GRTiQ Podcast: 82 Roneil Rumburg

The following podcast is for informational purposes only. The contents of this podcast do not constitute tax, legal or investment advice. Take responsibility for your own decisions, consult with the proper professionals and do your own research.

Roneil Rumburg (00:00:19):

We use it pretty actively. If you ever go to dashboard.audius.co, that’s all pulling from a set of subgraphs that are curated. And then we were also fortunate to get to be among the first to use the decentralized graph offering as well. So, all of Audius’ graph engagement happens via the decentralized product. So yeah, we’re big fans of The Graph.

Nick (00:01:14):

Welcome to the GRTiQ Podcast. Today I’m speaking with Roneil Rumburg, co-founder and CEO at Audius, a web3 dapp that connects fans directly with artists with the mission to empower a new generation of artists on the decentralized web. As is made clear throughout this interview, Roneil’s entrepreneurial journey and the success of Audius are both perfect examples of what’s possible in web3. Roneil, a self-described builder, is a brilliant founder and entrepreneur. It’d be easy to overlook Roneil’s early success because of his soft spoken and humble approach, but as you’ll hear, his background is very impressive, including time studying at Stanford University, a period where he worked at Kleiner Perkins, a well-known investments firm, and eventually following his passion for building to unique experiences working as an entrepreneur, which culminated in the launch of Audius. During our discussion, Roneil talks about his early start in software engineering, which started in high school, his drive as a builder, the backstory behind Audius and the unique nature of this three-sided platform, and a great discussion of how Audius uses The Graph. As always, I started the discussion with Roneil by asking about his educational background.

Roneil Rumburg (00:02:34):

I’d say that’s a great question. I am mostly, I guess, from the perspective of the skills I use professionally, I’m mostly informally trained or self-trained. I learned to be a software engineer just by being part of some of the early iOS like script kiddie, and then later actively trying to work on those things myself. That was how I learned to kind of program and do things like that. But yeah, went to public schools growing up, went to Stanford for university and studied computer science there. Was already kind of a software engineer doing that work professionally in high school. So it ended up being just kind of a waste of a college education for the most part, but made a lot of good friends and had a good time. But because of that, got to finish up a year early too. But that’s it for me. Just undergraduate and nothing else.

Nick (00:03:30):

I’m interested about this computer science degree, and you’re not the first computer scientist to be on the podcast, but I’m never really sure what that education actually means. And I’ve heard a lot of different answers. So how would you describe to people who say, what does a degree in computer science mean?

Roneil Rumburg (00:03:45):

Yeah, I think what it means is you can take a bunch of tests and otherwise do math problems and things. There’s very little correlation between what they teach you there and what you actually do in software engineering. It’ll give you a good theoretical backing for the practical work that you do, but ultimately, most of the kind of practical stuff that you’re actually doing is not going to be coming from school. So see, it’s sort of a lightweight math degree with a little bit of software project work. That’s how I would describe it.

Nick (00:04:21):

And I’m loving this fact that here you are in high school, you eventually go on to get a degree in computer science, but you’re actually a software developer in high school. What’s going on at that time? What are you working on?

Roneil Rumburg (00:04:32):

Oh, so yeah, like I said, I got my start with some of the iOS, and then actually, to a greater extent, on the programming side in some of the Hackintosh communities. So this Hackintosh movement was to get, when Mac OS first started running on Intel machines, there was a group of people who worked on hacking on that to get it to run on commodity PC hardware. So kind of was part of those circles and working on things. And I learned my way around the command line interface and then ultimately putting together some really, really simple kernel extensions and things to support certain hardware that I had that I couldn’t get supported otherwise. And then, that ended up leading to, I spent a summer doing SisAdmin work at a local IT consulting company in the Washington DC suburbs where I grew up. That was summer of 10th grade or so, and then just kept doing stuff like that.

(00:05:37):

So yeah, I don’t know, I guess I slowly fell into doing that work just because I was interested in it and enjoyed it, and then had a friend who was working at that company who told me, oh yeah, you can also use an older upperclassmen in high school. He was like, oh yeah, you probably would be fine, do well here. You should check it out. I don’t know. Yeah, I feel like these answers are always unsatisfactory. What it really came down to is just I enjoyed tinkering with various things and tinkering ended up being fortunate in leading to a career direction that made sense.

Nick (00:06:19):

Well, in preparing for this interview and doing some research on you, I was struck by two threads that I hope to explore with you today. The first one is, you’re the quintessential builder, right? Somebody that gets their hands dirty and knows how to build in technology. And we’re going to talk a lot about Audius and some of the things that you’re working on there. But on the other hand, there’s this entrepreneurial journey I see in you and maybe some passion for that space. You’ve done it in investments and now you’re doing it as an entrepreneur yourself. Do you see yourself as an entrepreneur? Is that kind of how you think about your career and what you want to do for a living as being an entrepreneur?

