? And we go into the Cron-job example, we go into how would you short a token? That's a tricky thing, how do you short a token. At first, you're like, oh there's probably some mechanism to do that. It's not like you borrow the token, you want to short, and you swap it for something else, then maybe you leverage that up. But like having to go through that thought process of how that works. You've tackled four challenges, let's just get on a call and a lot of me getting on the call with him at that point is figuring out where they're going. Are they going to create a new product? Are they going to become an auditor? Are they going to be a developer for sushi? Who is this person and what does their path look like? Asking them what they'd like to do, seeing what they'd like to work on.

So there's a little bit of just like actual placement going on there. But some of it's still like refining the method, if I get on and start talking to you and you've gone through the first three challenges and there's like a giant blind spot. And I asked you about it and it is obvious and it's like shoot this isn't working, I've got to fill that blind spot. One of them is gas, I'll find that they'll get through it and still not quite completely understand how the gas price, gas limit stuff works. That's like a fundamental thing which makes me think we've got to maybe change up how these challenges work and maybe have some kind of gas-based challenge where you discover how gas limit and gas price works. And then we can kind of refine that speed run and make it better.

Sina [00:43:09]: Maybe even including layer twos because that's a place where gas gets more interesting. You go from the main net to layer two, and that's also going to become more relevant to a developer's journey.

Austin [00:43:23]: What's interesting there with layer two is when you see a web two guy come into the space, he uses Ethereum like a database. He or she has a big old struck with everything they want to store and their big whole array of strikes where everything is going on the chain and they're using it as a database that does not work

on the main net. You got to think about gas. You have to think about costs. You have to be efficient, but on L two or some of these side chains, you can kind of treat it like a database, and it kind of works like that. So there's this interesting thing where maybe it will be easier to on-board people, going back to the NFT its way easier to explain it with NFT. Maybe it's going to be easier to on-board people if they can treat it like a database at first and have those big struts and have it on a side chain and then have the moment where, okay, we got to go to the main net, well, everything's going to cost $200 each time you interact with this thing. You may want to look at making it more of an event-based system or something like that.

Sina [00:44:24]: You'll learn quickly if you're spending 200 bucks of gas. So what's the BUIDL GUIDL?

Austin [00:44:31]: So we've got Scaffold-ETH, we've got the templates. All of a sudden we have these starter kits and challenges and tutorials. This kind of community is building around this thing and I'm realizing I need help first of all, but also I need to incentivize these folks to keep building cool things. And so the BUIDL GUIDL, and it's just recently I'm realizing is kind of like UBI for developers, where it's like if you're a developer that's gone through the speed run I know that you have the chops to eventually land at an auditing firm or build your product or go to a place in this space. I know you're destined for that, I'm just going to keep you around and keep your orbit tight by streaming your ETH. So I stream ETH to maybe 30 or 40 developers and artists, it’s a stream mechanic that I built myself where it's not sabiler. If you need a stream ustabiler and it's safe, super cool.

I needed to build a custom mechanism. I needed to set it up so I could stream to them, but I needed to set it up so they had to withdraw and price their work. And the other really important piece is they need to be able to disappear for a month and come back and pick up where they are. What we're finding is these orbits, like people will be around for a while, they'll be doing other jobs. They're working for $3 at once. You know, like web three is not going to be a lifestyle.

Sina [00:46:01]: You have one job that you go into the office full time and you clock in and out.

Austin [00:46:06]: Yes, exactly. So, the stream needs to work in a way where there's some cap and I think of it more like a battery charging up. The stream charges up to some 1 ETH or 0.05 ETH  and then they withdraw and they price it and they say here's the PR that I worked on or here's the tutorial I built or here's the new challenge I built. And then that stream slowly starts to charge up again. So it gives me the ability to have all these asynchronous workers, I can dump money to these streams. I make sure to just keep the streams full, make sure people that turn in and work, but it's not like I'm not tracking hours. It's just so low touch, but it just keeps folks around and keeps building and incentivizes them to create new challenges and create new tutorials, so that's the BUIDL GUIDL in a nutshell, is an incentivization layer around Scaffold-ETH that UBI for developers, kind of funds developers keep them kind of in this space and building things.

