Boxer  0:00  

Welcome to a new episode of the Weekly Wizard. I think this is the 10th instance now. So, there’s a little diversity here. It's great. And today, I have with me Andrew from Mir. 

And if you're in the doxing community, you should be familiar with Andrew. He's pretty much our most prominent and most helpful and just best human life community member. So yeah, very happy to have you on. And yeah, welcome.

Andrew  0:32  

Thank you. Thanks for having me here.

Boxer  0:35  

Yeah. Can you do a brief introduction of yourself, please?

Andrew  0:41  

Yeah, sure. My name is Andrew Hong. I grew up in California and came to New York to run away from home basically. I've been in New York ever since. I started to get the long spiel now or should I get…

Boxer  0:57  

No, no, I'll just think…okay. Yeah.

Andrew  1:02  

Like four years since college, but yeah, data science. I'm here. Basically, it was the first day to hire the team last August. 

Working on everything from basic analytics to engineering to like some science stuff. But whoa, I'm sure we'll get to that later. 

Boxer  1:27  

But a lot of people aren't really starting on data science, but you're doing other stuff as well. No.

Andrew  1:30  

Yeah, okay. Okay. All right. My brain is like scrambled. I'm like sewage. So, I used to basically do just a bunch of research projects on the side until like any anything in the Web3 data space that looks fun to me. 

So, everything from like mental stuff and art blocks auctions. The way I got involved with Mir was doing like social graph based ERC 20 airdrops. Right now, I'm working more on like Neo for J style network graph analysis. 

But we can dig into that later because that's like a whole new rabbit hole that I've basically disappeared into for the last two months. So yeah, and then I went through education stuff. 

That's obviously almost always going on…try to put our content…should have something fun to announce on that in the next week or so. 

Boxer  2:24  

Interesting. Yeah, when we work together briefly on the our network course, I talked to my team internally, and I was like, Andrew is the most productive person in life. 

So, it's certainly…you have your hands in a lot of…or you're participating all throughout the ecosystem. It's always amazing to see what you come up with. So, Andrew, I'm sure you've looked at your portfolio or in the market today. 

And it's not the best. And yeah, we're both data analysts. So, we're not market analysts. That's a stark difference. But what do you think we're left with now? Once everything has gone down? 

Tara has imploded. Apparently, they're scammers in the NFT space, which is like, oh my god, big surprise. Might have guessed. Who would have guessed? So after, I guess, two years of a good life in crypto and feeling very rich now we're back to baseline reality. 

So, what's your like? What are we left with? What have we actually built?

Andrew  3:40  

Okay, what have we built? I mean, what are we left with portfolio wise? Just pain. I haven't even checked my portfolio in the last few days. Building I mean, I think, God, I want to say we've built and learned a lot just because looking at where we're losing money and value. 

It is getting more and more sophisticated. We aren't just making the same exact dumb mistakes anymore. Like that. 

The most recent one well, okay, for the most part we are sometimes…you still do this and see the same reentrancy hacks, but for the most part, we've gotten out of the basic hacks.

But yeah, even with the whole Terra de Pegging that happened, that was some very sophisticated financial markets stuff. 

You know, just rereading into it, it wasn't like…there's by no means simple. There are institutional players involved. I think even though a lot of money was lost, and it basically crashed Bitcoin into…what what is it? 2027? 28? 

Whatever it is now. That's painful financially. It's like okay, it's getting more and more sophisticated. So, I think that's still a pretty strong signal. 

And obviously things we've built in the last two years is insane. No one knew what NFTs were, a year and a half ago. You know, so I think we filled a lot. I'm still bullish. And as long as ETH is above 1000, I'm not going to cry myself to sleep.

Boxer  5:26  

Yeah. Yeah, I think about how many people we've picked up, and especially how many builders and how many smart people, the space has picked up in the last two years. 

And obviously, that is fueled by financial speculation because smart people look for easy ways to make money. They find crypto, and they're like, “Oh, this is actually like a really cool playing field.” 

But then they realize, like, hey, there's actually more to this decentralized casino that we're building. Rather, it does feel like that uncertainties…but rather the underlying infrastructure here. 

That is really what is exciting. And what's interesting here, and I guess a lot of people share that story. So yeah, I'm definitely a lot of failed experiments in the last few years. But also, we've defined and found a lot of valid use cases. I guess, I would you agree with that take?

Andrew  6:31  

Yeah, yeah, I think the use case is like, they're slowly maturing. I feel like the use cases are only really usable by people like us who understand crypto. It’s not really usable for people who are my next door neighbor, probably even just writing on Mir would be hard for them. 

You know? And if even something as simple as writing can be difficult. We don't even need to get into DeFi.

Boxer  7:00  

Yeah, there were a few startups who had encouraged the back end, like easy API. 

Andrew  7:09  

I don't know how they're doing what they're doing. Do you find Mala thesis I think just went out the window for a little bit. Yeah.

Boxer  7:21  

They shouldn't have banked on Anchor, I guess. Yeah. Okay. So mark it down. We will probably be in Goblin mode for quite some time. So why the fuck should anyone still work in Web3, or like you're running this? 

Career boards or job boards or however you call it? Pellets? I think it's in the description below the below the video. 

Andrew  7:57  

So, why should anyone still try to get into a pretty easy answer. It's still early, and I feel like the best place…not even getting into the tech wise, but  the best place to build a career is where there's a ton of unchartered territory, a ton of money, a ton of potential impact. 

And smart people that aren't trying to keep things like you don't have 10 certificates you need to take or some really weird long job application process to get into the space, you know? 

You write one good Twitter thread on data analysis, and I guarantee you'll probably start getting interviews and maybe even offers. That's the barrier to entry. 

It sounds like a low barrier, but it's actually hard for people to learn how to contribute and communicate that way. Because people are so used to like,” Oh, let me go take this Udemy course” and then find a job board and apply, you know. 

But it's like, no, we don't do that here. The opportunity is just huge. And there's so much freedom. 

And that alone is already something I don't think he can really find in any other industry. And then he add all the tech is just like, okay, it's almost a no brainer. For me at least, it's a no brainer. Everything that I build that near I can then use anywhere else, you know, like day one. 

And anyone I want to work with, if I admire a data analyst on another team, like Michael from the optimism team, I can work with him on anything whenever I want. It's not like if I asked him for help. I'm the only one who benefits you know or vice versa. That's huge.

Boxer  9:44  

Yeah, I told him you read obviously. The choir here playing the devil's advocate. So, being on that topic, how did you actually start getting into crypto?

Andrew  10:00  

Yeah. COVID hit in 2020, and I was at my banking job. I did credit analysis back back in the old days. Obviously, none of my tech worked because they were trying to figure out remote login and whatnot. 

So, I had a bunch of free time. And I was like, alright, I can either sleep and watch Netflix or try to learn a new skill. And of course, I slept and watch Netflix for probably a month before I was like, okay, it's April/May this does not look like it's gonna just end. I should probably do something more productive. 

That was when we thought like, “Oh, by June, we're going to be out,” you know. But yeah, I had so much free time. I started looking into crypto because I saw someone talk about a unit swap. And I was like, okay, this is something that would affect my job and finance a lot. 

If it becomes big, I should probably at least have a basic understanding of it. And this was when DeFi pulse was still under like 2 million or something. And like AUM and whatnot.

Boxer  11:13  

If I pull it was still a thing, like at all.

Andrew  11:17  

Like DeFi pulls, screenshots every week, look another 300k. But yeah, I was like just dabbling on the side reading a bunch of these white papers. Honestly, none of it technically made sense to me. 

It took a while before the technical stuff kind of clicked. And thanks to Linda, she’s on Twitter. Great person. She put a Tweet where she was like, “Hey, anyone who's like trying to become like an ETH dev in some sort of way, I'm gonna hold this like Zoom session.” It was on some XYZ date. I forgot the date. 

And just like that, I was filling the Google form out. I signed up. So, this was like July or something of 2020. So, I filled out that form. 

And it was Linda Trent Pinups, Austin Griffith, and then like two or three others in the space, and they just talked to like 30 of us and walked us through the full like development stack back then. It was like react, andwe'd layer before it became hard hat. 

And like slitted, solidity for something with the 5 or 0.4. And talking to them helped me a lot and understanding the technical side of the space because otherwise, I would never have found those tools because Google Search sucks for trying to search or develop anything for crypto. You will not find what you need. 

Like, I don't know why it's that bad. It’s likehaving that workshop taught me like, oh, these are the tools you need to know. And then Trent was like, “Hey, you should do ETH global hackathon.”

And it was like a 30 day hackathon. When I was like, alright, nothing else really to do. So for those 30 days after work, I probably spent like fiveor six hours every day just like coding and learning solidity and Java scripts. 

And after that month, I was like, okay, at all. It all makes sense at a basic level. But yeah, that's basically how I went from zero to 100. Yeah. Big shout out to the Bankless guys as well. 

That was what I was reading to learn about Uniswap and whatnot. So, on like a media side of things. That's interesting.

Boxer  13:44  

So, 2020 was the first time you actually heard of crypto or you did any? I mean, not heard, but like, actually…not like Bitcoin mining in your dorm rooms or anything like that.

Andrew  13:59  

No, I was vanilla as hell. In college, old idea, just played ping pong. Did some like basic quant trading? And like, okay,

Boxer  14:13  

Yeah. What's your background? Like university wise?

Andrew  14:18  

Oh, I did econ and math at NYU.

Boxer  14:21  

So, you're not even a developer person or anything like that. You just now started developing in Web3. And that's like your first…

Andrew  14:31  

So, I started developing when I was at the…in my banking rule because everything in banking is PDFs and excel sheets. And I hated working in Excel and PDFs. 

So, I learned Python because Python you can have image CSV, like read the PDF put into tables and automate your Excel and PowerPoint. I literally wrote scripts to take the PDF of like the 10k. 

Put it into Excel with the spread and then screenshot the spreads into PowerPoint. And that used to take people like a day to do. And I just automated it. So, I could one click do in 15 seconds, and I'd have the rest of my day to study. 

And I can say this now.It would take me all day even with automating the emails. The email goes on at like 4pm. But that's how I learned to…

Boxer  15:30  

Been there done that. Yeah, if you are slave to the programs you are you supposed to be using that to make the program's work? I think slave is a hard word. Servant. I think that's better. 

I don't know. I apologize for foul language if anyone takes offense. So yeah, um, you just dove right in and then now you work in Mirror. So, what was the progression from being like, I know some of this stuff to actually getting a full-time position somewhere.

Andrew  16:06  

Yeah. So, I think everyone's journey that they went through is somewhat messy. Because one, the technical stuff is hard, and the startups make almost no sense. The fact that there's no product market fit in anything, you know, even back then it's not very known was great. 

So, it's literally…I knew I wanted to join a crypto startup, but there is no way for me to tell which of these DeFi protocols are going to be around in like, five months. There's literally no way of knowing, you know, so it was like really hard for me to figure out like, okay, I know the basic technicals. 

But there's no job place for me to apply to. I don't really want to join a VC sided period where I was just basically freelancing. So, I did some solidity freelance work with spectral finance as a start. And then I ended up joining consensus for five months or so full-time from I think it was March 21 to like August 2021.

I joined consensus as a business analyst because I was like, oh, maybe enterprise consulting will be fun. It wasn't my cup of tea. A lot of great people. There's a ton of smart people.

‍

It basically helped me understand the space better, and be like, oh, here's the history of the space and how these types of startups got to where they are today. And here are the key people that you should try and meet at the startups. 

So that really helped. I ended up being like, okay, I don't want to do solidity development because optimization and security is not my gig. 

And I'm like there's gonna be people who are 1000 times better than this, even if like solidity. devs get paid like three, four, or 500k. I'm just like, that's just not. It's not my thing. 

So, I went back to my data roots and started posting more about doing stuff. In April, I want to say of last year, I think that's when we first interacted. I was asking you a bunch of questions on what are these tables? 

How come none of my queries are running? Why is the timing, and you helped me…here's the basic sources and all that. 

And I think as soon as I started tweeting…the stuff I was doing in June…I started blowing up on Twitter more. Once you're like…start building a following on Twitter. 

It's just become such a strong cycle of meeting more involved, smarter people opportunities, and then it keeps building until you've hit a point of like, okay, I can't keep doing 10 side projects. I need to find the main commitment. 

And gram CTO of Mirror DM’d me and was like, “Hey, do you want to do this freelance project for us for like a right token, AirDrop?” and I was like, okay, I'm tired of DeFi. NFTs are starting to take off. 

And  I do want to try to learn how to work in a consumer app. And the freelancer project went well. I loved working with the Mirror team. I think they liked working with me. And then we just decided, alright, let's just make this a full-time gig. 

Boxer  19:35  

Yeah, cool, long-winded down. The journey has a lot of arcs. I didn't know you were at Consensus. That's very surprising to me. Yeah. That's probably the only guy on Twitter who was at Consensus and doesn't have X consensus.

Andrew  20:00  

I think I used to, and then I ran out of character space. And it was excellent talent board. 

Boxer  20:14  

So, I guess…what drove you forward or what really helped? While you were learning how to Web3…personal connections…you mentioned this one Zoom call, which seems to have had some transformational value to you. 

So, in general, how did you approach trying to learn as much as possible, like learn in an effective manner in this very convoluted space?

Andrew  20:41  

Yeah, um, I think it was, like once I learned the importance…this is gonna sound so cliche…but once I learned the importance of community…and Web3 reshaped my whole approach. 

Because I realized what phase not a space where you just study and work on your own and grow that way. Like you can do that. 

But it's gonna be very, very slow. Because even for data, for example, the amount of data content that's out there, even now, very, very little, like extremely little, you know. 

So, if you try to study that on your own, and just try to figure out through doing queries yourself without talking to anyone in the community, you're in for a world of pain, you know? 

It's like, okay, everything's based off communities, there are key players that move across all these communities. How do you learn to track that pulse of people moving and new problems and communities coming up? 

So that you know what to learn next? It’s kind of a way that I thought about it…like I wrote a…hold on…I'll basically the mainland head versus like talk to people. And like ask them who…

Boxer  21:51  

What do you mean by key players to? Instead like an actual person?

Andrew  21:55  

Yeah, like actual people? Yeah. Um, so I like people like Austin Griffith, for example, like looking at the stuff he's building, and the people that he's talking with or looking at ETHGlobal and seeing who are they partnering with? 

Who are these two sponsoring all the time, you know, and you're gonna start seeing on all these live streams and whatnot, and then this course, you're gonna start seeing some of the same usernames keep popping up? 

You know, or like Patrick Collins, you know, you're gonna see the same names.

Boxer  22:25  

Yeah, I see. I see what you said. Okay, like the last DeFi summer, the end of 2020. There was a bunch of these chords where I used to run into the same people all the time. 

No, nowadays, not so much just because we have so many people now. But back then it was like, every Discord server was a few hundred people…like the probability of running into someone, you know? 

Yeah, I don't know. But that's very interesting. So, you were trying to uncover these hidden networks and follow the trail and see what these people are up to and follow them.

Andrew  23:06  

Yeah, my biggest insight for me is then just writing about what you learn, and putting it out there, so every time I learned a new framework or thought around something, I would write about it, and put it out there. 

So, when I was studying wallets, I would study the full history of wallets up to looking at all of the GitHub commits of Metamask. And seeing like, oh, this is how walls were built up. 

And then I'd write an article about composability of wallets. So, one I have notes to look back at, into what I share. It's like, oh, that's gonna get me…that caught me a conversation with Dan Finley from Metamask. 

And we had a call, you know, andthat would not have happened if I hadn't put the article out there. You know, if I just done research and tried to reach out to him, it would have been a lot harder. Yeah. 

But the main thing… I wrote…I don't know how to put this. I put it in our private chat. But this is back when I used to read on Medium. I wrote this thing. I'm going to share my screen.

Boxer  24:03  

Yeah, Thomas in the background can put this in the actual chat.

Andrew  24:08  

I wrote this thing. I'm networking in crypto. And essentially, you have to try to understand how to navigate content, conversations, and communities. 