Roneil Rumburg (00:06:52):

Honestly, no. I guess I’ve never, I mean, I think that’s an apt description, but I never really thought about that as directed at myself. I think I just like building stuff. And when there are opportunities to build things and then hopefully get people to use them, I enjoy doing that. I guess it’s very simple. Yeah, there wasn’t, and that’s, I think, probably what’s coming across in how I’m answering a lot of these things. There wasn’t some master plan I was executing against here, you know? I just followed what I enjoyed doing and tried to keep doing more of that. And that’s actually what led me away from investing at that time was I just really missed building stuff and that’s why I stepped away from that and wanted to get back to actually making things. But I didn’t really set out to go found companies or do things like this. It more came of, I felt like there were cool opportunities to build things from scratch that people hadn’t built before, hadn’t tried before, and hopefully get people excited about those things and get them to use those things.

Nick (00:08:07):

Being a builder is, in a lot of ways, being an entrepreneur, but of course, there’s large organizations with teams that build within them, so it’s not always exactly that way. But I still feel like personally when I’ve interviewed entrepreneurs or when I’ve interviewed people, they say I’m less of an entrepreneur, I’m more of a builder. The skill sets are the same. These are people that can look off into the distance and maybe identify a need that hasn’t quite materialized yet and then they can do all that really hard work from going zero to one. What have you learned, in that experience, if I’ve characterized it correctly, of being a builder? Kind of the skills or the lessons you’ve learned as you reflect back on all the things you’ve built?

Roneil Rumburg (00:08:47):

I think there are a few salient ones, but first and foremost, I think when I’ve tried to build things that weren’t directed at a need or a thing that I directly faced, usually things didn’t work out because I didn’t actually really understand why I was building what I did. So I think that’s one aspect. There’s another one, which is in some cases when I’ve built things to try to solve the problems that I had. There’s nothing, I guess, nothing wrong with this direction because I still do use some of these things myself, but if I didn’t stop to ask, does anyone else want this or would anyone else care about this? That was what sort of doomed the first project that I tried to go after with one of my college friends in 2014. We built this Bitcoin peer-to-peer payment product, think Venmo for Bitcoin.

(00:09:42):

And the reason we built it was because he and I had been doing some mining and other things and we had a small group of friends that was as well. At that time, we were mining these scrypt altcoins, at that time, Namecoin and Dogecoin and things like that. This was pre Ethereum and pre tokens, everything. So yeah, we were mining those things. We would exchange everything we earned to Bitcoin and then we just all had a little bit of Bitcoin and we wanted to start paying each other back for dinner, doing things like that with it. And there wasn’t really a good way to do that. So we were like, why don’t we go build a thing to do that? And it was a good product and we grew to 25,000 or so monthly spenders on there, but we could not figure out how to grow past that.

(00:10:29):

And we realized that the thing that was broken in our market was everyone wanted to receive Bitcoin. No one wanted to send it. If you think about who is the kind of bitcoin accumulator archetype, that’s basically everyone who was in crypto at that time. No one actually wanted to be spending their Bitcoin. They all wanted to amass it. And if one side of your two-sided marketplace doesn’t want to use it, that’s kind of a problem. So it didn’t end up working out. So those are two more salient lessons and I know they’re sort of opposing of one another in a way, which is, one is that I feel like I need to understand why I’m building the thing I’m building to be able to do a good job with it, rather than trying to build for other people. But two, if you only build for yourself, you’re probably not going to get anywhere if your ambition is to build something that a lot of people use. If it’s not and there’s the time and a place for that too, I think that’s totally fine also.

Nick (00:11:33):

I want to ask you an incredibly difficult question and I’m very interested in your answer. And you may not have an answer for this, which would be totally fine because it is a super difficult question. But have you thought about, or do you have advice for what point a builder or an entrepreneur should stop? What’s the milepost at which you might have that epiphany or you should have that epiphany, this isn’t going to work? I mean, it seems so fuzzy and hazy to me that people can spend years of their lives and maybe there was an indicator at some point that they should have paid attention to you. Do you ever think about this?

Roneil Rumburg (00:12:08):

Yeah, I definitely do. Because when I stopped working on that Bitcoin payment thing, for example, it wasn’t entirely obvious that, had we chosen to continue, things would work out. It’s definitely possible that that happened. There were two competitors of ours at that time, and it’s funny in retrospect to think about them. We’re a Circle. So Circle prior to doing USDC was a payments company. They wanted to be this really great fiat to crypto on-ramp and then let people make payments on the crypto side. They didn’t go after the identity problem as much as we did. And then there was another company which I actually can’t even remember what they called themselves at that time, but it’s Blockstack now. So Blockstack prior to becoming Blockstack was, I think it was one name. It was basically a crypto native, kind of like ENS before ENS, if you think about it that way.

(00:13:09):

It was a way to attach a wallet to an identity that you register and then you could pay people against the identity. So it was another different take on the same market that we were trying to go after. It turned out all three of us were going after a market that didn’t matter and those two pivoted away to work on different things we wound down. And I don’t know, maybe we should’ve tried to find a new idea or done something else. So, I think also personally, at that time, I was nearing the end of my rope financially to be able to just keep building stuff without earning money and that was also a factor in the decision. So yeah, I think you kind of know it in your heart of hearts when something, the direction you’re going in is not going to work out or at some stage of working on it, you kind of get there.