Sina [00:47:07]: It was honestly like one of the coolest things seeing you build this because we were talking about this idea of Austin's army, there are all of these developers who are kind of going through your materials and you need help building things. So the naive solution would have been, let's just hire or a more interesting solution would have been let's give a grant to three of these developers so they can spend more time on this.

And then you just came up with the BUIDL GUIDL which is like super crypto native.

Austin [00:47:40]: It's hard to give grants right? I would say grants are pretty slow second to maybe audits. Audits are probably the biggest pain point in this space but grant giving is a tough thing and it's hard to know who needs a grant, there's a lot of noise in this space. So even just being able to not have to F around with grants and be like I'm going to sell my uni-token drop. I'm going to sell my GTC token drop. I've dropped these valueless governance tokens, but I can exchange them for ETH and I can stream that ETH to developers. So it was just like, I just shortcutted the whole thing. Thank goodness for Uni and some of these other drops, right. They gave power back to some of these early users to help this ecosystem flourish more. So I'm able to sell that stuff that was dropped to me and stream it to developers, to keep building cool things and generic components on top of Ethereum.

Sina [00:48:34]: So what is the life of one of these developers look like?  Are they doing other things in this space, how do you work with them?

Austin [00:48:46]:  I think that they're working on different stuff, sometimes they're working for multiple DAOS. There's definitely a conversation I have with a developer as soon as I can tell this guy or girl is going to deal a lot of damage in the space, they're a damaged dealer, they can sit down and build cool stuff. Usually, I'll say what do you want to do? How do I keep you around? And it's often that they'll either say, I just want to get into space. I want to be in space full-time. And then what's interesting is once they get into the space full-time they don't want to do one thing full-time, they want to contribute to this DAO because it's interesting and it has something to do and they want to maybe help out with NFTs over here. It usually comes down to once they're safe and secure and like what they're doing, they rather kind of dabble in a few different areas and be able to phase in and out of things and build stuff that's interesting to them. It's not a traditional hiring process where I'm just bringing them in to do a task, it's much more open and trying to find stuff that they're interested in learning enabling them.

Sina [00:49:55]: One of the things I wanted to ask you about, and this makes me kind of move to that is that were the first quadratic freelancer. So what was that, what happened there?

Austin [00:50:09]: So it was early Gitcoin grants rounds, right?  Gitcoin was testing out how this whole quadratic math thing could work, it needs civil resistance, and how they can create civil resistance. But really, it was just a grant where the community could fund that grant and then there was a big matching pool from the EF saying, whoever kind of wins the love of the people gets the most money from the matching round. Kind of a popularity contest, but more like just getting a signal from the community. So early signal in the community was, "Hey, this Austin Guy seems to be building a lot of things and trying to raise all boats." I can say that a million times, I can even raise some boats here and there but it's hard to sell that sometimes to the EF or whoever. Like, "Hey, can I get a grant for this? I'm doing good things I promise." And then it's like, okay, "Tell us exactly what you're going to do and what are the milestones?" And that was harder for me, I'm very kind of all over the place and creative and doing weird things, something will happen and very unstrategically was not planning on that thing working as well as it did, but it did. So there's some play there with it, I had a hard time getting funding, but I had a huge community signal. That was kind of the multiplier that allowed me to get those matching funds and I think like 70 grand. I got my master's in electrical engineering, the first job I made $70,000 a year. And that moment as a quadratic freelancer kind of working nights and weekends, trying to get into the Ethereum space, trying to provide value all of a sudden, there's this moment where I get an annual pay check from the community to keep doing what I'm doing. And I think that was the first quadratic freelancer

Sina [00:52:09]: That was 70 K for a year?

Austin [00:52:11]: Yeah, I think so. Well, it was one or two Gitcoin grants rounds matched together. It was probably more like a 20 and then a 50 or something like that, I don't exactly remember. But it was enough money to fund a developer to go full-time if he was scrappy about it, he or she.

Sina [00:52:29]: So how did that feel like to be supported by the community in that way? Did you expect it?