And it's basically, what I realized is everything is fractal in Web3. Because in Web3, if you have a new problem come up, right, whether that's in math or DeFi or whatnot. It’s really…it's a decision tree. 

It's just a subset of an existing problem. Right? And someone from an existing community is going to probably go and lead and build the new community. They’re trying to solve that problem. 

Right? So, it's all connected and all these people are moving around these communities, right, but it's all fractal. 

You know and this is different from traditional industries where it's like, oh, these fractal communities are happening, but they're only happening within Google or only happening within Facebook. 

And that doesn't go across industry, but it went through…every single topic is fractal and connected. And once you learn to see it that way, then you can start looking through like, okay, what are the layers of things you have to understand…the problems and communities around tokenomics around memes protocol, governance, and now there's a lot more back then it was much easier. 

But it's basically like I have a mental map like this of different meetings. And if I see someone move from one to another, I'll be like, oh, there must be a link that I'm missing between these communities. 

Otherwise, why would they suddenly be reading about this or studying this? Right? And then I'd either go and ask them or just keep an eye on the data and be like, is there something happening?

Boxer  25:49  

So basically, a third dimension to this picture that you just showed, and like, it turns out, they're really close in another dimension.

Andrew  25:58  

It's really fun to think about it that way. Because then it's like…all these problems that look like splitting branches suddenly connect. There's all these Web3 moments where you’re just like, oh, shit, lmind blown. Oh, that can be connected that way…that changes everything about how I thought about this, you know? 

And yeah, but you have to have a basic framework. Otherwise, you're just gonna get so lost, trying to read this on Twitter and be like, oh, what's trending? What's the most important like…? And it was?

Boxer  26:33  

What does your basic framework even look like?

Andrew  26:37  

For me, it’s all notion notes like…

Boxer  26:40  

No, no, like, more on a conceptual level? Like, are you only trying to follow data people nowadays or only? 

I don't know…you're probably like in this Mirror and FTE writing bubble, and then the data bubble, and then probably some other bubbles that I didn't even know about. But how did…because you're saying…you can't just consume everything…but rather you need to focus on what you're doing. 

So, how do you decide that? And how does that inform your information diet as well?

Andrew  27:11  

Yeah, yeah, definitely. Um, I normally try and pick just one topic at a time to follow. So, right now, it's social graph analysis, right. And that's going to be the only thing that I pay attention to feed wise for the next month or so. 

There's gonna be certain key people that I follow. I'll still pay attention to the stuff they post, but I try not to get distracted by…even if I see something cool in the med space, I'm like…okay, that's cool. 

But it doesn't connect to what I'm trying to learn right now. I'm not going to waste my brain time on this stuff. 

I don't know if that makes sense. But it's like very much you follow a decision tree of like, okay, what affects what I'm working on right now. What's gonna be big in like three or four months, and what actually has enough content to read about and you find that Circle? It's like, okay, this is the thing.

Boxer  28:13  

Yeah, I see. I see. So very much the focus approach, like that's this tariff stuff.

Andrew  28:18  

I'll read whatever thread goes in the telegram or slack, but I'm not going to… because I feel like that's gonna get wet through data burnout. 

And we've all been through this. It's just like, oh, every new thing that comes out. I'm gonna try to write queries and do a dashboard on it. 

You know, and you could do that for two months, but you're gonna be so burned out and tired. That like, they're not going to have any output for another few months.

Boxer  28:48  

They'll be…would like to have a word with you. Yeah, he's been doing this for so long now. He’s crushing it. Yeah, I guess he doesn't have any other job. 

Right. So, this is his full-time job to to do this. And meanwhile, you also building on Mirror, and you have responsibilities there. So, I guess it makes it easier if you were…I wouldn't.

Andrew  29:19  

I wouldn't. Even if I was doing full-time I would burn out by now.

Boxer  29:24  

Okay, fair enough.

Yeah, your journey is one of a kind. It's pretty interesting that you go from being I don't know…like very non-technical and then being like one of the…I don't know….thought leaders in the Web3 data space or something like that in a year or something. It's pretty amazing.

Andrew  29:50  

Only in crypto. You can't do that in other industries. No, yeah.

Boxer  29:53  

Oh, one thing that I forgot to bring up…um, this whole fractal thing that you were talking about. Yeah. All of that is enabled by all of us being on the same infrastructure, right? Is that your take on this? Yep. Yeah. Okay.

Andrew  30:12  

I think that's the only reason it's all self similar connected. Whatever someone does on one end of the network affects everyone else and all the infrastructure.

Boxer  30:26  

A lot of butterfly effects happening there. Yeah. Very interesting. So, we've talked a lot about neuro, but I know what Mir does, but I don't really know what the role does. So, the elevator pitch you go. 

Andrew  30:44  

Yes, I think I need to give some beer history first. Obviously, okay, the people here we're all. I think we're all very smart. We love to build very enthusiastically for all the fairly easily distracted. So, I think last year was like, oh, you started as a writing platform in 2019 or 2020. 

And then, Daluz and NFTs, both started becoming popular at the same time. And it became like, oh, how many different ways can we do crowd funds and doubt tooling? 

Because that's where all the attention and money is at, you know, so we ended up building, so many different five or six different versions of crowd funds. Like we were like Dutch auctions, we built splits as useful. 

So, this is more focused, but when basic auctions had like reservable editions and whatnot. And it was just 10 or 12 different token races…well…iit was like 10 to 12 different products all in one. 

And it got to the point where it's like we have to spread out, and we need to refocus again, you know. The the last four or five months or so have just been like… 

Alright, go back to the core of writing NFTs, build on L2s, and really build like a Dow writing flywheel where you have a dowel that we already have a lot of the best writers on Web3 publishing on here, right. 

But there needs to be a strong economic flywheel that needs to get better… like stronger collector, flywheel, and better reason for the doubt to exist, and ways for them…like writers to participate in the first place, you know. 

So that's the focus of a lot of what you'll see comes out over the summer. We've already planned it, and it's ton of QA and fixing and removing things that we're like, alright, we don't need to do build just for Dows and their tools anymore. 

Because if you think about it, Dows are a big opportunity, but the number of dials that are actively using all over the lot of these downswing platforms, in like flexible ways, and more than just connect to discord. 

It's very…maybe 20 bigger Dows…yeah not just like for once I have like over 1,000,002 million at least…and they are actively doing more than just one NFT sale. Yeah.

Boxer  33:20  

Yeah, I see what you're saying. So, there's so much dopamine but not a lot of active significant dollars to get around, basically.

Andrew  33:29  

Yeah because a lot of downs…they have an initiation phase, which is really strong, but they don't have…besides something like noun style. 

They don't have like sorry, I live next to it. They don't have some economic token flywheel where they have figured out like, oh, NFT issuance membership tie in and connect to next season or whatnot, you know. It’s so early, and that's still going to take a while. 

So, alright, let's just build for writers, which makes sense. And there is a flywheel for…versus just building for dowels where it's one. There aren't that many into, like, if you're trying to build for super rare versus like Blackhand or something, good luck. They're gonna ask for two very different things, you know? 

Boxer  34:21  

I see. So, maybe let's take a step back. So, you guys have built a writing on which anyone can host their own publishing or writings and mint that thing as an NFT to whoever supports this piece of writing that correct assessment?

Andrew  34:45  

Yeah, yeah. So okay, given the elevator pitch, right. 

Boxer  34:49  

Can we skip that phase?

Andrew  34:53  

Given planation like, okay, you think about economic models. We have our writers today. Need em is pay-per-view, right? They literally…you get paid more if someone subscribes to media while reading an article, which is completely random, you know? That's the only way you get a larger paycheck. 

Otherwise, it's pretty arbitrary pay-per-view. They started doing bounties where it's like, oh, if you're one of our top 100 writers of this month, we'll give you 50 bucks. You know, I think I was in their top 1000, and they gave me 50 bucks. 

And I was like, okay, q sub sack is like, you know, pay-per-subscription subscriber, right? Which is…it's better. But it's still homogenous, right? Every subscriber is the same. 

You could have a Patreon-style tiered-subscription. But there's not really much way of customizing or showing how much you like someone's writing with those traditional models, right? 

And I'm not even talking about medium pay-per-view. But when you have the NFT is one, it's like, you could still use it in a subscription style method of like, “Hey, if you buy an NFT, then you're subscribed to me for the month, you still have that.” 

But it's also like, it gives the collector and the author a lot more community. Personalization saying, hey, I want to find someone who really identified with that article, you know? Or someone who's really committed to…it's like, I don't…okay, the elevator pitch is getting to walk. 

But basically, it's like, you're giving them a way to like, subscribers to distinguish themselves from other subscribers, right? And building a more personal relationship with the author. You're giving the author a much better way of monetizing, I think, than just like a flat fee because if every subscriber is homogenous, you can only do really a flat fee. Right. 

But if you know, hey, I have 1000 subscribers, but I have 10 huge fans, there's something I can do with a pricing model specifically for the fans that are willing to pay like $500 per article, you know, and it's like that gets your writers mind thinking like, oh, maybe there's more I can do with this. 

And then you add that to a dowel of like, Oh, if I'm a great author, I can be part of the Mirror doubt of it with a ton of other great writers. 

Like, there's also a lot you can do with that as well. And like that stuff that can't be done what to and so that's where mirror is trying to focus on, so work on whittling that down.

Boxer  37:29  

So, that was a good pitch, please send me the deck after the…get you in the round.

On the one hand, you're building this platform, which is open for anyone. And then on the other hand, you also try to facilitate a writer dowel, which is about producing world class content. 

And I guess it's Web3 expertise, or are people writing about like…I don't know…how to make the best honey in Alabama or something as well.

Andrew  38:00  

Great question. I think right now it's one through expertise. Obviously, we would love to have someone writing about the best honey in Alabama. But, I think that's just one of the technical barriers of what…

So, right now, just having the top with three writers figuring out an economic model for them and a membership and collector model that works for users. And wrapping that into a towel. 

Because it's how you like…that's one of the strongest competitive points, right? It's not really just your product. Anyone can afford your product…like Mir is not that hard to fork. 

Like someone's literally wrote…I forgot who wrote the code for Kashmir as like a dev tutorial, you know, because all of our design problems…yeah all of our design objects for what it's called that sits as a solid NPM package…

So, you could use and rebuild near at least the basic surfaces in a day, you know, but if you have a strong down…a strong community….that's what keeps people there. Yeah. So, that's…

Boxer  39:12  

Yeah, and also for Mirror, I think it's a lot of brand awareness as well. You guys have built a very, very strong brand. So if somebody now…yeah, I don't know. I would much rather click on a neuro link than a medium link already. 

And then if somebody just faux Mirror and call to…what do I know? Then like, I'd be very confused about that link even. Yeah, definitely besides the point. Yeah, there's definitely some defensibility there. 

Um, but so what do people do in this writer doll? Do they change thoughts like revenue each other's pieces work together to make a bigger impact? What's the model there?

Andrew  39:55  

Yeah, well, okay, right now, it's just focused on curating other writers that's literally the one use case of the Dow right now. It's like every week, members will get together and be like, hey, what's some of the best writing we've seen on near each week and Rafa, the builder on Twitter, leads our Dow. 

And he really just like spearheads talking with members, talking with the team, finding the best writers, and getting them on Tamir. Patrick helps with that as well. So, that's like the main focus of the doubt because it's like we got to get a ton of great writers first and build up the Treasury. 

And then we can start funding a bunch of fun initiatives and bounty programs and leaderboard incentives and all of that…or even just focused figuring out what we want to do with the governance token, you know? 

All right now, it's literally just like keep a very focused one thing…your only proposals are, which writers do you want to see to join the Dow? You know, and that's it.

Boxer  41:06  

Okay. Yeah. But sometimes simplicity is the beauty of things, especially in this very noisy world of Web3. 

It's sometimes good to really hunt down and be like, you can give the dashboard the star or not the star, nothing more. You can or not…that's it.

Andrew  41:29  

Yeah, just keep it simple. But that's the key. That's the key.

Boxer  41:34  

So, can you share a bit of the data that you've been surely pulling about neural and doing?

Andrew  41:41  

Yeah. Ah, okay. Let's see, how do I want to start this? I'll start with the concrete before it just goes into fully abstract. I mean, this is just like a basic NFT dashboard. I don't know how much to talk about. 

This was basically your typical NFT like show the mint stuff. Okay, let me just go maybe one by one. We're putting more and more metadata on to on-chain, so it's a lot easier to be like, alright, let's say you were just looking on Dune, and this would link out from a leaderboard dashboard which shows which NFTs are trending right now. 

You can click here and see the name, the description, and total supply and mid-status with…this would show how many NFTs are left. And since it's closed, you can just click on it. It takes you to quixotic shout out to Michael for figuring out the HRF stuff there. 

It shows the basics on purchases, and it’s raised with a bot removal column because optimism bots to things. You have to give a scale if I don't remove bots. This number is in the hundreds of thousands.

Boxer  43:02  

I was gonna say like this optimism NFT was super, super successful.

Andrew  43:07  

Yeah, they're like five…I think it even times out to like 500,000. Okay. Everything has to have a no boss filter.

Boxer  43:17  

Yeah, I think how do you do that? Or just Michael’s?

Andrew  43:23  

So, the way I do this is just like looking at people who only call the contract directly. Because a lot of the bots use a self destruct contract basically to mint things. 

It’s like, okay, the people are our boss. It shouldn't be the people who met through our front end and call the contract directly. It's not the best, but it works for now.

Boxer  43:46  

Yeah, that's actually…so, there's 480,000 transactions made by bots on optimism just for this mirror NFT.

Andrew  43:56  

So, they batched it in this contract, so they can make 10,000 at a time and one transaction. It was whatever 480,000 divided by two…like 500 transactions.

Boxer  44:11  

Wait, so the bots minted 10,000 NFTs at once?

Andrew  44:15  

Per transaction so they have a contracts that batch minutes and then self destructs when it's done. So you get refunded the gas costs for like deploying the contract and whatnot. And it's actually pretty advanced. It took me a while to look through it.

Boxer  44:31  

Yeah, that sounds…I mean, yeah, good on them for figuring that out. But I mean, Michaels working optimism…like have you seen your adversary? Don't you think Michaels got to notice?

Andrew  44:44  

Yeah, we're talking about like, is there anything we can do about this? Anyways, that's still discussion. Okay. We'll keep that private. I have some wallet history stuff here. 

Since I think for every mint, it's nice to know who's transacting, so in the media, wallet has been on optimism for 52 days. This is how much gas they have spent. Most of them have spent…they've had most of them have only had eight transactions. 

And they only have like 0.00 to 5 ETH in their wallet, which is not much on main net. But that's enough to buy like 500 writing and FTS on…yeah. So, at first, I was disappointed, but I thought about…and I was like, you know, that's good. 

That's a good amount of play money. Classic distributions, and then this is an author leaderboard. But it's not going to make sense until the more entries tries to use you in a deeper dive. The data exploration dashboard for people…because obviously, we're gonna have this data in like our front end. 

But it's like, if someone just wanted to explore things quickly, they should be able to click around dashboards. And the secondary market, I think this is fairly standard across NFT dashboards, just highest price, lowest price volume, like getting their hold or distribution of like 5% of people hold 47 tokens. 

27% of people hold one token, you know. It's just like being able to quickly look at this stuff and be like, alright, how many people are at each token range? And then trading protocols, this is probably something I'm going to work on a lot more. 

But like looking at your top traders and just being like, hey, how many tokens have they traded? How many did they meant? How long have they been part of the ecosystem like part of this collection? Sold? Purchased? Bought? 

Pretty basic stuff. But it's good enough for you to be like, alright, let me put in any address and just see what's going on. I feel like this is like table stakes…went through data analysis now. I think it did take us a while to get to stuff like this. 

Boxer  47:09  

I feel like, yeah, we're growing together. We're growing together. So, this is the optimism data. Yeah. Do you have any idea how many Mirror NFTs got mentored on? In total, I guess you're on main net Polygon as well. Or only main? 