(00:14:06):

And I think there are two ways to respond to that feeling. One is to just push through. The other way is to change course. And in that case, we didn’t have enough financial runway to actually change course, so we’re like, all right, we just wind down. But yeah, when my co-founder and I at that time talked through those options, I think we came to the consensus that it probably made sense to step away just given that there’s really not much more that we could financially afford to do at that time without resorting to debt or doing things like that.

Nick (00:14:43):

I appreciate that answer. It’s very insightful and I’m sure my listeners will as well. I want to ask you this other question about being a builder. You’ve been at it a long time, started doing it in high school, went to Stanford, which is a highly respected university here in the United States and across the world. You’re a young person and you could have gone into web2 and worked at any large company. You got great pedigree, you got great skills, and yet you decided to go off and venture into web3 and crypto and build there. So, my first question is, why would someone like you want to do that? And the kind of follow up is, do you think that’s true of all younger builders right now in the world that are trying to make an impact? Are they all migrating into web3 and crypto as well?

Roneil Rumburg (00:15:29):

Yeah, so why I ended up in it, I had a very keen interest in distributed systems stuff in high school and then early on in college, and I guess this is the common theme with a lot of what I’m telling you. This just happened to be the case. I don’t know why I was interested in some of those problems and problem sets and that’s what ended up leading me to crypto was seeing that this felt like the area that you could do greenfield distributed systems work. That space as a general category is, and I know any true academic that works in that area would be very unhappy with me characterizing this way because I’m sure they don’t feel this way, but feels kind of played out. The solutions to most problems there are pretty well understood, right? That is to say, in classical distributed systems, there is not typically this lack of trust between the participants in the system.

(00:16:37):

So if you assume that everyone, all these machines trust each other, there are pretty well understood ways to find consensus to a question or do something like leader election or even to do peer discovery for example. But when you layer in, what if these machines don’t trust each other? And that is the problem space that crypto opened up for the first time, I think. And well, it’s not the first time. There is prior work there, but there was an explosion of new work in that area, I think, that was created by that direction of work. And Nakamoto Consensus was a truly novel solution to that sort of problem space. How do you find consensus in a situation where folks cannot trust one another? And that’s where there are certain assumptions about Nakamoto Consensus. If more than 51% of hashing power on the network is honest, then you know that you can get to an honest answer even if 49% of it is dishonest.

(00:17:41):

So there’s some really cool stuff there, but I don’t know, I got excited about what are the new things that we can build around this and the areas that we can explore. And that’s what ended up leading me to this. But yeah, prior to starting to mine in college was my first time interfacing with crypto. I did internships in college at two kind of classical distributed systems companies. One was called Arrow FS. It was a really cool product that, kind of think Dropbox, but peer to peer. So if me and you and five other people work at a company and we have a shared folder within the company, we could actually kind of share that folder between one another without all of that stuff being stored on a server somewhere. This was in like 2011, right? This was before Bitcoin was really a thing or anything. I thought it was just cool tech and really, really great team working on it.

(00:18:40):

And then, in 2012 I worked at this company, Nebula, that was an open source development company working on this thing open stack that Rackspace and a few others used to orchestrate their cloud products. It is sort of like an open source version of AWS is how I describe it. But orchestrating these systems and doing kind of practical distributed systems work rather than theoretical. But yeah, I think because of that, when I saw some of the crypto, what was happening in crypto was exciting to me and I felt like there was just cool things potentially to be built in that area.

Nick (00:19:25):

And do you think young people are heading into this space? Are you symptomatic of what next gen talent are interested in and the kind of career moves they’re making?

Roneil Rumburg (00:19:34):

Yeah, so I think what’s changed from 2013, ’14 when I started working on things in crypto compared to today is that the kind of tooling to build things is much, much better today than it was back then. I think you as a solo developer today or a couple of people can build pretty incredible stuff in crypto now and ship it and have that see the light of day. So I think that has changed. So too has, I think, the network effects around these underlying networks which opens up more opportunities for higher level building. Something like the Bitcoin peer-to-peer payment thing I worked on in 2014. That was trying to climb up a hill without shoes or something, right? Because there’s only so much that two people could build when the market for what we are building didn’t exist yet. But I think there are now channels to reach distribution to users.

(00:20:38):

There are channels to short circuit a lot of the upfront work that you might need to do, just everything from code libraries that you can import from to well understood practices for auditing and things like this. I mean, none of that existed back then. And I think that lack of structure attracted a certain type of wild west mentality and also that sort of set of individuals, and I think the space has matured a lot since then, which is both good and bad. I think it’s good in that, to your question, young people are coming into the space in droves right now, because when I talk to even younger members of the Audius ecosystem or across other projects, think about it this way. If you spend a year in crypto, you’re probably among the top five, 10% of experts at it. If you go do that for AI or for any other well established area of software, there’s a lot more competition.