Austin [00:52:36]: I get lots of people in the community saying, thank you for what you do, like you're so important, but when it comes down to me asking for a grant or trying to get that, the validation from the top is never really there. But there was this moment after the Gitcoin grants round where Vitalik wrote a blog post and my picture was in it and that was like goose bumps amazing. To get recognition and Vitalik even talks about how he would have never funded me, I'm just a knucklehead building stuff. But the fact that Gitcoin grants did this signal thing where the community said, "Yo, V this Austin Guy is doing some stuff you should check it out, and then that helped him take notice "Okay, maybe we should fund this knucklehead even though he's not like a Giga brain, he's doing good stuff and doing UX and doing on boarding." To me, the biggest validation was from Vitalik, but I had the community signal there for a while, but it was more of a pat on the back, I'm still super poor.

I'm still working nights and weekends here. I still have to have a full-time job, I know you guys think that I'm helping out and building things, but there's no putting your money where your mouth is until that moment. That's when I got that validation from the top, hearing from Vitalik that this dude may help me feel like my purpose was there and I was being effective.

Sina [00:54:09]: You started from a truly like grassroots sort of a way, it was like very bottom up. And then just through doing good work over and over and over again, you earned your way through this. I'm so happy that you're here now, all of these good things have happened and it makes me feel good about this community genuinely and your situation so it's awesome.

Austin [00:54:40]: I feel super lucky to be here. I love my job and I work a stupid amount of hours, but it's because I want to, I enjoy it and I love what I do. It used to be different, but I love what I do and I'm so glad to be here. I'm so lucky to be able to be effective in this role.

Sina [00:55:00]: I've seen you even when you're talking about the BUIDL GUIDL, just streaming ETH to developers, keeping them close to that orbit. Streaming them the ETH before they've done the work, you really kind of give people the benefit of the doubt and lean in. To what extent has your own experiences, like the fact that you spent years going through this effecting affected the way you go about it today?

Austin [00:55:28]: What I can even lead into there is that I was a real knucklehead when I got in, I got in and I thought I was going to be effective and I thought I was going to build something that was going to be awesome. It took time for me to like learning, so the strategy or the learning there is that you almost want to reward effort at first. When the blind man comes in and he starts feeling the elephant and he decides that he wants to push the wall over or climb the ladder, like whatever he thinks it is but it's not that thing you still need to like, reward him for working through that because he's about to learn that this is an elephant and there's a whole thing behind it. But at first, if we reward effort and then as they get through that tour of duty, we start rewarding efficacy. But at first, we were just, "Hey, you're going through the speed run, keep going." You don't know what you're talking about and you're not going to get rich. You think you do, but you're not. I just need to get you through this speed run and get you through this mental model, get you through the tour of duty, and then on the other side of that, you're going to be a veteran that your third, fourth, and fifth build are just steps along the way but your six build is a banger and you just release Loot. Loot is a very edged case example, but your sixth build is going to be something that the community likes, I guess that is the way I should put that.

It takes a while to build things and throw them away and if you reward effort at first, it's a lot easier to get a person to be like, not so married to that thing that they're ready to throw it away and try something new.

Sina [00:57:09]: In some ways, you're a teacher like amongst many things, you're a builder but you're also a teacher. I wonder if that transition from initially rewarding efforts and knowing that this person is doing the best they can seems like something that a great teacher would do. I'm wondering what you think about the question of what makes for being a great teacher versus being merely a good teacher?

Austin [00:57:39]: So actually both of my parents are public educators in a small school and I was raised to be a public educator type. My dad is just a great teacher and I think that there's empathy, and there's an understanding of how a person learns. There's presenting it in the right way, there's a narrative you take them through. There are a lot of like soft touch things that come with educating that I think I was just lucky enough to have grown up or have around me and now I love doing it. In this particular field, it's wonderful because it's like I'm teaching gold miners to mine, it's like I'm giving you the keys to do some really neat things. It's not like I'm teaching you though I don't know...

Sina [00:58:29]: Learning sequel

Austin [00:58:30]: Yeah, learning sequel. A sequel is cool, it's powerful but it's more than that. It's like teaching the mental model, teaching the Gotchas, and like freeing you to go build some crazy decentralized app or whatever. So it's a really rewarding process for me to teach that stuff.

Sina [00:58:48]: Did you know you were going to be a teacher or is this something that emerged over time?

Austin [00:58:54]: I always thought that, well I didn't want to be a public educator in a school with a hundred kids. There are not a lot of upsides there, I'm sure that my math teacher, Mr. Carpenter in eighth grade was teaching me physics and teaching me how to build a Linux machine, and teaching me how to write C-code. I'm sure that there are times when he and I both have this thought, “Man, I'm so glad I got to do that."