Andrew  47:28  

No, not on Polygon? Let's see. Mention it on Main net? That's a good question. I actually don't know if I track that. I only track ETH raised from all of your side hustles. 

Boxer  47:43  

Exactly. Yeah. Yeah, I actually don't know. It's interesting. How many, at least how many unique writers know that it actually doesn't get transmitted on-chain? Right?

Andrew  48:03  

I could check for unique. Yeah, I could do unique writers. This is probably pretty close. I'd say 9700 or so have probably minted. And then their IDs are empty. I guess cuz we looked at this we're like, okay, let's just recommit and do this on optimism. Yeah. Yeah. This is like the front facing data. I can dive into my big data projects, if we want.

Boxer  48:32  

Your social media or social graph stuff. Yeah, maybe real quick. 

What made you guys go to Optimism? The only two, if you can talk about it, obviously. I know.

Andrew  48:49  

Yeah, I can definitely talk about that. I think a lot of it was just timing of like, we've been talking to a bunch of LTE teams. We wanted to have a chain where we could both implement NFTs and also write race and a bunch of our other products on. 

We just found through a bunch of conversations, Optimism was a good partner to work with at the time and worked with our existing infrastructure on Alchemy because a lot of other chains like Alchemy don’t fully support…they're still working in beta, so it's really just like…everything came together. 

It was like okay, let's just do this on optimism for now. Sorry, my dog is having a fit. And yeah, it's gonna be the only chain okay because I think it's like…it's not a DeFi product where it's like, oh, I can have each pool on its own chain, and they don't affect each other. 

But writing NFTs, I feel like it gets our consumer products in general.I feel like it gets confusing if one writers running on Polygon and another ones writing on optimism. 

And if I tried to do some composability thing or trading thing across my collection, like it just feels like it can get confusing really quick. 

Like, oh, which NFT sits where? And like, what if a writer minted on Polygon, but now they want a mental optimism because there's no more volume on optimism? Do I have the same NFT on multiple chains like do they even have funds on?

Boxer  50:23  

Do those chains like? Yeah, very interesting problem space. I guess that could be solved if the user interfaces. 

So, the wallets solve this in a better manner. So, you don't even notice on which shengyuan. But for the time being, this certainly feels like the the way to approach this.

Andrew  50:45  

100%. Yeah. It's just less dev overhead. As I'm sure you all know, adding each additional chain, it's not as simple as changing chain ID.

Boxer  51:00  

Just call the nodes, man, no problem. No problem.

Andrew  51:04  

Just add a new node.

Boxer  51:07  

Yeah. Interesting. Yeah. And maybe before we…Yeah, I don't know, if you want to dive into your social graph data, I'm certainly very interested in what you're what you're picking up.

Andrew  51:19  

I can talk about from like a theoretical level because I think like this, this is like one of the things that's kept me very interested in the space from a data perspective. Identity has always been super interesting in Web3, right? 

That's the whole reason I did the social graph shop and whatnot in the first place. Because you have your wallet with all of your transactions, all of your NFTs and tokens, right? 

And there's basic identity building blocks. You can build off of that just like, okay, who bought what NFT? 

And then you have layers of context on top of it, which are like your different communities. Which is more than just saying like, “Oh, is this NFT from our blocks or not?” It's saying, “Oh, if you bought a fidenza, who else bought a fidenza?” with you. 

And that's your community. Right? So that's how you start to build this social graph of interactions. And basically, I had a split into just trees. 

Last time, not last time…like the last two years basically I spent thinking about a decision tree of like, did you do this? How long did you do it? For? What community? And who else did this with you? 

That was the basic side of it. And I kind of…I don't know why it took me so long to realize this. But about like two months ago, I started looking at this and being like, okay, if you were to build a recommender algorithm, right? 

For NFT content, the same way that you build it for like Netflix, could you just build it off of like a matrix of users? And are they buying the same NFTs? Or could you just do it that way? 

And the more I looked into it, the more I realized, okay, a big difference in Web3, is the ownership perspective, right? It’s like if I watch something on Netflix, or if I don't watch on Netflix, it doesn't really affects me or other users, you know? Like, whether or not I watched…

Boxer  53:34  

Yeah, just consumption, like it doesn't change. And the same flip stays switched off…

Andrew  53:41  

The same thing on Twitter or anywhere else. If I like something, it doesn't really…I didn't hurt anyone. And I didn't spend anything to get it, right? But one three is ownership base, which means everything now has game theory attached to it, right? 

Which means if I just…if you just do a basic recommender algorithm with your matrix ceases ones and zeros. That doesn't do as good job of…they might help predict like oh, what stuff do I want to read on here? 

But it doesn't help predict what stuff do I want to collect on you, right? Because the collection aspect is now dependent on what the supply is. 

And if I've already been collecting with this group of people, I have to look at did those like…I, as the data analyst analyzing the character, have to look at that same group of people collecting the NFT. Because your social graph, the game theory aspect of it now becomes a predictor of whether or not someone's going to collect that NFT or not…not just do I like that content? 

Ultimately, you want to drive views and the more views you drive, the more collectors you're probably going to get, but  there's probably a better social graph algorithm that you can build to specifically get people to collect. So, you have to figure out, from a network graph perspective, okay, this writers put out 10 articles, who are the people collecting those articles? 

How are they related? And what is the tipping point that causes them to collect the article? You know, maybe it's these two key people in the group have to collect, and then these other five nodes are going to collect, or maybe it's these two people collect. 

And there are four people over here who always just buy from the secondary market later, like with a two-month or three-month delay, you know? Once you find these relationships, you can start building. It's basically a random graph algorithm on top of a game theory algorithm. 

And obviously, in the interface, all you see is XYZ article and ABC collectors, right? And it looks really simple. Yeah. 

Behind the scenes, it's like we've carefully chosen what NFTs are being shown and which collectors are being shown, you know, because we know the social graph of what's affecting you to collect, so you start building these community-based identities.

Boxer  56:13  

Alright, so that's already in production, right?

Andrew  56:16  

No, no, God, no. I was gonna say I'm painting the image of like, why?

Boxer  56:24  

No, that makes so much sense. And then also, it’s based on whoever's looking at it at the moment, showing you want to make it more common? Yes, yeah. And then you're gonna need a lot of compute power.

Andrew  56:38  

Yeah, I'm working on that now…on like, looking at existing collections, but like no NFT collection or no NFT creator has put out things consistently enough. A high-enough quantity for this to be analyzable. 

So, a lot of this stuff is theoretical that I'm looking at now. But it hit me with, oh, you have to have, obviously, your typical reference matrix. You have to have the random graph, but you really have to capture the game theory element. 

And there's not many real world examples of this, like the best examples, international economies, you know, where like, all of your nodes are these different countries, and you're trading commodities, right? 

And these commodities come out and editions of XYZ at a time. And you might like certain creators more than others, right? 

And that's what I'm reading to understand, like how to do this, because it doesn't exist in traditional tech. Like there's no reason for Facebook. So, apply game theory to what's led me to the show…

Boxer  57:43  

Isn’t there kind of a “Couldn't you also look at the VC space?” where it's like if one of the VCs goes into round, two other nodes will most likely follow? I think that immediately comes to mind.

Andrew  57:57  

Yeah, that's a great one. I hadn't thought of this and maybe make a lot of money.

Boxer  58:05  

I don't know about that. But if you want to do further reading, that sounds like similar dynamics would happen there. So, you very much think of this consumption aspect and this ownership aspect, in very, very, very different terms. 

So, the consumption algorithm wouldn't be the same as the actual ownership. I run. And that's very profound. I just realized this, and I'm like, holy shit. Yeah.

Andrew  58:38  

I was like, it makes it's obvious once.

Boxer  58:40  

You said…yeah, yeah. People should recommend things they want to own because that's something that I want to associate with. And like that doesn't get to build my identity now.

Yeah, exactly. Yeah. And what happens in Web2 social media today is that we get recommended content that kind of like…upsets us or keeps us on the platform longer. 

You're not what would be the ideal for me to… I don't know be a better person. If you really want to take it. Like, I didn't know, it sounds pretty absurd. But the content that you consume, basically programs you and kind of…oh man…

Andrew  59:23  

Yeah, exactly. So, I want to do that for writing, where the stuff you read is built off building your identity, not just off of what's going to get you to stay on the page, you know? Because we're not paying-per-view. We're not going to have ads, so we don't have to optimize for that, right? 

We can purely optimize for whatever helps the collective the most. Own the most writing NFTs like that helps everyone else in the ecosystem, you know, and think about the best part of this All this data is open. 

Once it's set, you can show this to the authors, and the authors can look at their community and explore this live, you know? It's all on-chain probably…we can build this into it like…

Boxer  1:00:13  

You can very easily then…I talk, I don't know, like traditional book authors. They have some kind of idea which other authors are being asked by their readers but not directly and…

Can they even contact them? There's probably some telephone number of whatever. But it's very hard to actually. And in with three, you can actually…like there's a like mirror profile attached where it's like the Twitter name is Vijay, and you can just talk to the guy.

Andrew  1:00:38  

Now again, that's my identity ecosystem. Salting service off of that…amazing. But the algorithm has to be written first. And that's if anyone watching this wants to help. Please DM me on Twitter. Because I'm learning from scratch.

Boxer  1:00:58  

Oh my God. That certainly is very profound.

Andrew  1:01:04  

That's what's been getting out of the bed for the last two months. So, I need to figure it out.

Boxer  1:01:11  

Okay, I see. So yeah, I guess, coming from that, what is the long-term game from for Mirror? 

Do you guys see yourself as like? Do you want to replace Medium? Or do you think we can take this even farther?

Andrew  1:01:29  

I mean, Table Stakes is replacing Medium. Table Stakes like base? That's the very minimum? Oh, yeah.

Boxer  1:01:38  

Okay, let's see. Sorry.

Andrew  1:01:41  

I'm using corporate speak. Bare minimum would be that I mean…basically, the goal is to build a strong dow of writers and an economic moat-like flywheel that works for all kinds of writers. You know, not just the big ones. 

And then, sub goals, building something where collectors can really showcase and build their reading identities. But that's very abstract and unexplored. If I've bought like 20 read and like 20 Writing NFT is that the best way of showcasing that?

Boxer  1:02:23  

Yeah. I also kind of…who looks at it? Yeah, good question. When Tinder integration or like what?

Andrew  1:02:32  

Exactly? Yeah. I don't know. But that's also more of a question to the design team than to the data team. My brain is not optimized for that. Yeah.

Boxer  1:02:41  

Um, I can certainly see a world where, I don't know, other people have a book shelf in the background, and they prominently place the books that they're very proud of that are like their ideology or that have really shaped their thinking. 

And yeah, transferring that into the eye, that actually makes a ton of sense. And then you can think about it. So, it doesn't make sense for me personally. I don't…I can. It's a bit weird to have these bookshelves in the background. 

But yeah, I can certainly see how other people would enjoy that. Very, very. Yeah. Eventually, I could basically host. There's Kindle self-publishing today. And then, I could just drop her PDF book on neural, I guess.

Andrew  1:03:41  

Yeah, he could. I mean, you could do that now. Technically. Yeah. But I mean, the hope is, if you're doing something like that, you can find a much better model for it on Mir. Because right now, you do no matter what. Like content on Mir, it's all the same, right? 

And that's not really ideal. So, the hope is like, oh, you find some customizable model that works for different levels of content. But that's like we can't really think of that stuff. 

It's more of like, oh, some writer is going to be writing on Mir for a while and then have some idea. Then talk to us, and we'll be like, okay, that makes sense. Yeah.

Boxer  1:04:30  

Yeah, certainly a lot of ideas to be explored there. There's not just…here's some text different ways of having written content. Yeah, certainly. Yeah. 

And I really enjoy Mirror and love what you guys have been doing. All the side tracks have also been very interesting to observe. What's maybe the most interesting project that got funded through…one of these sales or however…

Andrew  1:05:03  

Oh? Oh, most interesting.

Boxer  1:05:08  

Most successful I don't know…like what the team's favorite is city, Dow, where they bought like… real world plans.

They actually did that. Yeah. So they're like…what was it called Delaware? No. There is some Dow which I don't cooperate in the US somehow. Is that how they…?

Andrew  1:05:30  

Yeah, they did like a crowd fund where you could buy parcels of that land, basically. So, it's like a land sale but not in the metaverse? 

Probably going to…probably it's been long enough other deed was a while ago. That's not going to hurt people's souls. But they can actually do stuff out there land. Yeah.

Boxer  1:05:56  

So, there are no NFTS out there minted via Mirror, who verifies the ownership of a certain real world land asset? Yeah. Okay. How much was it? Like one parcel of land?

Andrew  1:06:10  

Ah, I'm looking now. I don't remember.

Boxer  1:06:14  

But in the order of like, 1000s…10…1000s…100…1000s.

Andrew  1:06:20  

I will have the answer soon. Okay, they have a slightly complicated…because they did a crowd fund first. And then they have like a membership thing for who can like you had to be part of the crowd fund to participate. Okay, I see. And, anyways, that's like one of the favorites.

Boxer  1:06:54  

Okay. Yeah, that sounds sounds very interesting.

Andrew  1:06:58  

One of the membership tokens was like seven years. Okay. So not insane. And like, crypto terms.

Boxer  1:07:06  

Yeah, depending on when this was…like 20 to 30k or 15 to 30k. Yeah, stuff like that. Yeah. Interesting. Interesting.

Andrew  1:07:20  

Yeah. Other favorites. Blackhand…obviously a favorite. It's like a Web3 based crypto gaming organization. So, they have like Halo team…they have an apex team. 

They funded like 1.3 million on mere eSports. Organization. Sorry, yes. E-sports organization. And they have a few players on their roster. They're super cool. 

Etherium film was like the biggest one. For sure. And obviously, we're still very excited to see what comes out of it when the film drops. That one raise 1000 ETH. Which is also like a lot. 

Okay, now that's a lot. Yeah, for formula. Yeah, obviously, it spans a spectrum. But now, we've mostly just seen like a bunch of like 20 to 30 crowd funds. I would say for random people's smaller projects. 

I don't personally prefer that, or it's huge…1000…these projects were more just like these random communities that we've never heard of. And like we'd look at the crowd fund. We're just like, what is this?

Boxer  1:08:48  

Yeah, that's the beauty, right? Like any community. 

Andrew  1:08:51  

Yeah, exactly. I guess. I would have never seen this on Kickstarter. Like people funded their whole…someone funded a crazy Dungeons and Dragons campaign off of it. I'm like, oh, okay.

Boxer  1:09:06  

Okay. That's cool. That's interesting. But wait, it's more fun to do a Kickstarter where people actually get stuff back or like an NST back or something or in a GoFundMe kind of way as well, where people are like…I'm in the US and now bankrupt.

Andrew  1:09:29  

It's marketed in a Kickstarter kind of way. Like most of them are like, Oh, if you participate, oh, sorry. Like a big example. A good example is like mad realities TV. They funded a dating crypto dating show. 

And they fund it on the air, and they have like three tiers of NFTs where it's like, oh, based on how much you contributed, you get more voting power of which candidates go on the show? You know? 

Boxer  1:09:56  

It’s super interesting. Yeah. They're really interesting.

Andrew  1:10:00  

They just raised 11 million, I think. Excuse me. Like a seed round. Not on here.

Boxer  1:10:11  

Yeah, but then 11 million seed round. What the fuck? I guess I've seen. They're actually producing videos. They've already had one. Okay. Okay, so I don't need just like a bunch of computers and developers.

Andrew  1:10:27  

They're like immediate. Okay. Okay, I think paradigm was involved. So, I trust it. They gotta get…they got a really smart team.

Boxer  1:10:35  

Yeah, it sounds really interesting. I'm just…

Andrew  1:10:40  

Hey, like all these crowd funds they have…promises in the entry, but Mirror is not responsible for….