(00:21:42):

There’s a lot less greenfield work to do. And that combination of things like crypto feels like the frontier still, I think, in technology, and that attracts a certain type of person. But I think the risk is lower now. A lot of people in 2014 when I was working on that asked exactly the same question you just asked me, which is like, why don’t you go work at Google or Facebook or one of the places that you’re supposed to work? Someone like you is supposed to go there and supposed to do those things. And I just didn’t really want to. And I was like, well, I can always go work there in six months or a year if this doesn’t work out, so why not? And this is more fun for me. That was basically all the rationale, right? Today is super, super different. Right here we’re recording in August of 2022.

(00:22:33):

None of those big companies are hiring right now. So if you’re a young person, there’s actually, as we’re starting to go into a potential recession, whatever, there’s not as much of a path for you to follow. Whereas in 2014, when I was wrapping up at school, there was a path. There was a path that most people followed. It’s like go work at Fang and cut your teeth and then maybe in a few years or four years or five years, then you go do interesting work. I don’t mean to speak about that pejoratively. I think there’s a lot of interesting work that they do there too. But you’re not going to have any sort of meaningful level of agency over the work that you do at a company like that. And that’s okay. I mean, some people choose to make that trade off and there’s nothing wrong with that. That wasn’t for me, I guess.

Nick (00:24:49):

After your time at Arrow FS and Nebula, which you just described there, you obviously got some education and did other things, but you ended up for a while at Kleiner Perkins working on the investment side of crypto and investing in technology. What did you do there during the time you were there and any lessons come out of that that kind of led you to where you are today?

Roneil Rumburg (00:25:10):

Yeah, so I was covering two areas primarily on the investment side, crypto and computer vision, or as a subset of AI more broadly. So there was kind of a renaissance of sorts happening in AI at that time or that really started in 2013, 2014 or so around what these convolutional neural networks were able to do and accomplish against older problems. We were just every year at that time improving by leaps and bounds on what image recognition and things along those lines could do. So I spent time there and then I spent time on crypto. So I did the seed round for Lightning Labs, for example, the Lightning Network on Bitcoin. Got to participate in seeds for a few other cool companies during that time as well, one of which is still operating but ended up pivoting away from crypto to some extent. But it was actually yet another payments company along the lines of what I was excited about.

(00:26:16):

So yeah, it was a really, really great time for me, I think, to just understand a different level of the game. As a software engineer, building things for other people. It was cool to see what the big leagues of building for yourself looked like compared to what I had tried to do previously. And it also, I think, helped teach me or I did go into that role with the intention of wanting to learn how to think about business opportunities more analytically, because I think, I felt like that was the mistake that I made on the Bitcoin peer-to-peer payment thing was not thinking critically or really at all about who else is going to use this, how big is the market, what’s our path to building a market if one doesn’t exist? If there is a market, how big is it? How quickly is it growing?

(00:27:09):

All these really basic things that I just never really considered. I was just like, this can be built, it’s cool, it’ll be fun, so let’s build it. That was the extent of my 20-year-old mind thought through the question. I was like, it can be built, so it should be built, so let’s build it. So working as an investor was really, really valuable from all of those directions. It helped me have some convening power with respect to getting to meet really cool, exciting people, and a lot of that network ultimately helped a lot with my work on Audius afterwards.

(00:27:48):

So yeah, it was a really great time there, but I just missed building stuff and I think that’s ultimately what led me to think about a next step and what might make sense for me to work on. And the idea for Audius actually came while I was at Kleiner in like 2015 or so. But the tech, I think, for us to do, to build what we wanted to, my co-founder Boris and I were talking about it back then, just didn’t exist. So we shelved it, but in 2017, the stars aligned for us to work on something together. We were like, oh, I wonder if that music co-op thing could work today? And turned out, it could, at least so far, and there’s more work to do.

Nick (00:28:36):

I want to ask you a lot of follow up questions about Audius and the origins there and what you’re building, but before we turn our attention there, just a couple follow-up questions. You said you went to Kleiner Perkins largely to kind of build out your skillset and understanding of just business analytics or analysis, what makes a good business opportunity versus what doesn’t make a good business opportunity. And you mentioned some things there, but a lot of entrepreneurs pass by that stuff. Maybe it’s too academic, maybe it is too constraining on something that’s so passion driven. So based on that education you went there to get and some of the things you learned, is it your position that entrepreneurs should spend time thinking about these types of things? Is that what you ended up understanding, those things are critically important?

Roneil Rumburg (00:29:20):

Yeah. Yes and no, I would say. Yes, you need to have some level of understanding that was certainly higher than my level of understanding prior to learning those things, but I also think if you understand those things too much, they kind of constrain your thinking to some extent. Not all businesses need to be profitable from day one so long as you understand a path towards sustainability. And sustainability doesn’t necessarily mean profitability. And I think that’s also one of the cool things about the crypto movement broadly is we’re opening up different ways to sustain open source development over time that aren’t necessarily value extractive from the existing world or existing parties. And that’s really cool. But putting all that aside, I did read those blog posts, I took some courses in college about those things. Without the context of actually thinking about how I would apply those in situations that I myself was facing, all that stuff just went in one ear and out the other.