I'm so glad that he helped me and I'm sure he enjoyed teaching me that stuff, but that's not at scale. I want to be able to have that wonderful thing, but like with thousands of people and teaching them more, being able to scale it up. I thought maybe a college professor could be a cool role. So I had thought about teaching but it just didn't have the leverage I wanted it to until now.

Sina [00:59:50]: This is what you're doing is the coolest version of me, it's kind of crazy like having a BUIDL GUIDL where you're streaming to developers it's pretty wild.

Austin [01:00:04]:  It's so cool to go to these conferences or Devcon or whatever, and have like random people would be like, "Bro you taught me everything, thank you so much you got me through all this stuff." And it's so heart-warming to get that from random people, not me and Mr. Carpenter who spent hours and days together. This random person I've never met before was able to go through my curriculum and probably there are videos of me. They've probably had an encounter with a version of me in a video and they feel like maybe we're friends and I don't know this person yet, but then all of a sudden I get to know them and hang out and it's cool but it's a different scale of education for sure.

Sina [01:00:50]: You're simultaneously at the forefront of education too, alongside Ethereum and crypto.

Austin [01:00:57]: And maybe like funding decentralized workforces or figuring out what a decentralized workforce looks like and how to fund them and maybe what funding developers look like? Hopefully like some of these tools we're building with the moonshot kind of fit with that. We see things going towards DAOS, we see things going towards governance, what are all these tools that they're going to need and then being able to build and supply those.

Sina [01:01:27]: Well maybe let's move to moonshot now. So what is the moonshot collective?

Austin [01:01:32]: So Scaffold-ETH was the template, the starter kits, BUIDL GUIDL was kind of like an incentivization layer, it was kind of like UBI around that. And then Awalkie kind of came in and said, let's put a multiplier on the BUIDL GUIDL, do this similar thing but at a bigger scale. So that the moonshot collective is sort of the Venn diagram of where BUIDL GUIDL ends and moonshot collective begins is kind of blurry, it's a similar thing. Some people are in both, but the moonshot collective is a work stream of the Gitcoin DAO. It was funded with 40k GTC and we're funding and mentoring developers to build coordination tools and public goods funding and just thinking it through. Vitalik gets on stage and talks about how we should have signed in with Ethereum and then immediately on the Trello board of the moonshot collective is to build a sign-in with Ethereum starter kit.

So we're mentoring people, we're getting them to build those things. We're finding what the ecosystem needs and building out those tools. Another one is the gnosis-safe starter kit, you go talk to Dallas right now you realize that a lot of these DAOS have been constructed in the past with these intricate smart contracts, they're built without having a bunch of people use them. They're built with these complicated systems without knowing if it's going to have a product market fit if people are going to use it the way they think people are going to use it. So there's this theory that I have, that what DAOS need is just a multi-sig and they need to be able to have a dynamic front end that can settle to that multi-sig. So I use this example everywhere, but I talk about a stability DAO, say you wanted to build a stability DAO where X amount of people all stake one ETH, and then they all get to decide on where all that ETH goes by voting. Should we be stable? Should we just hold ETH or should we leverage up?

And you can imagine a front end of a sad face, a medium face, and a smiley face, that's all that the UI is. All those hundred people, don't need to have an intricate knowledge of a gnosis safe front end or how signing transactions or signing on a multi-sig work. They land on a website, they connect their wallet and then they say, "I'm feeling sad face today." And what happens in the background is you've created an experience to them that feels like you're voting in a DAO, but really what you're doing is if I click a sad face, I'm just signing the transaction that says we should move our money into die. If enough of those signers all click that sad face then the transaction gets the M of in that it needs and it executes and the money moves to die or leverage is up or whatever. And so the theory here is that we're going to need a bunch of tools and we're going to need to inject funds out of a DAO and we're going to need to have a process that still settles to a multi-sig so it's decentralized, but a process that we can design an experience around. We can design an experience another good example is a result oracle.