Boxer  1:10:51  

…no securities you guys…

Andrew  1:10:58  

Fom a technical perspective, it's like purely GoFundMe style. Yeah, no, okay. You raise how much you raise, you get these tokens? They're there or an NFC or whatever? Yeah. Yeah. Okay. Interesting. So, yeah, cool. 

Boxer  1:11:14  

Yeah, it’s very interesting that you guys started all this writing platform and then spun off into Kickstarter. 

And now you're back to…let's actually make a great…it's certainly an interesting development. I don't know. It kind of makes sense in terms of like distributed ownership economy where it cut out the middleman, enabling all of these projects to thrive. 

And obviously, if you want to introduce something to your community, you're gonna be using some kind of blog post article. And that seems to be a very natural connection between those two things. 

Andrew  1:11:51  

Yeah, exactly. Yeah. And also just the crypto space is very distracting. Just think about how many NFT drops were happening last year and constitution dollars last year…

Boxer  1:12:05  

Yeah. People bought the Doom book. They realized afterwards they don't have the IP.

Andrew  1:12:15  

I haven't heard any.

Boxer  1:12:16  

So, it's still the best story.

Andrew  1:12:19  

Point being, there's so many distractions like that. And it's just like, I don't know. Yeah, that's fine. It's hard to say.

Boxer  1:12:26  

Yeah, but you guys have done a great job. I guess, all of these things have been a success. No. Is there? Is there one point where you're like, we should have not…like this was a total? Oh, they probably reserve auctions. 

Andrew  1:12:37  

But that was like very early. Like you've probably…it's like basically bizarre auctions. Yeah, we had that. And it kind of works.

Boxer  1:12:52  

Yeah, but for written content. It's a bit. Yeah, maybe for a poem or something. That was backwards. Somebody who writes poems.

Andrew  1:13:00  

I'm probably…I don't know. I don't know. 

Boxer  1:13:06  

Sounds like an interesting avenue.

Andrew  1:13:08  

I could do a text search. But that was back one of the days where I thought like, oh, all of these writing NFTs are going to be one of you know…so it makes sense to have an auction for it. 

But then…it became apparent, why limited to one collector at some random price point? We should make these additions. And as soon as you have additions, that auction becomes pointless. 

And it's like, okay, if someone wants to use our auction, they need to mint it on Zora or OpenSea and use our auctions. And it's like, okay, that's a lot.

Boxer  1:13:45  

Yeah.

Andrew  1:13:47  

It made sense at the time. And then we deprecated it because it's like, okay, we're never not…

Boxer  1:13:55  

I see sec. Somebody learned something during that, I hope.

Andrew  1:13:59  

Yeah. A lot of learnings a lot. Yeah.

Boxer  1:14:04  

Yeah. So, that's Mirror. I've certainly been very enlightened during our conversation. No, I've never really thought about this whole ownership consumption like the battle that's happening. 

Very cool. Yeah. So, Web3 education. You certainly are one of the great heroes in the data analytics space, bringing education forward. So, where do you think the general with three education landscape is today, and how could we maybe improve upon it? 

Andrew  1:14:41  

Yeah, that's a great question. Something I've thought about a lot. I think, Okay, last year, I think we're in a space where it's a technical…I'm just going to talk about data education. I don't want to get into the rest of it because it's complex. 

Yeah. But on the data side of things, I think last year where most people were struggling was all the technical aspects of data. 

This was even before the dude had all the great abstractions tables that it did now. There was the decoding process. It was eventually, but it's not as nice of flow as it is now. Yeah, right. 

I definitely had a few times where I was like, wait, this is not the right address. I think it's just the technical knowledge that people were not as good. Even Etherscan wasn't as strong as it is now. 

So, last year, my focus is just putting out technical content of like, okay, we needed like help. People understand at a basic level how contracts work, how the three tables do work, how to vote, and how these ERC standards work. 

Because that was where most of the questions I got from people were coming from. And then that all culminated in our network for like the 3030 students Thursday's thing, where it's like, okay, I think people are starting to get the technical side of things, but they're not understanding the community side. 

I don't know what it is. But a lot of the data analysts I talked to would play with crypto data. They would find it fun. And they wouldn't do anything else with it because it's not like hackathons where it's like, oh, I can build this like new escrow app and get paid by sponsors to build this app and put it out there. 

And it stands alone, right? For data stuff. You do some obscure query. No one's gonna read your query. You probably don't really know how to interpret it as in like someone new to the system. You probably don't know how to interpret it in the context of everything else, right? Because there's so much noise coming around. 

If you just throw out a TVL metric, people are gonna be like okay. Most people are too scared to tweet anyways. I was like, Okay, this is a problem, we need to teach people how to engage with the community. 

So, that's what I did on our course. I think we did a good job of basically teaching those 30 people not only is this how you do Krypton analytics, but this is how you should do a write up of it. 

And you should be tweeting this every single day. We were basically pressuring them. Like tweet this. Put this out. I'm gonna tweet this for you. And that got like…I think quite a few of them. Obviously, 15 of them are hiring full-time now.

Boxer  1:17:42  

It’s absolutely mental. I think my internal count is 12 or something, but maybe I had a few people. It's absolutely crazy. 

Yeah, it was out of 30 people, and there was like…I think 27 actively participated. So basically, 55% of people actually have full-time jobs now or 50/50 or something. 

Andrew  1:18:07  

But it's constantly learned how to engage with the community. Like they went from just attending the calls and asking a few questions to actively…I saw a lot of them were like active in the doing discord. 

Even after the class is over…they'd be answering questions or… like Grace and Tuition would put out like articles on Twitter of stuff they're looking at. Seeing that was like, okay, it clicked for them. This is a community-based game. 

It's not just like, analyze, put out dashboards, and wait for people to ask them questions about it. Because if you just write a dude, create a new dashboard, and put it there, and don't do anything, it almost doesn't matter how good of a dashboard it is. No one's gonna go and be like, oh, hey, look at this great dash.

Boxer  1:18:57  

Yeah, in very rare cases, if you've really hit the nail on the head with like…if you have like a great Luna…you wouldn't like people search for Luna and then they're like, oh, this is like the best dashboard. 

Let’s actually pull the data from there. Let’s make a Twitter thread using this guy's data. If somebody starts consuming your dashboard autonomously, then the chances of that are usually pretty slim. Yeah.

Andrew  1:19:30  

I agree. Yeah. Even when that does happen, that person normally doesn't have a Twitter. 

Boxer  1:19:35  

There’s no way. Yeah. Recognition use you.

Andrew  1:19:39  

Yeah, it doesn't help you, you know. It's like, okay, cool that’s Twitter…no one's gonna be like, oh, wait, let me go look at that handle and try to contact me, you know what I mean? Like, no one does that, you know, so my focus now has been like, alright, how do we teach people to be more community engaged? 

That's probably going to be my focus for this year. Because I think like last year was like, Oh, hey, here's this Azureus. Do this analysis. Get the daily sum of whatever trades on this unit. Swap the two versus three thing, right? 

Versus this year. It's like, alright, I'm working, I'll drop the tea. I'm working on like a 20 or 30 problem practice set, basically have like data questions. It’s like, Hey, if you want to be a data analyst in the space, you need to be able to do these like 20 questions, but they're formatted in a way of like you saw something on Twitter. There's no links to any addresses or anything. It's some obscure thing going on. It’s basically stuff like that. 

Boxer  1:20:48  

It’s like a real world example of like a treasure hunt.

Andrew  1:20:51  

Exactly. Because I realized you need to teach people to start from like a Twitter thread. So, they're forced to go and talk to the communities. I can't just keep putting out your address. 

Here's the table schema. Calculate like that because people don't struggle with the sequel side for the book. Okay. For the most part, the sequel side, I think there's enough resources on that learning. Yeah. 

But going from a random tweet about Luna to actually analyzing contracts, that's the hardest step, you know. So, that’s, I think, a big opportunity from an education standpoint.

Boxer  1:21:28  

Yeah. Interesting. Teaching even more like not DeFi context, but rather the whole environment, like yes, this early Metaverse environment where it’s like, “How the fuck do I actually navigate all these different like, resources?” 

Yeah, dogs off the project…and this code of the project…and kind of like…can I ask what are just sequels this year? Exactly. That's super interesting.

Andrew  1:21:54  

So, I think we need to find a way to get people there. Yeah, that's a big opportunity. Anyways, we'll see what happens. That's one of my goals for the next…

Boxer  1:22:04  

I see. Yeah, I'm certainly excited for the Andrew Twitch streams, Andrew's treasure hunting data, or something. Yeah.

Andrew  1:22:13  

But, yeah. I don't know. I haven't figured out the business opportunity side of it. If you're gonna ask me about that, I don't know how to generate money off. That's not my stress.

Boxer  1:22:30  

Maybe you could sell it as an additional neuro.

Andrew  1:22:34  

It's crossed my mind. Problem and hefty. Got it. You gotta do what you gotta do.

Boxer  1:22:47  

Yeah, cool. Um, so yeah, Web3 education. I don't really think we are also running out of time. So, if Thomas is here, he probably has some questions gathered from the chat. And usually, I’ve done them before. So, you guys, Thomas.

Thomas  1:23:15  

Alright, so we've got a couple of questions. I'll start with this one from Hilldobby. Why the username? Let me know. 

Andrew  1:23:23  

Okay, so that’s a classic. I think a lot of us have this. This was just my gamertag from middle school. I had this crush on this girl named Emily. So, it stands for I love Emily because it was unique enough that it was never taken on any gaming server. 

My other username was like XMen35. I used that for years, but then like that started getting taken. So, I was like, okay, you know what? I have this other username from this crush that I had. I've just kept with it. She has no idea.

Thomas  1:24:14  

What are some unintuitive lessons you've learned about engaging in Web3 communities?

Boxer  1:24:22  

Oh, unintuitive I mean, thank you was the most unintuitive. Maybe…give you time to think it's…don't assume that anyone is too important to talk to you. Like usually if you like you can write the main developer of some projects like at 3 AM my time. And they, for some reason, actually answer, and this has happened and shouldn't have at times. 

Always feel free to like just DM anyone. People in the space  are super happy to help you, get forward, teach you, so take mentorship. 

But being in a position where people do reach out to me, like, try to have something like substantial that you actually ask me, because I don't really like…I like I talked to people…like shit on and like…the conversations that I enjoy….

It's when I actually see that the other person is getting ahead. Don't start a conversation with “Hey, man, how you doing?” No small talk, please. 

Like this doesn't… we are here to get something done. And if we've talked a bunch of times, then we can be like, “Hey, how do you like…” This is like a very different dynamic. 

I don't know, We're not like…we went to robot normies where you have to be like, “Hey, are you doing?” No, I'm great. Like, oh, yeah, I'm great as well.

Let's get to the extra like…no, just fucking ask me the question that you have. And don't ask, “Can I ask you a question, please?” 

But reach out to people. People are super helpful. I think that's very under appreciated. And a hidden gem in the space pretty much.

Andrew  1:26:26  

Yeah. I think my one and only biggest thing I say to everyone is contribute. Don't apply. If there's some community you want to get involved with…same with what Boxer was saying…don't just go into Disco and say, “Hey, can I help with something?”

Yeah, I personally hate when people just say, “Hey, how do I get involved? Can I help here? I'm a geek.” 

I'm sure you can. But like the people that I ended up working with the most are actually helping. They look at what I'm working on or what someone's working on. 

And they start to try to do something with it. And it's usually pretty ugly, you know, not gonna lie. But it's like, you see that they put in some sort of effort, right? 

And it's like, oh, this is someone who wants to contribute and has that blood, I'm gonna go reach out to them and like, try to help them on this or try to get them involved, you know? 

And like that gets you going in the system…not just sitting there applying on LinkedIn. Like please. Please don't spend time on LinkedIn applying to jobs.

Boxer  1:27:33  

Yes. You'll just wreck yourself. You're now probably working at the DeFi mullet startup. Funds an anchor.

Andrew  1:27:43  

Yeah, that's my biggest advice.

Boxer  1:27:45  

If you can financially afford it…that's always a part of this…where it's like some people can't just spend like…if she doesn't have time to contribute for free. 

So that’s always something to consider. But then…I don't know. They would have to not have a job currently and not be able to side hustle for some time. I don't know. That's usually like…I think the side hustle is very approachable. 

And then if that team actually starts hiring in the area that you've been contributing to…of course, they're gonnago to you first because you're already on boarded. 

Right? There's no question…who else would you actually hire? Then, this guy was like…obviously…does the stuff already and like Genesis should…

Thomas  1:28:34  

100% David says he'd love it if Andrew would discuss metrics dial since he's a councilmember.

Andrew  1:28:45  

Oh, yeah, sure. Metrics now is a spin-off of flip sides of bounty program. And a much more open and collaborative way. I think they're still very early on. I would say they don't have…they've worked on like a tokenomics model on the white paper. 

They do a few fun workshops here, and they're definitely like a good place to get involved in data if you're like…specifically, I would say if you're looking at non…not just your core EVM chains. I see like bad timing now. 

But if you're looking at Terra then they would have been like the best place to start. Still the best place to start is looking at Terra. But yeah, probably one of the bigger community-specific not product-specific data communities out there. 

I would say, if you're looking at Terror, look away. But yeah, great people a lot. I mean, again, I'm going back to my key people that you're gonna see the same people from across data in all these communities. 

So like, just find one of them and start chatting. Getting involved. I'm not gonna say, “Oh, get started here,” then like just start all over and overanalyze it. Yeah.

Thomas  1:30:23  

All right, great. Let's have one more from us to DeFi. The future of Web3 data analytics will require in your opinion, which languages and tools solidity…Russ, Python, SQL, graph, QL, etc.

Andrew  1:30:39  

I hope you don't have data analytics and solidity. Honestly, in traditional Web2, I would say Pythons most important. And Web3 sequels more important than Python by far for data analytics rules. 

For data engineering, you need to know Python both for like purely data analytics if you're splitting your time 5050 between SQL and Python…just do 100% of your time. And SQL. Just don't even worry about Python. Because dude has figured out all the Python for you. 

Basically basically, you need Python. Like get those Dakota tables, right? But like dudes already done that for you. Just do SQL and like touched Python. We absolutely need to. 

And if you want to do graph data and SAS then you have to cipher, which is like new for JS language. But that's not…I wouldn't worry too much about Graph QL, solidity, JavaScript, and one that's like nice to know. Like, if you have the time to learn then go ahead, but it's not necessary.

Thomas  1:31:55  

Awesome. I think that wraps up the questions. Amazing.

Andrew  1:31:59  

I realized I'm still sharing my screen.

Boxer  1:32:02  

Yeah, that’s the way to go.

Yeah. Anything you want to leave with us? Maybe some…what's your favorite next feature of Dune? Or what would you be really excited about if we just release party data?

Andrew  1:32:20  

API data into these main tables.

Boxer  1:32:29  

Miguel will be very excited for that API product manager.

Andrew  1:32:35  

I'll tell you what I'm doing right. I can't keep doing abstractions. PRs for the date…I want to bring it…

Boxer  1:32:41  

Oh, yeah.

Andrew  1:32:44  

Right now is export from my API, put into a JavaScript file or SQL file, PRP like, Hey, can you add these 50k rows?

Boxer  1:32:55  

That's…yeah…

Andrew  1:32:57  

That's my top request.

Boxer  1:32:59  

Okay, see? Definitely doable. We hope to be able to provide you with service soon.

Andrew  1:33:09  

Other than that…only last thing I'll say is if you're looking to get a data Web3 role, please apply to the talent board. I think it's in the description. It's also my Twitter. But it's just alemi.palette.com/talent.

Boxer  1:33:28  

It's in there. Yeah. I just put a Dune drop in there today as well. So oh, you still need to approve that. 

Andrew  1:33:37  

I need to prove it? Yeah. But yeah. Amazing. Yeah.

Boxer  1:33:43  

Thanks so much for coming on. Andrew. Really appreciate your work across the space. Like really most productive person I know. Yeah, always great talking to you and see you soon. Bye. Bye.