(00:30:25):

There are these people who sit all day and read hacker news and the sort of various Twitterati of startup hustle porn, but I don’t think those things are actually helpful without some context to put them in. Because now I can think about, oh yeah, I did make these mistakes and now I understand where I could have used these tools and this knowledge to avoid making those mistakes, right? Without that, it’s like you’re just learning a bunch of abstract concepts and they’re not rooted in any reality that you are thinking about and facing and dealing with on a regular basis. So that’s not to say I think there is a time and a place and a value for business education for example, right? I just think it should actually honestly come later in folks’ career.

(00:31:24):

And I don’t know, I feel like now if I went and got an MBA or something, I feel like I’d actually get value from that because I have all this context. But there are probably faster ways to learn the same things than spending two years in school. So I don’t know, I think to each their own. And to your question of should entrepreneurs do this or should entrepreneurs not do this, broadly, I don’t like giving people advice characterized in that manner because I think, ultimately, to build something interesting that’s different to the world, you have to have a contrarian view to that. And I think I’ve come across successful founders who have, think of any possible permutation of what level of education did you have? How did you go about building it?

(00:32:17):

There are plenty of people I know who’ve been super successful just building whatever cool idea came to them without ever thinking analytically about it and it happened to work out, right? And sometimes it doesn’t work out, but maybe they just had an intuition for what the world would also want where maybe I didn’t. So I don’t think there’s a right way to do this, I guess, is what I’m trying to say in probably too many words here. But you find your own way one way or another and then either it’ll work out or it won’t work out and you’ll lick your wounds and you’ll figure out what to do differently next time.

Nick (00:32:54):

Then the second follow up question I want to ask before we turn our attention to Audius is, for those listening outside the United States, Kleiner Perkins is a very well regarded with a long history of investment success, so it’s an investment firm located here in the United States. And the fact that you worked there and did some incredible work is so impressive. What do you make of the occasional criticism that crops up on Twitter and maybe other venues that when these large established investment firms begin seeding capital into web3 or crypto projects, that in essence what’s happening behind the scenes is you have centralization? You’ve got all the same people that invested and built web2 investing in the shadows in web3, and therefore, fundamentally, it’s not as decentralized as everyone would hope or perceive.

Roneil Rumburg (00:33:43):

Yeah. So I’ve definitely seen that criticism. I think it is worth thinking critically about, right, because the goal of what all of us are building broadly in this movement and different people have different takes on this, but I do think broadly everyone working in this ecosystem would probably agree that the goal is to be more decentralized than web2 things were and everything else. But I don’t see how fundraising from web3 specific venture firms is any different from fundraising from web2 specific venture firms. There’s still centralization risk created by any form of investment. So if you take that argument and agree with it, then the question becomes a broader one, which is, why take investment at all? Right? And I think the practical reality of the situation is it costs money to build things. And if you can, there’s kind of a key pivotal moment in these projects, I think, where they cross the panopticon, if you will, from being largely centralized to being decentralized.

(00:35:01):

And that usually happens at the time the network launches, right, where you cross this threshold that is you can’t really undo or cross back where the network is launched. It exists and it’s out there in the world. I am super excited to see folks experiment with and explore other ways of doing capital formation that are more decentralized, capital formation being the process of folks collecting capital to be spent building something and then ultimately shipping it to the world. But I do think it takes capital to build things in this space, otherwise, the only people who would get to build are people who are already wealthy. And it either means people who are wealthy enough to afford to pay other people, or at the very least, people who are wealthy enough to go six months, a year, two years without income. And I think that would be really unfortunate.

(00:35:58):

Ultimately, I think that is an even more centralized, worse outcome from any rational or logical perspective than raising money from investors to build things. Because if you can’t raise money to build things, the only way to build something is to already be an incumbent who’s very wealthy. So I think this idea that people can go years and years without making money and putting food on the table and everything else, that is a very sort of privileged view of the situation and of the world that ultimately favors decentralized incumbents. So I was super excited to see how other folks explore different avenues for collecting capital for the purpose of building projects. But yeah, I think in today’s world and today’s landscape, I don’t see significant issues with this happening right now. And it’s a free market. So as better form of mechanisms for capital formation come along, if this argument proves out to be true, the one that this is creating centralization risk and centralization problems, the free market aspect of the space will determine that as folks find better ways to form capital that don’t create that risk.

Nick (00:37:23):

So Roneil, let’s talk about Audius. That’s why we’re meeting here together today and I’m super excited and I’ve been waiting a long time to interview you, so I’m very grateful for your time and your willingness to join the podcast. For listeners that may not know what Audius is, can you give just a brief elevator pitch on what Audius is and what the vision here is for the project?

Roneil Rumburg (00:37:43):

Yeah. Audius is a music community and discovery platform that puts the artist in control. And that puts the artist in control is really what we see the key differentiator being between Audius and a lot of the other music streaming and fan engagement tools that exist today. The way Audius achieves that is by being fully decentralized and Audius uses a number of crypto native types of infrastructure tooling to enable that, including The Graph and Ethereum, Solana and IPFS. So that combination of enabling technologies is kind of allowing this community operated commons for music distribution to exist. Where Audius is at today, the company that I work for could go away tomorrow and all of this would keep working so long as there are folks in the community that still want to use it to share things and listen and everything else. And I think that’s a really important kind of differentiator.