If a DAO wants to see a hundred new people join the DAO and so what they'll do is they'll create a results oracle, and they'll put four or five signers on it that are trusted members of the community. And they'll say "Did we get a hundred people in" and those be a yes button and a no button. The yes button signs the transaction to send it along to someone else and then the no button sends the money back to the DAO, it's still decentralized. It's super easy to use. You don't have to write any complicated, smart contracts. You just need a react developer and there are a million react developers in the world. You can quickly iterate on the mechanics of a DAO, and the experience of a DAO by just setting them up in the front end and having them settle to gnosis safe And that's just one example of like one thing we're working on in the moonshot.

Sina [01:05:47]: That's super cool, so that's just one of many projects. To make sure I'm following, gnosis safe is a multi-sig. And when you set one up, you say how many signatures out of the total possible signatures and you need to kind of trigger an action. You have to know who these people are ahead of time when you're handing it off to the react developer, but then basically the only API that the react developer needs to know about is that they can sign or they cannot sign. And they kind of iterate on the front end, building various types of experiences and you're kind of simplifying that the way that they interact with the gnosis safe, smart contract.

Austin [01:06:41]: Exactly, that's a great way to put and that's that gnosis safe build is just one build. There's also, we're going to release here at M-con I think tomorrow morning tip.party, which is just a fun little thing. We're on a call with a bunch of people from the DAO and we want to reward them with a little GTC, it's actually kind of a complicated thing. You've had to collect everybody's address, you've got to send it out. We built this super simple little app, you go to tip.party, you type in some words, so we'll be on a call with 20 people and I'll say the word is clown shoe. And everyone types in clown shoes and signs it and then I get this big list of everyone showing up, what's even better is it does ENS a reverse resolution. So I get some civil resistance there too. I see everyone's name show up and if you don't have your name, you're out. If you don't have ENS that's the first on boarding process I sponsor someone to get you ENS, so if you don't have ENS you're out. But it's an easy way for me to look through everyone and distribute GTC, I just say a website and I say a word and I get a list of everyone over here and I click one button and distribute GTC to those folks, so that's another example of a tool that we needed to coordinate to fund developers.

A third one is similar to Coordinape, we're doing this called quadratic diplomacy right now. So quadratic diplomacy it's a complicated name and is just some app, let's say tip party and we'd done this tip party, and hit the milestone of the main net, right? It took two weeks, and in two weeks it went from this is an idea to this thing is in production and we're using it and we need to pay out to whoever helped make that happen. So at that milestone, we all get on a call and we have a little ceremony where we say, who did what, stand up and say your piece. And then we give everyone 25 votes and then everyone on the call votes, how we should pay out that milestone. Say we allocate five ETH to the main net and then 10 ETH once five other DAOS are using this tool or something like that. So we have this milestone, we have this amount that we need to pay out, but we can use quadratic diplomacy to just pull the group and have the group vote, then we use quadratic math to kind of damp it down. We use the same formula that they used for Gitcoin grants where it's the square of the sum of the square roots. And so if you get lots of little votes, it's more powerful than one big vote, basically. So that helps us kind of pull the builders, find out how we should pay it out, and then in one button click pay out that 5k in, in GTC to the builders who built the thing.

So it's kind of like building tools to help us fund to keep building tools, to help us coordinate. We're trying a lot of prototyping and that two-week window is key, the fact that we can go from, this an idea on a zoom call too. I'm using this thing on the main net to distribute funds on a zoom call two weeks later and that's thanks to Scaffold ETH and some great builders, but that really quick iteration loop is helping us test a lot of different things and we're just getting started.

Sina [01:10:04]: That's super exciting. I mean, this sounds like one of the funniest possible things for a developer to work on. How can people get involved in this?

Austin [01:10:15]: So I say just do the speed run. If you Google ETH-DEV speed run, there are going to be some challenges and if you go through those challenges you're going to get an education and you're going to learn about building an Ethereum. But you're also going to prove that you know this stuff, like another build from the moonshot its codenamed scaffold directory. So as people complete these challenges, we're going to have them sign the output of any of these challenges is a URL. You deploy a full live working app, so they're going to sign that URL and turn it into the directory. So now you're going to have this list of 30, 40, 50 developers and how far they are through the speed run. If you're looking to hire a Dev, you can kind of look through that list and pick them out. If we're looking for who fell over, we can go find them and say, "Hey, you didn't complete this challenge, how can we help you out?" So there's this nice directory that we're trying to build that will help keep track of where developers are on their journey. Make sure they're progressing and then when someone says, "Hey, I'm looking for a web three" we are like, "Yo, go look at this website, here's 60 of them or whatever.