Boxer  0:00  

Welcome to a new episode of the Weekly Wizard. I think this is the 10th instance now. So, there’s a little diversity here. It's great. And today, I have with me Andrew from Mir. 

And if you're in the doxing community, you should be familiar with Andrew. He's pretty much our most prominent and most helpful and just best human life community member. So yeah, very happy to have you on. And yeah, welcome.

Andrew  0:32  

Thank you. Thanks for having me here.

Boxer  0:35  

Yeah. Can you do a brief introduction of yourself, please?

Andrew  0:41  

Yeah, sure. My name is Andrew Hong. I grew up in California and came to New York to run away from home basically. I've been in New York ever since. I started to get the long spiel now or should I get…

Boxer  0:57  

No, no, I'll just think…okay. Yeah.

Andrew  1:02  

Like four years since college, but yeah, data science. I'm here. Basically, it was the first day to hire the team last August. 

Working on everything from basic analytics to engineering to like some science stuff. But whoa, I'm sure we'll get to that later. 

Boxer  1:27  

But a lot of people aren't really starting on data science, but you're doing other stuff as well. No.

Andrew  1:30  

Yeah, okay. Okay. All right. My brain is like scrambled. I'm like sewage. So, I used to basically do just a bunch of research projects on the side until like any anything in the Web3 data space that looks fun to me. 

So, everything from like mental stuff and art blocks auctions. The way I got involved with Mir was doing like social graph based ERC 20 airdrops. Right now, I'm working more on like Neo for J style network graph analysis. 

But we can dig into that later because that's like a whole new rabbit hole that I've basically disappeared into for the last two months. So yeah, and then I went through education stuff. 

That's obviously almost always going on…try to put our content…should have something fun to announce on that in the next week or so. 

Boxer  2:24  

Interesting. Yeah, when we work together briefly on the our network course, I talked to my team internally, and I was like, Andrew is the most productive person in life. 

So, it's certainly…you have your hands in a lot of…or you're participating all throughout the ecosystem. It's always amazing to see what you come up with. So, Andrew, I'm sure you've looked at your portfolio or in the market today. 

And it's not the best. And yeah, we're both data analysts. So, we're not market analysts. That's a stark difference. But what do you think we're left with now? Once everything has gone down? 

Tara has imploded. Apparently, they're scammers in the NFT space, which is like, oh my god, big surprise. Might have guessed. Who would have guessed? So after, I guess, two years of a good life in crypto and feeling very rich now we're back to baseline reality. 

So, what's your like? What are we left with? What have we actually built?

Andrew  3:40  

Okay, what have we built? I mean, what are we left with portfolio wise? Just pain. I haven't even checked my portfolio in the last few days. Building I mean, I think, God, I want to say we've built and learned a lot just because looking at where we're losing money and value. 

It is getting more and more sophisticated. We aren't just making the same exact dumb mistakes anymore. Like that. 

The most recent one well, okay, for the most part we are sometimes…you still do this and see the same reentrancy hacks, but for the most part, we've gotten out of the basic hacks.

But yeah, even with the whole Terra de Pegging that happened, that was some very sophisticated financial markets stuff. 

You know, just rereading into it, it wasn't like…there's by no means simple. There are institutional players involved. I think even though a lot of money was lost, and it basically crashed Bitcoin into…what what is it? 2027? 28? 

Whatever it is now. That's painful financially. It's like okay, it's getting more and more sophisticated. So, I think that's still a pretty strong signal. 

And obviously things we've built in the last two years is insane. No one knew what NFTs were, a year and a half ago. You know, so I think we filled a lot. I'm still bullish. And as long as ETH is above 1000, I'm not going to cry myself to sleep.

Boxer  5:26  

Yeah. Yeah, I think about how many people we've picked up, and especially how many builders and how many smart people, the space has picked up in the last two years. 

And obviously, that is fueled by financial speculation because smart people look for easy ways to make money. They find crypto, and they're like, “Oh, this is actually like a really cool playing field.” 

But then they realize, like, hey, there's actually more to this decentralized casino that we're building. Rather, it does feel like that uncertainties…but rather the underlying infrastructure here. 

That is really what is exciting. And what's interesting here, and I guess a lot of people share that story. So yeah, I'm definitely a lot of failed experiments in the last few years. But also, we've defined and found a lot of valid use cases. I guess, I would you agree with that take?

Andrew  6:31  

Yeah, yeah, I think the use case is like, they're slowly maturing. I feel like the use cases are only really usable by people like us who understand crypto. It’s not really usable for people who are my next door neighbor, probably even just writing on Mir would be hard for them. 

You know? And if even something as simple as writing can be difficult. We don't even need to get into DeFi.

Boxer  7:00  

Yeah, there were a few startups who had encouraged the back end, like easy API. 

Andrew  7:09  

I don't know how they're doing what they're doing. Do you find Mala thesis I think just went out the window for a little bit. Yeah.

Boxer  7:21  

They shouldn't have banked on Anchor, I guess. Yeah. Okay. So mark it down. We will probably be in Goblin mode for quite some time. So why the fuck should anyone still work in Web3, or like you're running this? 

Career boards or job boards or however you call it? Pellets? I think it's in the description below the below the video. 

Andrew  7:57  

So, why should anyone still try to get into a pretty easy answer. It's still early, and I feel like the best place…not even getting into the tech wise, but  the best place to build a career is where there's a ton of unchartered territory, a ton of money, a ton of potential impact. 

And smart people that aren't trying to keep things like you don't have 10 certificates you need to take or some really weird long job application process to get into the space, you know? 

You write one good Twitter thread on data analysis, and I guarantee you'll probably start getting interviews and maybe even offers. That's the barrier to entry. 

It sounds like a low barrier, but it's actually hard for people to learn how to contribute and communicate that way. Because people are so used to like,” Oh, let me go take this Udemy course” and then find a job board and apply, you know. 

But it's like, no, we don't do that here. The opportunity is just huge. And there's so much freedom. 

And that alone is already something I don't think he can really find in any other industry. And then he add all the tech is just like, okay, it's almost a no brainer. For me at least, it's a no brainer. Everything that I build that near I can then use anywhere else, you know, like day one. 

And anyone I want to work with, if I admire a data analyst on another team, like Michael from the optimism team, I can work with him on anything whenever I want. It's not like if I asked him for help. I'm the only one who benefits you know or vice versa. That's huge.

Boxer  9:44  

Yeah, I told him you read obviously. The choir here playing the devil's advocate. So, being on that topic, how did you actually start getting into crypto?

Andrew  10:00  

Yeah. COVID hit in 2020, and I was at my banking job. I did credit analysis back back in the old days. Obviously, none of my tech worked because they were trying to figure out remote login and whatnot. 

So, I had a bunch of free time. And I was like, alright, I can either sleep and watch Netflix or try to learn a new skill. And of course, I slept and watch Netflix for probably a month before I was like, okay, it's April/May this does not look like it's gonna just end. I should probably do something more productive. 

That was when we thought like, “Oh, by June, we're going to be out,” you know. But yeah, I had so much free time. I started looking into crypto because I saw someone talk about a unit swap. And I was like, okay, this is something that would affect my job and finance a lot. 

If it becomes big, I should probably at least have a basic understanding of it. And this was when DeFi pulse was still under like 2 million or something. And like AUM and whatnot.

Boxer  11:13  

If I pull it was still a thing, like at all.

Andrew  11:17  

Like DeFi pulls, screenshots every week, look another 300k. But yeah, I was like just dabbling on the side reading a bunch of these white papers. Honestly, none of it technically made sense to me. 

It took a while before the technical stuff kind of clicked. And thanks to Linda, she’s on Twitter. Great person. She put a Tweet where she was like, “Hey, anyone who's like trying to become like an ETH dev in some sort of way, I'm gonna hold this like Zoom session.” It was on some XYZ date. I forgot the date. 

And just like that, I was filling the Google form out. I signed up. So, this was like July or something of 2020. So, I filled out that form. 

And it was Linda Trent Pinups, Austin Griffith, and then like two or three others in the space, and they just talked to like 30 of us and walked us through the full like development stack back then. It was like react, andwe'd layer before it became hard hat. 

And like slitted, solidity for something with the 5 or 0.4. And talking to them helped me a lot and understanding the technical side of the space because otherwise, I would never have found those tools because Google Search sucks for trying to search or develop anything for crypto. You will not find what you need. 

Like, I don't know why it's that bad. It’s likehaving that workshop taught me like, oh, these are the tools you need to know. And then Trent was like, “Hey, you should do ETH global hackathon.”

And it was like a 30 day hackathon. When I was like, alright, nothing else really to do. So for those 30 days after work, I probably spent like fiveor six hours every day just like coding and learning solidity and Java scripts. 

And after that month, I was like, okay, at all. It all makes sense at a basic level. But yeah, that's basically how I went from zero to 100. Yeah. Big shout out to the bank was guys as well. 

That was what I was reading to learn about Uniswap and whatnot. So, on like a media side of things. That's interesting.

Boxer  13:44  

So, 2020 was the first time you actually heard of crypto or you did any? I mean, not heard, but like, actually…not like Bitcoin mining in your dorm rooms or anything like that.

Andrew  13:59  

No, I was vanilla as hell. In college, old idea, just played ping pong. Did some like basic quant trading? And like, okay,

Boxer  14:13  

Yeah. What's your background? Like university wise?

Andrew  14:18  

Oh, I did econ and math at NYU.

Boxer  14:21  

So, you're not even a developer person or anything like that. You just now started developing in Web3. And that's like your first…

Andrew  14:31  

So, I started developing when I was at the…in my banking rule because everything in banking is PDFs and excel sheets. And I hated working in Excel and PDFs. 

So, I learned Python because Python you can have image CSV, like read the PDF put into tables and automate your Excel and PowerPoint. I literally wrote scripts to take the PDF of like the 10k. 

Put it into Excel with the spread and then screenshot the spreads into PowerPoint. And that used to take people like a day to do. And I just automated it. So, I could one click do in 15 seconds, and I'd have the rest of my day to study. 

And I can say this now.It would take me all day even with automating the emails. The email goes on at like 4pm. But that's how I learned to…

Boxer  15:30  

Been there done that. Yeah, if you are slave to the programs you are you supposed to be using that to make the program's work? I think slave is a hard word. Servant. I think that's better. 

I don't know. I apologize for foul language if anyone takes offense. So yeah, um, you just dove right in and then now you work in Mirror. So, what was the progression from being like, I know some of this stuff to actually getting a full-time position somewhere.

Andrew  16:06  

Yeah. So, I think everyone's journey that they went through is somewhat messy. Because one, the technical stuff is hard, and the startups make almost no sense. The fact that there's no product market fit in anything, you know, even back then it's not very known was great. 

So, it's literally…I knew I wanted to join a crypto startup, but there is no way for me to tell which of these DeFi protocols are going to be around in like, five months. There's literally no way of knowing, you know, so it was like really hard for me to figure out like, okay, I know the basic technicals. 

But there's no job place for me to apply to. I don't really want to join a VC sided period where I was just basically freelancing. So, I did some solidity freelance work with spectral finance as a start. And then I ended up joining consensus for five months or so full-time from I think it was March 21 to like August 2021.

I joined consensus as a business analyst because I was like, oh, maybe enterprise consulting will be fun. It wasn't my cup of tea. A lot of great people. There's a ton of smart people.

‍

It basically helped me understand the space better, and be like, oh, here's the history of the space and how these types of startups got to where they are today. And here are the key people that you should try and meet at the startups. 

So that really helped. I ended up being like, okay, I don't want to do solidity development because optimization and security is not my gig. 

And I'm like there's gonna be people who are 1000 times better than this, even if like solidity. devs get paid like three, four, or 500k. I'm just like, that's just not. It's not my thing. 

So, I went back to my data roots and started posting more about doing stuff. In April, I want to say of last year, I think that's when we first interacted. I was asking you a bunch of questions on what are these tables? 

How come none of my queries are running? Why is the timing, and you helped me…here's the basic sources and all that. 

And I think as soon as I started tweeting…the stuff I was doing in June…I started blowing up on Twitter more. Once you're like…start building a following on Twitter. 

It's just become such a strong cycle of meeting more involved, smarter people opportunities, and then it keeps building until you've hit a point of like, okay, I can't keep doing 10 side projects. I need to find the main commitment. 

And gram CTO of Mirror DM’d me and was like, “Hey, do you want to do this freelance project for us for like a right token, AirDrop?” and I was like, okay, I'm tired of DeFi. NFTs are starting to take off. 

And  I do want to try to learn how to work in a consumer app. And the freelancer project went well. I loved working with the Mirror team. I think they liked working with me. And then we just decided, alright, let's just make this a full-time gig. 

Boxer  19:35  

Yeah, cool, long-winded down. The journey has a lot of arcs. I didn't know you were at Consensus. That's very surprising to me. Yeah. That's probably the only guy on Twitter who was at Consensus and doesn't have X consensus.

Andrew  20:00  

I think I used to, and then I ran out of character space. And it was excellent talent board. 

Boxer  20:14  

So, I guess…what drove you forward or what really helped? While you were learning how to Web3…personal connections…you mentioned this one Zoom call, which seems to have had some transformational value to you. 

So, in general, how did you approach trying to learn as much as possible, like learn in an effective manner in this very convoluted space?

Andrew  20:41  

Yeah, um, I think it was, like once I learned the importance…this is gonna sound so cliche…but once I learned the importance of community…and Web3 reshaped my whole approach. 

Because I realized what phase not a space where you just study and work on your own and grow that way. Like you can do that. 

But it's gonna be very, very slow. Because even for data, for example, the amount of data content that's out there, even now, very, very little, like extremely little, you know. 

So, if you try to study that on your own, and just try to figure out through doing queries yourself without talking to anyone in the community, you're in for a world of pain, you know? 

It's like, okay, everything's based off communities, there are key players that move across all these communities. How do you learn to track that pulse of people moving and new problems and communities coming up? 

So that you know what to learn next? It’s kind of a way that I thought about it…like I wrote a…hold on…I'll basically the mainland head versus like talk to people. And like ask them who…

Boxer  21:51  

What do you mean by key players to? Instead like an actual person?

Andrew  21:55  

Yeah, like actual people? Yeah. Um, so I like people like Austin Griffith, for example, like looking at the stuff he's building, and the people that he's talking with or looking at eat global and seeing who are they partnering with? 

Who are these two sponsoring all the time, you know, and you're gonna start seeing on all these live streams and whatnot, and then this course, you're gonna start seeing some of the same usernames keep popping up? 

You know, or like Patrick Collins, you know, you're gonna see the same names.

Boxer  22:25  

Yeah, I see. I see what you said. Okay, like the last DeFi summer, the end of 2020. There was a bunch of these chords where I used to run into the same people all the time. 

No, nowadays, not so much just because we have so many people now. But back then it was like, every Discord server was a few hundred people…like the probability of running into someone, you know? 

Yeah, I don't know. But that's very interesting. So, you were trying to uncover these hidden networks and follow the trail and see what these people are up to and follow them.

Andrew  23:06  

Yeah, my biggest insight for me is then just writing about what you learn, and putting it out there, so every time I learned a new framework or thought around something, I would write about it, and put it out there. 

So, when I was studying wallets, I would study the full history of wallets up to looking at all of the GitHub commits of Metamask. And seeing like, oh, this is how walls were built up. 

And then I'd write an article about composability of wallets. So, one I have notes to look back at, into what I share. It's like, oh, that's gonna get me…that caught me a conversation with Dan Finley from Metamask. 

And we had a call, you know, andthat would not have happened if I hadn't put the article out there. You know, if I just done research and tried to reach out to him, it would have been a lot harder. Yeah. 

But the main thing… I wrote…I don't know how to put this. I put it in our private chat. But this is back when I used to read on Medium. I wrote this thing. I'm going to share my screen.

Boxer  24:03  

Yeah, Thomas in the background can put this in the actual chat.

Andrew  24:08  

I wrote this thing. I'm networking in crypto. And essentially, you have to try to understand how to navigate content, conversations, and communities. 