(00:38:43):

So today, Audius serves about 7 million fans on a monthly basis and that fluctuates from month to month. Around 250,000 artists and over a million tracks now shared on the network. So it’s been a really fun, wild ride over the two and a half years since the network launched and the project started in early 2018. Actually, around the time The Graph did, that’s how we got to know them. I got introduced to Yaniv Tal in 2018 when they were just getting started. We were just getting started because we shared an investor at that time and got to see everything happen thereafter.

Nick (00:39:21):

So Audius puts the creator at the center. By way of contrast, what traditional music industries got somebody else at the center and therefore a solution like Audius may be necessary.

Roneil Rumburg (00:39:34):

Exactly right. The existing tools for distributing content really are not, they aren’t free markets in any sense of the word. Spotify controls pricing and access to content and also controls, from the artist’s perspective, the ability to reach fans that they have. And Audius is not directly competitive with Spotify, actually. We’re going after a very different set of use cases and monetization mechanisms and everything else. But Audius is premised on the foundational belief that artists should have control of their fan base and should be able to do with that what they wish. They should not be subject to and beholden to the whims of these intermediaries deciding that, oh, we’re not going to let you message your fans anymore. We’re not going to let you see how they found your content, how much they’re listening to it. And you actually don’t get that information through existing channels today, which is even crazier.

(00:40:35):

That’s a whole nother rant and direction of conversation we could go down. But suffice to say, the existing kind of music industry power structures don’t allow for really any control whatsoever at the artist level of what experienced fans are getting when they interact with them. And we believe, as a result of that, folks are leaving a lot of money on the table. This idea that all fans are the same, that all fans are worth the same amount to you as an artist. Those are ideas that we seek to challenge and demonstrate that there is space for segmentation of your fan base and there is space for exploring the different price points and price levels that fans might be willing to engage with you, so long as they’re getting some value proportionally back in return for that, proportionally from their perspective.

Nick (00:41:30):

Well, I’m a user of Audius. I love the interface and I love the tool and ways to engage with it, but one of the things I’m kind of passionate about the project myself is I think it’s this really great analog or example, if you will, of what web3 intends to do. So you mentioned there it removes the middle person. Most listeners of this podcast or any other podcasts know that web3 is all about removing unnecessary intermediaries, Audius is doing that, but number two, it’s putting the creator in an ownership position and kind of making the decisions at the center here. So I really love Audius and what you’re doing there.

(00:42:06):

I want to ask this question about double-sided markets, and for listeners that may not know what that term is, you see these things in Uber, you see these things in Airbnb where essentially somebody creates a technological solution that creates a platform by which the buy side and the sell side can interact more easily and more seamlessly. So in the case of Airbnb, people looking for a place to stay can connect through the technology with people who want to rent out their homes and maybe it’s cheaper or better than a hotel. So question is, do you see Audius as a double-sided platform where you’ve got the trick of attracting artists and creators to the platform as well as listeners and people that want to consume?

Roneil Rumburg (00:42:48):

We do. And I would actually take that a step further, which is to say we actually see Audius being a three-sided marketplace with those first two participants being who you named, artists and fans, the third being the node operators and developers that support this ecosystem. So, there’s a whole nother category of users of Audius that are participating to support the network and to get paid by the network in exchange for that work that they’re doing. And these three groups have to work symbiotically with one another for the whole ecosystem to thrive and to grow. But yeah, it’s a marketplace. That’s what Audius is. The thing being sold across that marketplace is engagement with an artist. That’s effectively the product being sold and different forms of engagement, having different price levels that artists can determine at their will.

Nick (00:43:50):

So let’s talk a little bit about revenue model and the money, right? And I only bring it up in that tongue in cheek way because a lot of the criticism that can be heaved at traditional music is that it robbed the artists and the financial mechanisms were broken. In the case of Audius, how would you explain in contrast to what I just described how the monetization for artists and these other two components of the platform work together?

Roneil Rumburg (00:44:18):

Yeah. So Audius is a marketplace like we described, and the lifeblood of any marketplace is exchange of value for services effectively. In our case, that’s users paying artists to access different ways of engaging with that artist, whether it’s listening to exclusive content, in the future, contemplated things like buying forms of digital merchandise, things like that. But ultimately, it’s an exchange of value from the fan for something back in return from the artist. And that payment flow happens in either the Audius native token or in USDC. And the network’s kind of agnostic to both. The Audius token is not really meant to be a transaction token, it’s really more meant to be backing that node operation side. And then from the artist’s perspective, it grants access to various features on the network that consume resources from the network. So that’s the way that things can be sustainable there.