Sina [01:11:29]: That's super cool. It makes me think of that education point again, where the curriculum like how far the person is, their credential is baked right in. They can just gradually fade into actually working and making money with it. It's cool.

Austin [01:11:46]: The gradual fading is really interesting because you do challenge one, you do challenge two. You work on a Dex then pretty soon we're on a call and we start talking about what you're thinking about building, and I have a big to-do list over here, and I'm constantly trying to figure out how I can give you one of these to do’s. So building a results oracle with the gnosis safe starter kit, top of my list to get done. I'm talking to a developer. He's just finished challenge three or challenge four. He's looking for his next thing and I say, "You want to build this results oracle." So it goes from there's this set curriculum it opens up to now you're checking things off my list for me and we're getting more and more work done.

Sina [01:12:27]: So basically, someone can go and do the ETH-DEV speed run and within a few days, be at a point where they're getting paid and they can even do this part-time. It's not a whole full-time thing they need to commit to, they're getting paid to basically prototype and ship stuff to main net Ethereum while being a part of this community and getting mentored by people like you.

Austin [01:12:55]: Yup, exactly and it's a lot easier to mentor. There's so much noise and so many people in the space, it's hard to know like me going on Twitter and saying, "Hey, get into my DMs, let's have some mentorship sessions." That worked once, but my calendar was full forever. Being able to have these challenges set up so you can kind of at least weed out folks that don't just don't want to sit down and do it, but they'll say, "Hey, I want to deploy an NFT and get rich, will you get on a call with me and validate my idea?" No, I don't want to do that but if you go through the simple NFT example and you go through the buyer mints and you build a price curve and you go through learning how it Dex, and then you say, let's get on a call and validate my idea. Put it on my calendar bro, get on the phone. Yeah.

Sina [01:13:41]: What is maybe gradually bringing this to a close? What are you most excited about right now?

Austin [01:13:51]: People like to put things in little categories, we had defy summer and then we have NFTs exploding and then there are governance tokens, now all of a sudden the gnosis safe and all these decentralized things have product market fit and that everyone says, what's next, what's next? I think we're going to get to a few things that are going to collide and it's L two and direct ramps. I think when you can ramp directly into an L two and have that smooth seamless web two, like experience, but still have it back to Ethereum. I think we're going to see a huge explosion of games and not top down games, but bottom-up games, where one guy deploys the loot system and another guy deploys the map generation system and some other developer creates the XP system and maybe you don't want some of those and you want some of the other ones. So I think we'll see a huge explosion of games and I think even zooming in on that, I'm excited and this is just me being nerdy. Who knows if this is actually a thing, but crafting games I can't wait to grab three or four NFTs and put them together and turn them into another NFT that then has this other utility for this other thing and just playing around with fun game mechanics and having it be quick and fast and cheap like that L two experience that we're hoping we get.

Sina [01:15:18]: How is that different than building things today? Like building the SVG NFT, is that when you were building games you're forgetting about this thing being useful, and just purely focusing on it being fun.

Austin [01:15:35]: I think it could be a useful game too. We could create a game that has to do with carbon credits. And if you don't take your car today and you check in and you prove that you rode your bike, then you get some badge or some token or something like that. The games don't have to necessarily be completely pointless. Going back to the Dao, the fun thing about games is it gives us a chance to play around in an adversarial environment and try things out to see if they'll maybe work in the real world. Like start applying some of these real-world economies in these games on chains that have real value and see how people interact and see how people work and maybe we're prototyping a new government or a new governance system within this game that eventually gets applied to other fields.

Sina [01:16:29]:  That's something we need to dive into another time.

Austin [01:16:31]: Yeah, for sure.

Sina [01:16:33]: All right, man, thanks for taking the time. It's really fun.

Austin [01:16:37]: Thanks for having me, man. Thank you for your guidance through a lot of the stuff too. Like we worked closely together for like a year or two there and I appreciate it.

Sina [01:16:47]: And Likewise, I mean I'm truly inspired by you, I think you have a heart of gold. And honoured to be hanging out with you here.

Austin [01:17:00]: Thanks man, let's go get a beer.