And it's basically, what I realized is everything is fractal in Web3. Because in Web3, if you have a new problem come up, right, whether that's in math or DeFi or whatnot. It’s really…it's a decision tree. 

It's just a subset of an existing problem. Right? And someone from an existing community is going to probably go and lead and build the new community. They’re trying to solve that problem. 

Right? So, it's all connected and all these people are moving around these communities, right, but it's all fractal. 

You know and this is different from traditional industries where it's like, oh, these fractal communities are happening, but they're only happening within Google or only happening within Facebook. 

And that doesn't go across industry, but it went through…every single topic is fractal and connected. And once you learn to see it that way, then you can start looking through like, okay, what are the layers of things you have to understand…the problems and communities around tokenomics around memes protocol, governance, and now there's a lot more back then it was much easier. 

But it's basically like I have a mental map like this of different meetings. And if I see someone move from one to another, I'll be like, oh, there must be a link that I'm missing between these communities. 

Otherwise, why would they suddenly be reading about this or studying this? Right? And then I'd either go and ask them or just keep an eye on the data and be like, is there something happening?

Boxer  25:49  

So basically, a third dimension to this picture that you just showed, and like, it turns out, they're really close in another dimension.

Andrew  25:58  

It's really fun to think about it that way. Because then it's like…all these problems that look like splitting branches suddenly connect. There's all these Web3 moments where you’re just like, oh, shit, lmind blown. Oh, that can be connected that way…that changes everything about how I thought about this, you know? 

And yeah, but you have to have a basic framework. Otherwise, you're just gonna get so lost, trying to read this on Twitter and be like, oh, what's trending? What's the most important like…? And it was?

Boxer  26:33  

What does your basic framework even look like?

Andrew  26:37  

For me, it’s all notion notes like…

Boxer  26:40  

No, no, like, more on a conceptual level? Like, are you only trying to follow data people nowadays or only? 

I don't know…you're probably like in this Mirror and FTE writing bubble, and then the data bubble, and then probably some other bubbles that I didn't even know about. But how did…because you're saying…you can't just consume everything…but rather you need to focus on what you're doing. 

So, how do you decide that? And how does that inform your information diet as well?

Andrew  27:11  

Yeah, yeah, definitely. Um, I normally try and pick just one topic at a time to follow. So, right now, it's social graph analysis, right. And that's going to be the only thing that I pay attention to feed wise for the next month or so. 

There's gonna be certain key people that I follow. I'll still pay attention to the stuff they post, but I try not to get distracted by…even if I see something cool in the med space, I'm like…okay, that's cool. 

But it doesn't connect to what I'm trying to learn right now. I'm not going to waste my brain time on this stuff. 

I don't know if that makes sense. But it's like very much you follow a decision tree of like, okay, what affects what I'm working on right now. What's gonna be big in like three or four months, and what actually has enough content to read about and you find that Circle? It's like, okay, this is the thing.

Boxer  28:13  

Yeah, I see. I see. So very much the focus approach, like that's this tariff stuff.

Andrew  28:18  

I'll read whatever thread goes in the telegram or slack, but I'm not going to… because I feel like that's gonna get wet through data burnout. 

And we've all been through this. It's just like, oh, every new thing that comes out. I'm gonna try to write queries and do a dashboard on it. 

You know, and you could do that for two months, but you're gonna be so burned out and tired. That like, they're not going to have any output for another few months.

Boxer  28:48  

They'll be…would like to have a word with you. Yeah, he's been doing this for so long now. He’s crushing it. Yeah, I guess he doesn't have any other job. 

Right. So, this is his full-time job to to do this. And meanwhile, you also building on Mirror, and you have responsibilities there. So, I guess it makes it easier if you were…I wouldn't.

Andrew  29:19  

I wouldn't. Even if I was doing full-time I would burn out by now.

Boxer  29:24  

Okay, fair enough.

Yeah, your journey is one of a kind. It's pretty interesting that you go from being I don't know…like very non-technical and then being like one of the…I don't know….thought leaders in the Web3 data space or something like that in a year or something. It's pretty amazing.

Andrew  29:50  

Only in crypto. You can't do that in other industries. No, yeah.

Boxer  29:53  

Oh, one thing that I forgot to bring up…um, this whole fractal thing that you were talking about. Yeah. All of that is enabled by all of us being on the same infrastructure, right? Is that your take on this? Yep. Yeah. Okay.

Andrew  30:12  

I think that's the only reason it's all self similar connected. Whatever someone does on one end of the network affects everyone else and all the infrastructure.

Boxer  30:26  

A lot of butterfly effects happening there. Yeah. Very interesting. So, we've talked a lot about neuro, but I know what Mir does, but I don't really know what the role does. So, the elevator pitch you go. 

Andrew  30:44  

Yes, I think I need to give some beer history first. Obviously, okay, the people here we're all. I think we're all very smart. We love to build very enthusiastically for all the fairly easily distracted. So, I think last year was like, oh, you started as a writing platform in 2019 or 2020. 

And then, Daluz and NFTs, both started becoming popular at the same time. And it became like, oh, how many different ways can we do crowd funds and doubt tooling? 

Because that's where all the attention and money is at, you know, so we ended up building, so many different five or six different versions of crowd funds. Like we were like Dutch auctions, we built splits as useful. 

So, this is more focused, but when basic auctions had like reservable editions and whatnot. And it was just 10 or 12 different token races…well…iit was like 10 to 12 different products all in one. 

And it got to the point where it's like we have to spread out, and we need to refocus again, you know. The the last four or five months or so have just been like… 

Alright, go back to the core of writing NFTs, build on L2s, and really build like a Dow writing flywheel where you have a dowel that we already have a lot of the best writers on Web3 publishing on here, right. 

But there needs to be a strong economic flywheel that needs to get better… like stronger collector, flywheel, and better reason for the doubt to exist, and ways for them…like writers to participate in the first place, you know. 

So that's the focus of a lot of what you'll see comes out over the summer. We've already planned it, and it's ton of QA and fixing and removing things that we're like, alright, we don't need to do build just for Dows and their tools anymore. 

Because if you think about it, Dows are a big opportunity, but the number of dials that are actively using all over the lot of these downswing platforms, in like flexible ways, and more than just connect to discord. 

It's very…maybe 20 bigger Dows…yeah not just like for once I have like over 1,000,002 million at least…and they are actively doing more than just one NFT sale. Yeah.

Boxer  33:20  

Yeah, I see what you're saying. So, there's so much dopamine but not a lot of active significant dollars to get around, basically.

Andrew  33:29  

Yeah because a lot of downs…they have an initiation phase, which is really strong, but they don't have…besides something like noun style. 

They don't have like sorry, I live next to it. They don't have some economic token flywheel where they have figured out like, oh, NFT issuance membership tie in and connect to next season or whatnot, you know. It’s so early, and that's still going to take a while. 

So, alright, let's just build for writers, which makes sense. And there is a flywheel for…versus just building for dowels where it's one. There aren't that many into, like, if you're trying to build for super rare versus like Blackhand or something, good luck. They're gonna ask for two very different things, you know? 

Boxer  34:21  

I see. So, maybe let's take a step back. So, you guys have built a writing on which anyone can host their own publishing or writings and mint that thing as an NFT to whoever supports this piece of writing that correct assessment?

Andrew  34:45  

Yeah, yeah. So okay, given the elevator pitch, right. 

Boxer  34:49  

Can we skip that phase?

Andrew  34:53  

Given planation like, okay, you think about economic models. We have our writers today. Need em is pay-per-view, right? They literally…you get paid more if someone subscribes to media while reading an article, which is completely random, you know? That's the only way you get a larger paycheck. 

Otherwise, it's pretty arbitrary pay-per-view. They started doing bounties where it's like, oh, if you're one of our top 100 writers of this month, we'll give you 50 bucks. You know, I think I was in their top 1000, and they gave me 50 bucks. 

And I was like, okay, q sub sack is like, you know, pay-per-subscription subscriber, right? Which is…it's better. But it's still homogenous, right? Every subscriber is the same. 

You could have a Patreon-style tiered-subscription. But there's not really much way of customizing or showing how much you like someone's writing with those traditional models, right? 

And I'm not even talking about medium pay-per-view. But when you have the NFT is one, it's like, you could still use it in a subscription style method of like, “Hey, if you buy an NFT, then you're subscribed to me for the month, you still have that.” 

But it's also like, it gives the collector and the author a lot more community. Personalization saying, hey, I want to find someone who really identified with that article, you know? Or someone who's really committed to…it's like, I don't…okay, the elevator pitch is getting to walk. 

But basically, it's like, you're giving them a way to like, subscribers to distinguish themselves from other subscribers, right? And building a more personal relationship with the author. You're giving the author a much better way of monetizing, I think, than just like a flat fee because if every subscriber is homogenous, you can only do really a flat fee. Right. 

But if you know, hey, I have 1000 subscribers, but I have 10 huge fans, there's something I can do with a pricing model specifically for the fans that are willing to pay like $500 per article, you know, and it's like that gets your writers mind thinking like, oh, maybe there's more I can do with this. 

And then you add that to a dowel of like, Oh, if I'm a great author, I can be part of the Mirror doubt of it with a ton of other great writers. 

Like, there's also a lot you can do with that as well. And like that stuff that can't be done what to and so that's where mirror is trying to focus on, so work on whittling that down.

Boxer  37:29  

So, that was a good pitch, please send me the deck after the…get you in the round.

On the one hand, you're building this platform, which is open for anyone. And then on the other hand, you also try to facilitate a writer dowel, which is about producing world class content. 

And I guess it's Web3 expertise, or are people writing about like…I don't know…how to make the best honey in Alabama or something as well.

Andrew  38:00  

Great question. I think right now it's one through expertise. Obviously, we would love to have someone writing about the best honey in Alabama. But, I think that's just one of the technical barriers of what…

So, right now, just having the top with three writers figuring out an economic model for them and a membership and collector model that works for users. And wrapping that into a towel. 

Because it's how you like…that's one of the strongest competitive points, right? It's not really just your product. Anyone can afford your product…like Mir is not that hard to fork. 

Like someone's literally wrote…I forgot who wrote the code for Kashmir as like a dev tutorial, you know, because all of our design problems…yeah all of our design objects for what it's called that sits as a solid NPM package…

So, you could use and rebuild near at least the basic surfaces in a day, you know, but if you have a strong down…a strong community….that's what keeps people there. Yeah. So, that's…

Boxer  39:12  

Yeah, and also for Mirror, I think it's a lot of brand awareness as well. You guys have built a very, very strong brand. So if somebody now…yeah, I don't know. I would much rather click on a neuro link than a medium link already. 

And then if somebody just faux Mirror and call to…what do I know? Then like, I'd be very confused about that link even. Yeah, definitely besides the point. Yeah, there's definitely some defensibility there. 

Um, but so what do people do in this writer doll? Do they change thoughts like revenue each other's pieces work together to make a bigger impact? What's the model there?

Andrew  39:55  

Yeah, well, okay, right now, it's just focused on curating other writers that's literally the one use case of the Dow right now. It's like every week, members will get together and be like, hey, what's some of the best writing we've seen on near each week and Rafa, the builder on Twitter, leads our Dow. 

And he really just like spearheads talking with members, talking with the team, finding the best writers, and getting them on Tamir. Patrick helps with that as well. So, that's like the main focus of the doubt because it's like we got to get a ton of great writers first and build up the Treasury. 

And then we can start funding a bunch of fun initiatives and bounty programs and leaderboard incentives and all of that…or even just focused figuring out what we want to do with the governance token, you know? 

All right now, it's literally just like keep a very focused one thing…your only proposals are, which writers do you want to see to join the Dow? You know, and that's it.

Boxer  41:06  

Okay. Yeah. But sometimes simplicity is the beauty of things, especially in this very noisy world of Web3. 

It's sometimes good to really hunt down and be like, you can give the dashboard the star or not the star, nothing more. You can or not…that's it.

Andrew  41:29  

Yeah, just keep it simple. But that's the key. That's the key.

Boxer  41:34  

So, can you share a bit of the data that you've been surely pulling about neural and doing?

Andrew  41:41  

Yeah. Ah, okay. Let's see, how do I want to start this? I'll start with the concrete before it just goes into fully abstract. I mean, this is just like a basic NFT dashboard. I don't know how much to talk about. 

This was basically your typical NFT like show the mint stuff. Okay, let me just go maybe one by one. We're putting more and more metadata on to on-chain, so it's a lot easier to be like, alright, let's say you were just looking on Dune, and this would link out from a leaderboard dashboard which shows which NFTs are trending right now. 

You can click here and see the name, the description, and total supply and mid-status with…this would show how many NFTs are left. And since it's closed, you can just click on it. It takes you to quixotic shout out to Michael for figuring out the HRF stuff there. 

It shows the basics on purchases, and it’s raised with a bot removal column because optimism bots to things. You have to give a scale if I don't remove bots. This number is in the hundreds of thousands.

Boxer  43:02  

I was gonna say like this optimism NFT was super, super successful.

Andrew  43:07  

Yeah, they're like five…I think it even times out to like 500,000. Okay. Everything has to have a no boss filter.

Boxer  43:17  

Yeah, I think how do you do that? Or just Michael’s?

Andrew  43:23  

So, the way I do this is just like looking at people who only call the contract directly. Because a lot of the bots use a self destruct contract basically to mint things. 

It’s like, okay, the people are our boss. It shouldn't be the people who met through our front end and call the contract directly. It's not the best, but it works for now.

Boxer  43:46  

Yeah, that's actually…so, there's 480,000 transactions made by bots on optimism just for this mirror NFT.

Andrew  43:56  

So, they batched it in this contract, so they can make 10,000 at a time and one transaction. It was whatever 480,000 divided by two…like 500 transactions.

Boxer  44:11  

Wait, so the bots meant to 10,000 NFTs at once?

Andrew  44:15  

Per transaction so they have a contracts that batch minutes and then self destructs when it's done. So you get refunded the gas costs for like deploying the contract and whatnot. And it's actually pretty advanced. It took me a while to look through it.

Boxer  44:31  

Yeah, that sounds…I mean, yeah, good on them for figuring that out. But I mean, Michaels working optimism…like have you seen your adversary? Don't you think Michaels got to notice?

Andrew  44:44  

Yeah, we're talking about like, is there anything we can do about this? Anyways, that's still discussion. Okay. We'll keep that private. I have some wallet history stuff here. 

Since I think for every mint, it's nice to know who's transacting, so in the media, wallet has been on optimism for 52 days. This is how much gas they have spent. Most of them have spent…they've had most of them have only had eight transactions. 

And they only have like 0.00 to 5 ETH in their wallet, which is not much on main net. But that's enough to buy like 500 writing and FTS on…yeah. So, at first, I was disappointed, but I thought about…and I was like, you know, that's good. 

That's a good amount of play money. Classic distributions, and then this is an author leaderboard. But it's not going to make sense until the more entries tries to use you in a deeper dive. The data exploration dashboard for people…because obviously, we're gonna have this data in like our front end. 

But it's like, if someone just wanted to explore things quickly, they should be able to click around dashboards. And the secondary market, I think this is fairly standard across NFT dashboards, just highest price, lowest price volume, like getting their hold or distribution of like 5% of people hold 47 tokens. 

27% of people hold one token, you know. It's just like being able to quickly look at this stuff and be like, alright, how many people are at each token range? And then trading protocols, this is probably something I'm going to work on a lot more. 

But like looking at your top traders and just being like, hey, how many tokens have they traded? How many did they meant? How long have they been part of the ecosystem like part of this collection? Sold? Purchased? Bought? 

Pretty basic stuff. But it's good enough for you to be like, alright, let me put in any address and just see what's going on. I feel like this is like table stakes…went through data analysis now. I think it did take us a while to get to stuff like this. 

Boxer  47:09  

I feel like, yeah, we're growing together. We're growing together. So, this is the optimism data. Yeah. Do you have any idea how many Mirror NFTs got mentored on? In total, I guess you're on main net Polygon as well. Or only main? 