(00:45:25):

And then the token governance network as well. So all of the creators, node operators, other participants that are actively staking the token have governance power over how Audius evolves and changes and what it does over time. So yeah, putting all these pieces together, the flow of money is from users purchasing things from artists on Audius. And 10% of every one of those payments gets captured by the folks who are staking and actively engaging with the tokens. So that’s how those folks operating servers and doing things like that can sustainably operate long term. So in the Audius ecosystem, 90% of that revenue is still going to the artist. If the artist chooses to opt out and self-host their own content on a node that they themselves run or they stake with their tokens, they are also participating in that side of the revenue. So that 10% is sort of the defacto if you want to push responsibility for managing your stuff onto the network, but there is a way to opt out of that.

(00:46:41):

So what that means is that folks who are helping to operate the network are capturing 10% of the payment flow that happens across it. Artists who are engaging on Audius can earn 90%, unlike the broader standard in music, this is pretty shocking when you say it, it’s about 12% of overall revenue making it to the artist. So music generates in aggregate $42 billion a year and 88% of that money is going to these various intermediaries that sit between the artist and the fan. So our perspective here was Audius is kind of the logical conclusion of disintermediation, which is that there is no company, there’s nothing sitting in between the artist and the fan.

(00:47:31):

Audius is effectively a set of tools that the artist can use to design their own monetization structure. And if they want to use the network’s kind of support in that, they can give up 10% of the revenue they generate, or if they don’t, they don’t have to. So that fee is being paid in proportion to resources consumed. If you don’t consume resources from Audius, you shouldn’t be paying. So that’s a lot. There’s a lot to unpack there, but that’s broadly how the flow of money works in this kind of view of the world.

Nick (00:48:07):

I’d be really curious to know, and I don’t want to assume any of it’s easy. But of this building a three-sided marketplace where you’re trying to attract artists, you’re trying to attract listeners and then node operators, which one of those has been the most difficult from your perspective where it’s like, oh boy, we’re taking care of these other two, it’s still a lot of work, but this third one, it’s a hard one. We got to figure out how to bring more of them to the platform. Has there been a case like that?

Roneil Rumburg (00:48:32):

Yeah, so we always think about and operate with the frame of reference that’s artist focused. So we always spend the most time on the artist prong of those three. And yeah, maybe this is a reductive view of the problem, but it allowed us to simplify the problem to a level that we’ve made meaningful progress against it, which is that if you can make artists happy on Audius and give them value and get them excited, they will do things on Audius that lead to fans coming. So we actually really spend zero time actively trying to solicit fans to Audius. We do look at metrics and stats around where their engagement is and then try to make that better and bigger and stuff like that, but we really don’t try and solicit them directly. Our thinking is that the more time artists spend on Audius and the more things they share here, that will manifest more fans listening and that’s largely played out to be true.

(00:49:37):

The node operator side is actually quite straightforward in that it’s really a simple economic question for them. And so long as the kind of economics around being a node operator are sustainable, number one, and number two, sufficiently interesting for them to allocate personnel to this, it’s also pretty straightforward. There’s a deterministic way to figure out how to make it make sense there. And there’s a robust enough ecosystem of node operators in the crypto community that was also pretty straightforward. So yeah, it’s always the artist. So the artist is the hardest, from our perspective, but that’s partly of our own design, I guess, is that we focus there.

(00:50:29):

The value props around Audius are generally artist focused. Audius in isolation isn’t really better for fans. It’s better in that they can access different things than what they get elsewhere, but ultimately, the existence of different things being experiences and otherwise on Audius comes from the artist. The artists have to have a desire to offer things on Audius that are meaningfully differentiated from what they offer elsewhere. And the way that we really try and think about and achieve that is by making sure that a fan on Audius is worth more than a fan on any other platform to the artist. And if the question to that answer is an affirmative one, it always makes sense rationally for artists to want to bring fans here and to build a fan base here because their work will be rewarded proportionally more on Audius.

Nick (00:51:54):

You mentioned earlier that you first became aware of The Graph in 2018, I believe you said, when Audius was getting started and you met Yaniv through a network connection there. Do you remember what your first impressions were of The Graph as you heard about the solution it was setting off to address? And did it become clear to you at the time how Audius might plug into that and need to leverage that solution?

Roneil Rumburg (00:52:16):

Yeah. So it made sense to me immediately, and it was really exciting for me, right, because we were kind of architecting the first version of Audius at that time. And The Graph filled a very large hole in our kind of architecture design for the network. There was a feature or a few features we were missing at that time from the first version of The Graph that precluded us from being able to use it right away. But as The Graph as a product matured, it ended up being a really great fit for us and we use it pretty actively. If you ever go to dashboard.audius.co, that’s all pulling from a set of subgraphs that are curated. And then we were also fortunate to get to be among the first to use the decentralized Graph offering as well. So all of Audius’ Graph engagement happens via the decentralized. As long as a year and a half ago, I think engagement on Audius was happening via the decentralized product.