Andrew  47:28  

No, not on Polygon? Let's see. Mention it on Main net? That's a good question. I actually don't know if I track that. I only track ETH raised from all of your side hustles. 

Boxer  47:43  

Exactly. Yeah. Yeah, I actually don't know. It's interesting. How many, at least how many unique writers know that it actually doesn't get transmitted on-chain? Right?

Andrew  48:03  

I could check for unique. Yeah, I could do unique writers. This is probably pretty close. I'd say 9700 or so have probably minted. And then their IDs are empty. I guess cuz we looked at this we're like, okay, let's just recommit and do this on optimism. Yeah. Yeah. This is like the front facing data. I can dive into my big data projects, if we want.

Boxer  48:32  

Your social media or social graph stuff. Yeah, maybe real quick. 

What made you guys go to Optimism? The only two, if you can talk about it, obviously. I know.

Andrew  48:49  

Yeah, I can definitely talk about that. I think a lot of it was just timing of like, we've been talking to a bunch of LTE teams. We wanted to have a chain where we could both implement NFTs and also write race and a bunch of our other products on. 

We just found through a bunch of conversations, Optimism was a good partner to work with at the time and worked with our existing infrastructure on Alchemy because a lot of other chains like Alchemy don’t fully support…they're still working in beta, so it's really just like…everything came together. 

It was like okay, let's just do this on optimism for now. Sorry, my dog is having a fit. And yeah, it's gonna be the only chain okay because I think it's like…it's not a DeFi product where it's like, oh, I can have each pool on its own chain, and they don't affect each other. 

But writing NFTs, I feel like it gets our consumer products in general.I feel like it gets confusing if one writers running on Polygon and another ones writing on optimism. 

And if I tried to do some composability thing or trading thing across my collection, like it just feels like it can get confusing really quick. 

Like, oh, which NFT sits where? And like, what if a writer minted on Polygon, but now they want a mental optimism because there's no more volume on optimism? Do I have the same NFT on multiple chains like do they even have funds on?

Boxer  50:23  

Do those chains like? Yeah, very interesting problem space. I guess that could be solved if the user interfaces. 

So, the wallets solve this in a better manner. So, you don't even notice on which shengyuan. But for the time being, this certainly feels like the the way to approach this.

Andrew  50:45  

100%. Yeah. It's just less dev overhead. As I'm sure you all know, adding each additional chain, it's not as simple as changing chain ID.

Boxer  51:00  

Just call the nodes, man, no problem. No problem.

Andrew  51:04  

Just add a new node.

Boxer  51:07  

Yeah. Interesting. Yeah. And maybe before we…Yeah, I don't know, if you want to dive into your social graph data, I'm certainly very interested in what you're what you're picking up.

Andrew  51:19  

I can talk about from like a theoretical level because I think like this, this is like one of the things that's kept me very interested in the space from a data perspective. Identity has always been super interesting in Web3, right? 

That's the whole reason I did the social graph shop and whatnot in the first place. Because you have your wallet with all of your transactions, all of your NFTs and tokens, right? 

And there's basic identity building blocks. You can build off of that just like, okay, who bought what NFT? 

And then you have layers of context on top of it, which are like your different communities. Which is more than just saying like, “Oh, is this NFT from our blocks or not?” It's saying, “Oh, if you bought a fidenza, who else bought a fidenza?” with you. 

And that's your community. Right? So that's how you start to build this social graph of interactions. And basically, I had a split into just trees. 

Last time, not last time…like the last two years basically I spent thinking about a decision tree of like, did you do this? How long did you do it? For? What community? And who else did this with you? 

That was the basic side of it. And I kind of…I don't know why it took me so long to realize this. But about like two months ago, I started looking at this and being like, okay, if you were to build a recommender algorithm, right? 

For NFT content, the same way that you build it for like Netflix, could you just build it off of like a matrix of users? And are they buying the same NFTs? Or could you just do it that way? 

And the more I looked into it, the more I realized, okay, a big difference in Web3, is the ownership perspective, right? It’s like if I watch something on Netflix, or if I don't watch on Netflix, it doesn't really affects me or other users, you know? Like, whether or not I watched…

Boxer  53:34  

Yeah, just consumption, like it doesn't change. And the same flip stays switched off…

Andrew  53:41  

The same thing on Twitter or anywhere else. If I like something, it doesn't really…I didn't hurt anyone. And I didn't spend anything to get it, right? But one three is ownership base, which means everything now has game theory attached to it, right? 

Which means if I just…if you just do a basic recommender algorithm with your matrix ceases ones and zeros. That doesn't do as good job of…they might help predict like oh, what stuff do I want to read on here? 

But it doesn't help predict what stuff do I want to collect on you, right? Because the collection aspect is now dependent on what the supply is. 

And if I've already been collecting with this group of people, I have to look at did those like…I, as the data analyst analyzing the character, have to look at that same group of people collecting the NFT. Because your social graph, the game theory aspect of it now becomes a predictor of whether or not someone's going to collect that NFT or not…not just do I like that content? 

Ultimately, you want to drive views and the more views you drive, the more collectors you're probably going to get, but  there's probably a better social graph algorithm that you can build to specifically get people to collect. So, you have to figure out, from a network graph perspective, okay, this writers put out 10 articles, who are the people collecting those articles? 

How are they related? And what is the tipping point that causes them to collect the article? You know, maybe it's these two key people in the group have to collect, and then these other five nodes are going to collect, or maybe it's these two people collect. 

And there are four people over here who always just buy from the secondary market later, like with a two-month or three-month delay, you know? Once you find these relationships, you can start building. It's basically a random graph algorithm on top of a game theory algorithm. 

And obviously, in the interface, all you see is XYZ article and ABC collectors, right? And it looks really simple. Yeah. 

Behind the scenes, it's like we've carefully chosen what NFTs are being shown and which collectors are being shown, you know, because we know the social graph of what's affecting you to collect, so you start building these community-based identities.

Boxer  56:13  

Alright, so that's already in production, right?

Andrew  56:16  

No, no, God, no. I was gonna say I'm painting the image of like, why?

Boxer  56:24  

No, that makes so much sense. And then also, it’s based on whoever's looking at it at the moment, showing you want to make it more common? Yes, yeah. And then you're gonna need a lot of compute power.

Andrew  56:38  

Yeah, I'm working on that now…on like, looking at existing collections, but like no NFT collection or no NFT creator has put out things consistently enough. A high-enough quantity for this to be analyzable. 

So, a lot of this stuff is theoretical that I'm looking at now. But it hit me with, oh, you have to have, obviously, your typical reference matrix. You have to have the random graph, but you really have to capture the game theory element. 

And there's not many real world examples of this, like the best examples, international economies, you know, where like, all of your nodes are these different countries, and you're trading commodities, right? 

And these commodities come out and editions of XYZ at a time. And you might like certain creators more than others, right? 

And that's what I'm reading to understand, like how to do this, because it doesn't exist in traditional tech. Like there's no reason for Facebook. So, apply game theory to what's led me to the show…

Boxer  57:43  

Isn’t there kind of a “Couldn't you also look at the VC space?” where it's like if one of the VCs goes into round, two other nodes will most likely follow? I think that immediately comes to mind.

Andrew  57:57  

Yeah, that's a great one. I hadn't thought of this and maybe make a lot of money.

Boxer  58:05  

I don't know about that. But if you want to do further reading, that sounds like similar dynamics would happen there. So, you very much think of this consumption aspect and this ownership aspect, in very, very, very different terms. 

So, the consumption algorithm wouldn't be the same as the actual ownership. I run. And that's very profound. I just realized this, and I'm like, holy shit. Yeah.

Andrew  58:38  

I was like, it makes it's obvious once.

Boxer  58:40  

You said…yeah, yeah. People should recommend things they want to own because that's something that I want to associate with. And like that doesn't get to build my identity now.

Yeah, exactly. Yeah. And what happens in Web2 social media today is that we get recommended content that kind of like…upsets us or keeps us on the platform longer. 

You're not what would be the ideal for me to… I don't know be a better person. If you really want to take it. Like, I didn't know, it sounds pretty absurd. But the content that you consume, basically programs you and kind of…oh man…

Andrew  59:23  

Yeah, exactly. So, I want to do that for writing, where the stuff you read is built off building your identity, not just off of what's going to get you to stay on the page, you know? Because we're not paying-per-view. We're not going to have ads, so we don't have to optimize for that, right? 

We can purely optimize for whatever helps the collective the most. Own the most writing NFTs like that helps everyone else in the ecosystem, you know, and think about the best part of this All this data is open. 

Once it's set, you can show this to the authors, and the authors can look at their community and explore this live, you know? It's all on-chain probably…we can build this into it like…

Boxer  1:00:13  

You can very easily then…I talk, I don't know, like traditional book authors. They have some kind of idea which other authors are being asked by their readers but not directly and…

Can they even contact them? There's probably some telephone number of whatever. But it's very hard to actually. And in with three, you can actually…like there's a like mirror profile attached where it's like the Twitter name is Vijay, and you can just talk to the guy.

Andrew  1:00:38  

Now again, that's my identity ecosystem. Salting service off of that…amazing. But the algorithm has to be written first. And that's if anyone watching this wants to help. Please DM me on Twitter. Because I'm learning from scratch.

Boxer  1:00:58  

Oh my God. That certainly is very profound.

Andrew  1:01:04  

That's what's been getting out of the bed for the last two months. So, I need to figure it out.

Boxer  1:01:11  

Okay, I see. So yeah, I guess, coming from that, what is the long-term game from for Mirror? 

Do you guys see yourself as like? Do you want to replace Medium? Or do you think we can take this even farther?

Andrew  1:01:29  

I mean, Table Stakes is replacing Medium. Table Stakes like base? That's the very minimum? Oh, yeah.

Boxer  1:01:38  

Okay, let's see. Sorry.

Andrew  1:01:41  

I'm using corporate speak. Bare minimum would be that I mean…basically, the goal is to build a strong dow of writers and an economic moat-like flywheel that works for all kinds of writers. You know, not just the big ones. 

And then, sub goals, building something where collectors can really showcase and build their reading identities. But that's very abstract and unexplored. If I've bought like 20 read and like 20 Writing NFT is that the best way of showcasing that?

Boxer  1:02:23  

Yeah. I also kind of…who looks at it? Yeah, good question. When Tinder integration or like what?

Andrew  1:02:32  

Exactly? Yeah. I don't know. But that's also more of a question to the design team than to the data team. My brain is not optimized for that. Yeah.

Boxer  1:02:41  

Um, I can certainly see a world where, I don't know, other people have a book shelf in the background, and they prominently place the books that they're very proud of that are like their ideology or that have really shaped their thinking. 

And yeah, transferring that into the eye, that actually makes a ton of sense. And then you can think about it. So, it doesn't make sense for me personally. I don't…I can. It's a bit weird to have these bookshelves in the background. 

But yeah, I can certainly see how other people would enjoy that. Very, very. Yeah. Eventually, I could basically host. There's Kindle self-publishing today. And then, I could just drop her PDF book on neural, I guess.

Andrew  1:03:41  

Yeah, he could. I mean, you could do that now. Technically. Yeah. But I mean, the hope is, if you're doing something like that, you can find a much better model for it on Mir. Because right now, you do no matter what. Like content on Mir, it's all the same, right? 

And that's not really ideal. So, the hope is like, oh, you find some customizable model that works for different levels of content. But that's like we can't really think of that stuff. 

It's more of like, oh, some writer is going to be writing on Mir for a while and then have some idea. Then talk to us, and we'll be like, okay, that makes sense. Yeah.

Boxer  1:04:30  

Yeah, certainly a lot of ideas to be explored there. There's not just…here's some text different ways of having written content. Yeah, certainly. Yeah. 

And I really enjoy Mirror and love what you guys have been doing. All the side tracks have also been very interesting to observe. What's maybe the most interesting project that got funded through…one of these sales or however…

Andrew  1:05:03  

Oh? Oh, most interesting.

Boxer  1:05:08  

Most successful I don't know…like what the team's favorite is city, Dow, where they bought like… real world plans.

They actually did that. Yeah. So they're like…what was it called Delaware? No. There is some Dow which I don't cooperate in the US somehow. Is that how they…?

Andrew  1:05:30  

Yeah, they did like a crowd fund where you could buy parcels of that land, basically. So, it's like a land sale but not in the metaverse? 

Probably going to…probably it's been long enough other deed was a while ago. That's not going to hurt people's souls. But they can actually do stuff out there land. Yeah.

Boxer  1:05:56  

So, there are no NFTS out there minted via Mirror, who verifies the ownership of a certain real world land asset? Yeah. Okay. How much was it? Like one parcel of land?

Andrew  1:06:10  

Ah, I'm looking now. I don't remember.

Boxer  1:06:14  

But in the order of like, 1000s…10…1000s…100…1000s.

Andrew  1:06:20  

I will have the answer soon. Okay, they have a slightly complicated…because they did a crowd fund first. And then they have like a membership thing for who can like you had to be part of the crowd fund to participate. Okay, I see. And, anyways, that's like one of the favorites.

Boxer  1:06:54  

Okay. Yeah, that sounds sounds very interesting.

Andrew  1:06:58  

One of the membership tokens was like seven years. Okay. So not insane. And like, crypto terms.

Boxer  1:07:06  

Yeah, depending on when this was…like 20 to 30k or 15 to 30k. Yeah, stuff like that. Yeah. Interesting. Interesting.

Andrew  1:07:20  

Yeah. Other favorites. Blackhand…obviously a favorite. It's like a Web3 based crypto gaming organization. So, they have like Halo team…they have an apex team. 

They funded like 1.3 million on mere eSports. Organization. Sorry, yes. E-sports organization. And they have a few players on their roster. They're super cool. 

Etherium film was like the biggest one. For sure. And obviously, we're still very excited to see what comes out of it when the film drops. That one raise 1000 ETH. Which is also like a lot. 

Okay, now that's a lot. Yeah, for formula. Yeah, obviously, it spans a spectrum. But now, we've mostly just seen like a bunch of like 20 to 30 crowd funds. I would say for random people's smaller projects. 

I don't personally prefer that, or it's huge…1000…these projects were more just like these random communities that we've never heard of. And like we'd look at the crowd fund. We're just like, what is this?

Boxer  1:08:48  

Yeah, that's the beauty, right? Like any community. 

Andrew  1:08:51  

Yeah, exactly. I guess. I would have never seen this on Kickstarter. Like people funded their whole…someone funded a crazy Dungeons and Dragons campaign off of it. I'm like, oh, okay.

Boxer  1:09:06  

Okay. That's cool. That's interesting. But wait, it's more fun to do a Kickstarter where people actually get stuff back or like an NST back or something or in a GoFundMe kind of way as well, where people are like…I'm in the US and now bankrupt.

Andrew  1:09:29  

It's marketed in a Kickstarter kind of way. Like most of them are like, Oh, if you participate, oh, sorry. Like a big example. A good example is like mad realities TV. They funded a dating crypto dating show. 

And they fund it on the air, and they have like three tiers of NFTs where it's like, oh, based on how much you contributed, you get more voting power of which candidates go on the show? You know? 

Boxer  1:09:56  

It’s super interesting. Yeah. They're really interesting.

Andrew  1:10:00  

They just raised 11 million, I think. Excuse me. Like a seed round. Not on here.

Boxer  1:10:11  

Yeah, but then 11 million seed round. What the fuck? I guess I've seen. They're actually producing videos. They've already had one. Okay. Okay, so I don't need just like a bunch of computers and developers.

Andrew  1:10:27  

They're like immediate. Okay. Okay, I think paradigm was involved. So, I trust it. They gotta get…they got a really smart team.

Boxer  1:10:35  

Yeah, it sounds really interesting. I'm just…

Andrew  1:10:40  

Hey, like all these crowd funds they have…promises in the entry, but Mirror is not responsible for….

Boxer  1:10:51  

…no securities you guys…

Andrew  1:10:58  

Fom a technical perspective, it's like purely GoFundMe style. Yeah, no, okay. You raise how much you raise, you get these tokens? They're there or an NFC or whatever? Yeah. Yeah. Okay. Interesting. So, yeah, cool. 

Boxer  1:11:14  

Yeah, it’s very interesting that you guys started all this writing platform and then spun off into Kickstarter. 

And now you're back to…let's actually make a great…it's certainly an interesting development. I don't know. It kind of makes sense in terms of like distributed ownership economy where it cut out the middleman, enabling all of these projects to thrive. 

And obviously, if you want to introduce something to your community, you're gonna be using some kind of blog post article. And that seems to be a very natural connection between those two things. 

Andrew  1:11:51  

Yeah, exactly. Yeah. And also just the crypto space is very distracting. Just think about how many NFT drops were happening last year and constitution dollars last year…

Boxer  1:12:05  

Yeah. People bought the Doom book. They realized afterwards they don't have the IP.

Andrew  1:12:15  

I haven't heard any.

Boxer  1:12:16  

So, it's still the best story.

Andrew  1:12:19  

Point being, there's so many distractions like that. And it's just like, I don't know. Yeah, that's fine. It's hard to say.

Boxer  1:12:26  

Yeah, but you guys have done a great job. I guess, all of these things have been a success. No. Is there? Is there one point where you're like, we should have not…like this was a total? Oh, they probably reserve auctions. 

Andrew  1:12:37  

But that was like very early. Like you've probably…it's like basically bizarre auctions. Yeah, we had that. And it kind of works.

Boxer  1:12:52  

Yeah, but for written content. It's a bit. Yeah, maybe for a poem or something. That was backwards. Somebody who writes poems.

Andrew  1:13:00  

I'm probably…I don't know. I don't know. 

Boxer  1:13:06  

Sounds like an interesting avenue.

Andrew  1:13:08  

I could do a text search. But that was back one of the days where I thought like, oh, all of these writing NFTs are going to be one of you know…so it makes sense to have an auction for it. 

But then…it became apparent, why limited to one collector at some random price point? We should make these additions. And as soon as you have additions, that auction becomes pointless. 

And it's like, okay, if someone wants to use our auction, they need to mint it on Zuora or Open Sea and use our auctions. And it's like, okay, that's a lot.

Boxer  1:13:45  

Yeah.

Andrew  1:13:47  

It made sense at the time. And then we deprecated it because it's like, okay, we're never not…

Boxer  1:13:55  

I see sec. Somebody learned something during that, I hope.

Andrew  1:13:59  

Yeah. A lot of learnings a lot. Yeah.

Boxer  1:14:04  

Yeah. So, that's Mirror. I've certainly been very enlightened during our conversation. No, I've never really thought about this whole ownership consumption like the battle that's happening. 

Very cool. Yeah. So, Web3 education. You certainly are one of the great heroes in the data analytics space, bringing education forward. So, where do you think the general with three education landscape is today, and how could we maybe improve upon it? 

Andrew  1:14:41  

Yeah, that's a great question. Something I've thought about a lot. I think, Okay, last year, I think we're in a space where it's a technical…I'm just going to talk about data education. I don't want to get into the rest of it because it's complex. 

Yeah. But on the data side of things, I think last year where most people were struggling was all the technical aspects of data. 

This was even before the dude had all the great abstractions tables that it did now. There was the decoding process. It was eventually, but it's not as nice of flow as it is now. Yeah, right. 

I definitely had a few times where I was like, wait, this is not the right address. I think it's just the technical knowledge that people were not as good. Even Etherscan wasn't as strong as it is now. 

So, last year, my focus is just putting out technical content of like, okay, we needed like help. People understand at a basic level how contracts work, how the three tables do work, how to vote, and how these ERC standards work. 

Because that was where most of the questions I got from people were coming from. And then that all culminated in our network for like the 3030 students Thursday's thing, where it's like, okay, I think people are starting to get the technical side of things, but they're not understanding the community side. 

I don't know what it is. But a lot of the data analysts I talked to would play with crypto data. They would find it fun. And they wouldn't do anything else with it because it's not like hackathons where it's like, oh, I can build this like new escrow app and get paid by sponsors to build this app and put it out there. 

And it stands alone, right? For data stuff. You do some obscure query. No one's gonna read your query. You probably don't really know how to interpret it as in like someone new to the system. You probably don't know how to interpret it in the context of everything else, right? Because there's so much noise coming around. 

If you just throw out a TVL metric, people are gonna be like okay. Most people are too scared to tweet anyways. I was like, Okay, this is a problem, we need to teach people how to engage with the community. 

So, that's what I did on our course. I think we did a good job of basically teaching those 30 people not only is this how you do Krypton analytics, but this is how you should do a write up of it. 

And you should be tweeting this every single day. We were basically pressuring them. Like tweet this. Put this out. I'm gonna tweet this for you. And that got like…I think quite a few of them. Obviously, 15 of them are hiring full-time now.

Boxer  1:17:42  

It’s absolutely mental. I think my internal count is 12 or something, but maybe I had a few people. It's absolutely crazy. 

Yeah, it was out of 30 people, and there was like…I think 27 actively participated. So basically, 55% of people actually have full-time jobs now or 50/50 or something. 

Andrew  1:18:07  

But it's constantly learned how to engage with the community. Like they went from just attending the calls and asking a few questions to actively…I saw a lot of them were like active in the doing discord. 

Even after the class is over…they'd be answering questions or… like Grace and Tuition would put out like articles on Twitter of stuff they're looking at. Seeing that was like, okay, it clicked for them. This is a community-based game. 

It's not just like, analyze, put out dashboards, and wait for people to ask them questions about it. Because if you just write a dude, create a new dashboard, and put it there, and don't do anything, it almost doesn't matter how good of a dashboard it is. No one's gonna go and be like, oh, hey, look at this great dash.

Boxer  1:18:57  

Yeah, in very rare cases, if you've really hit the nail on the head with like…if you have like a great Luna…you wouldn't like people search for Luna and then they're like, oh, this is like the best dashboard. 

Let’s actually pull the data from there. Let’s make a Twitter thread using this guy's data. If somebody starts consuming your dashboard autonomously, then the chances of that are usually pretty slim. Yeah.

Andrew  1:19:30  

I agree. Yeah. Even when that does happen, that person normally doesn't have a Twitter. 

Boxer  1:19:35  

There’s no way. Yeah. Recognition use you.

Andrew  1:19:39  

Yeah, it doesn't help you, you know. It's like, okay, cool that’s Twitter…no one's gonna be like, oh, wait, let me go look at that handle and try to contact me, you know what I mean? Like, no one does that, you know, so my focus now has been like, alright, how do we teach people to be more community engaged? 

That's probably going to be my focus for this year. Because I think like last year was like, Oh, hey, here's this Azureus. Do this analysis. Get the daily sum of whatever trades on this unit. Swap the two versus three thing, right? 

Versus this year. It's like, alright, I'm working, I'll drop the tea. I'm working on like a 20 or 30 problem practice set, basically have like data questions. It’s like, Hey, if you want to be a data analyst in the space, you need to be able to do these like 20 questions, but they're formatted in a way of like you saw something on Twitter. There's no links to any addresses or anything. It's some obscure thing going on. It’s basically stuff like that. 

Boxer  1:20:48  

It’s like a real world example of like a treasure hunt.

Andrew  1:20:51  

Exactly. Because I realized you need to teach people to start from like a Twitter thread. So, they're forced to go and talk to the communities. I can't just keep putting out your address. 

Here's the table schema. Calculate like that because people don't struggle with the sequel side for the book. Okay. For the most part, the sequel side, I think there's enough resources on that learning. Yeah. 

But going from a random tweet about Luna to actually analyzing contracts, that's the hardest step, you know. So, that’s, I think, a big opportunity from an education standpoint.

Boxer  1:21:28  

Yeah. Interesting. Teaching even more like not DeFi context, but rather the whole environment, like yes, this early Metaverse environment where it’s like, “How the fuck do I actually navigate all these different like, resources?” 

Yeah, dogs off the project…and this code of the project…and kind of like…can I ask what are just sequels this year? Exactly. That's super interesting.

Andrew  1:21:54  

So, I think we need to find a way to get people there. Yeah, that's a big opportunity. Anyways, we'll see what happens. That's one of my goals for the next…

Boxer  1:22:04  

I see. Yeah, I'm certainly excited for the Andrew Twitch streams, Andrew's treasure hunting data, or something. Yeah.

Andrew  1:22:13  

But, yeah. I don't know. I haven't figured out the business opportunity side of it. If you're gonna ask me about that, I don't know how to generate money off. That's not my stress.

Boxer  1:22:30  

Maybe you could sell it as an additional neuro.

Andrew  1:22:34  

It's crossed my mind. Problem and hefty. Got it. You gotta do what you gotta do.

Boxer  1:22:47  

Yeah, cool. Um, so yeah, Web3 education. I don't really think we are also running out of time. So, if Thomas is here, he probably has some questions gathered from the chat. And usually, I’ve done them before. So, you guys, Thomas.

Thomas  1:23:15  

Alright, so we've got a couple of questions. I'll start with this one from Hilldobby. Why the username? Let me know. 

Andrew  1:23:23  

Okay, so that’s a classic. I think a lot of us have this. This was just my gamertag from middle school. I had this crush on this girl named Emily. So, it stands for I love Emily because it was unique enough that it was never taken on any gaming server. 

My other username was like XMen35. I used that for years, but then like that started getting taken. So, I was like, okay, you know what? I have this other username from this crush that I had. I've just kept with it. She has no idea.

Thomas  1:24:14  

What are some unintuitive lessons you've learned about engaging in Web3 communities?

Boxer  1:24:22  

Oh, unintuitive I mean, thank you was the most unintuitive. Maybe…give you time to think it's…don't assume that anyone is too important to talk to you. Like usually if you like you can write the main developer of some projects like at 3 AM my time. And they, for some reason, actually answer, and this has happened and shouldn't have at times. 

Always feel free to like just DM anyone. People in the space  are super happy to help you, get forward, teach you, so take mentorship. 

But being in a position where people do reach out to me, like, try to have something like substantial that you actually ask me, because I don't really like…I like I talked to people…like shit on and like…the conversations that I enjoy….

It's when I actually see that the other person is getting ahead. Don't start a conversation with “Hey, man, how you doing?” No small talk, please. 

Like this doesn't… we are here to get something done. And if we've talked a bunch of times, then we can be like, “Hey, how do you like…” This is like a very different dynamic. 

I don't know, We're not like…we went to robot normies where you have to be like, “Hey, are you doing?” No, I'm great. Like, oh, yeah, I'm great as well.

Let's get to the extra like…no, just fucking ask me the question that you have. And don't ask, “Can I ask you a question, please?” 

But reach out to people. People are super helpful. I think that's very under appreciated. And a hidden gem in the space pretty much.

Andrew  1:26:26  

Yeah. I think my one and only biggest thing I say to everyone is contribute. Don't apply. If there's some community you want to get involved with…same with what Boxer was saying…don't just go into Disco and say, “Hey, can I help with something?”

Yeah, I personally hate when people just say, “Hey, how do I get involved? Can I help here? I'm a geek.” 

I'm sure you can. But like the people that I ended up working with the most are actually helping. They look at what I'm working on or what someone's working on. 

And they start to try to do something with it. And it's usually pretty ugly, you know, not gonna lie. But it's like, you see that they put in some sort of effort, right? 

And it's like, oh, this is someone who wants to contribute and has that blood, I'm gonna go reach out to them and like, try to help them on this or try to get them involved, you know? 

And like that gets you going in the system…not just sitting there applying on LinkedIn. Like please. Please don't spend time on LinkedIn applying to jobs.

Boxer  1:27:33  

Yes. You'll just wreck yourself. You're now probably working at the DeFi mullet startup. Funds an anchor.

Andrew  1:27:43  

Yeah, that's my biggest advice.

Boxer  1:27:45  

If you can financially afford it…that's always a part of this…where it's like some people can't just spend like…if she doesn't have time to contribute for free. 

So that’s always something to consider. But then…I don't know. They would have to not have a job currently and not be able to side hustle for some time. I don't know. That's usually like…I think the side hustle is very approachable. 

And then if that team actually starts hiring in the area that you've been contributing to…of course, they're gonnago to you first because you're already on boarded. 

Right? There's no question…who else would you actually hire? Then, this guy was like…obviously…does the stuff already and like Genesis should…

Thomas  1:28:34  

100% David says he'd love it if Andrew would discuss metrics dial since he's a councilmember.

Andrew  1:28:45  

Oh, yeah, sure. Metrics now is a spin-off of flip sides of bounty program. And a much more open and collaborative way. I think they're still very early on. I would say they don't have…they've worked on like a tokenomics model on the white paper. 

They do a few fun workshops here, and they're definitely like a good place to get involved in data if you're like…specifically, I would say if you're looking at non…not just your core EVM chains. I see like bad timing now. 

But if you're looking at Terra then they would have been like the best place to start. Still the best place to start is looking at Terra. But yeah, probably one of the bigger community-specific not product-specific data communities out there. 

I would say, if you're looking at Terror, look away. But yeah, great people a lot. I mean, again, I'm going back to my key people that you're gonna see the same people from across data in all these communities. 

So like, just find one of them and start chatting. Getting involved. I'm not gonna say, “Oh, get started here,” then like just start all over and overanalyze it. Yeah.

Thomas  1:30:23  

All right, great. Let's have one more from us to DeFi. The future of Web3 data analytics will require in your opinion, which languages and tools solidity…Russ, Python, SQL, graph, QL, etc.

Andrew  1:30:39  

I hope you don't have data analytics and solidity. Honestly, in traditional Web2, I would say Pythons most important. And Web3 sequels more important than Python by far for data analytics rules. 

For data engineering, you need to know Python both for like purely data analytics if you're splitting your time 5050 between SQL and Python…just do 100% of your time. And SQL. Just don't even worry about Python. Because dude has figured out all the Python for you. 

Basically basically, you need Python. Like get those Dakota tables, right? But like dudes already done that for you. Just do SQL and like touched Python. We absolutely need to. 

And if you want to do graph data and SAS then you have to cipher, which is like new for JS language. But that's not…I wouldn't worry too much about Graph QL, solidity, JavaScript, and one that's like nice to know. Like, if you have the time to learn then go ahead, but it's not necessary.

Thomas  1:31:55  

Awesome. I think that wraps up the questions. Amazing.

Andrew  1:31:59  

I realized I'm still sharing my screen.

Boxer  1:32:02  

Yeah, that’s the way to go.

Yeah. Anything you want to leave with us? Maybe some…what's your favorite next feature of Dune? Or what would you be really excited about if we just release party data?

Andrew  1:32:20  

API data into these main tables.

Boxer  1:32:29  

Miguel will be very excited for that API product manager.

Andrew  1:32:35  

I'll tell you what I'm doing right. I can't keep doing abstractions. PRs for the date…I want to bring it…

Boxer  1:32:41  

Oh, yeah.

Andrew  1:32:44  

Right now is export from my API, put into a JavaScript file or SQL file, PRP like, Hey, can you add these 50k rows?

Boxer  1:32:55  

That's…yeah…

Andrew  1:32:57  

That's my top request.

Boxer  1:32:59  

Okay, see? Definitely doable. We hope to be able to provide you with service soon.

Andrew  1:33:09  

Other than that…only last thing I'll say is if you're looking to get a data Web3 role, please apply to the talent board. I think it's in the description. It's also my Twitter. But it's just alemi.palette.com/talent.

Boxer  1:33:28  

It's in there. Yeah. I just put a Dune drop in there today as well. So oh, you still need to approve that. 

Andrew  1:33:37  

I need to prove it? Yeah. But yeah. Amazing. Yeah.

Boxer  1:33:43  

Thanks so much for coming on. Andrew. Really appreciate your work across the space. Like really most productive person I know. Yeah, always great talking to you and see you soon. Bye. Bye.

‍