(00:53:22):

So yeah, we’re big fans of The Graph. We’re very excited for the Solana support that’s promised and forthcoming. That’s a huge unlock for us. So Audius does still have to maintain its own native indexing workflow and structure. That’s not something our company maintains. It’s run by the network, but because there’s data on Solana that needs to be indexed and things like that. And similarly on the content side, there are a number of things in Audius. Like I mentioned, there are these nodes that are hosting content, metadata, all of that. The Audius network effectively acts as a crude CDN that’s built atop IPFS. So it’s pinning and keeping content available on IPFS that folks can pull. But it’s not our ambition to be either of these things. We’re a music distribution and community engagement tool chain, right? And we’re very excited for the day that the Audius network doesn’t have to own building and doing these things, but solutions don’t really exist in the market today that would let us get rid of what we have and operate at the level of scale that Audius does today.

(00:54:35):

So yeah, it’s always, and to be fair to all the teams building against those problem sets too, it is far easier technically to build a application specific solution to something like a CDN problem, or an indexing problem for that matter, than it is to build a generic version. There’s a reason why it was very easy for our ecosystem to build out its own indexing flow because you can very tightly scope the problem. Whereas if you think about The Graph, not only do they not have control of what they’re indexing, they don’t have control of how it’s being indexed. When I define a subgraph, I actually get control of that as a developer, which is really cool for a developer. It’s really damn hard for them as a project. So it makes sense that it takes longer, it’s much harder to build generic solutions. But on a long enough timeframe, the generic solutions are going to win out over the application specific ones, with some pointed exceptions to that, right?

(00:55:37):

If you think about Google or Facebook, companies at that scale, eventually revert and they actually bring all that stuff. They run their own data centers, they build out their own Facebooks like jacked up fork of my SQL is like so kind of different from normal my SQL, right? It’s effectively its own database. So there are exceptions to that, right, but I think it’s going to be a long time before anyone in crypto is operating at Facebook scale and would want to throw 10,000 engineers at the problem of being their own better index or better CDN or whatever. So I expect this to be, this is a transitory state, where today it does still make sense for Audius to have some of these functions be native to the network. But in 10 years, I’d be very surprised if they still were.

Nick (00:56:31):

If you had the opportunity to speak to all the developers listening to this, and here you are, you not only launched a successful dapp in the web3 space, but you’ve scaled it, which is a whole other thing. And they were curious, why use The Graph? What’s the underlying logic to even consider doing something like that? What’s your message to them?

Roneil Rumburg (00:56:50):

Yeah, it’s a tool just like all the other tools that we get to build with. It turns out, in this case, there really aren’t viable alternatives to what The Graph does other than build your own Indexer, right? And that’s a lot of technical debt and baggage to take on. If Audius were starting today and didn’t already have the legacy and the depth that it does, I don’t think we would have our own Indexer. In today’s world, I don’t think it makes sense to own that yourself. Yeah. So what The Graph is really, really useful for is being able to answer structured questions about data that exists on chain or in metadata blobs, like you can index metadata that’s pinned on IPFS for example, in our case. So all of that, it serves a very specific and very valuable purpose, which is you can have your users writing all this stuff on chain.

(00:57:56):

But here’s another way of describing the benefit is this Audius dashboard that I mentioned. Pulls all kinds of different on chain stats for how much is staked and who is staking it and what’s going on in governance right now and on and on and on and on. We actually, the first version of that, because one, you want to be able to have fallbacks to be able to query that data directly from the chain if you need it. But two, we wanted to make sure that the contract interfaces during the development process were exposing the right data that would be needed. When that first version of that dashboard launched, doing all of that querying directly from the Ethereum blockchain took on the order of three to four minutes load. It was insane. There were all these fourth and fifth and sixth order questions that had to be answered.

(00:58:52):

It was like, get me all the list of services. And then now that I can loop through the services and ask, get me all the list of people delegating to that service. And then I could loop through all the people delegating and say, how much have they delegated? So it’s like this horrible exponential blowup as more people stake on the network. Clearly it was going to get to a point where it would take hours to load the dashboard and things would probably time out at that point. So The Graph turned all of those queries into a single query, and that query, if it’s not cached, takes on the order of a second to a couple of seconds to compute. If it is cached, it’ll return in sub 100 milliseconds. So it’s quite a step function improvement to that.

(00:59:43):

So that’s just one way of using The Graph, but it’s a tool, right? It’s like a hammer. There’s a lot of different ways to use a hammer. I guess there are some wrong ways to use it, but there probably aren’t that many wrong ways to use it, I guess. So yeah, it kind of expands in your mind what are the things that I can solve with existing tools? That’s how I describe it.

Nick (01:00:11):

Thank you for that description. Super helpful for me as I listen to that description. And Roneil, I want to ask you the GRTiQ 10, these are 10 questions I ask each guest every week. But before I ask those questions, are there any news or announcements that are coming up related to Audius or things that you’re working on that you could share with listeners today?

Roneil Rumburg (01:00:29):

Yeah, so as an ecosystem right now, we’re really focused on fleshing out this monetization tooling that I mentioned and growing the ways that folks can use Audius to make money by charging their fans for different things. So yeah, there’s lots more to come along those lines. There are a lot of different tools in the pipeline that our team is working on some of, there are other folks in the community working on others of, that we’re really excited about. So I know that’s not a super specific answer, but that’s really where our head’s at over the next year or so is to be really focused on helping artists make the most of the fan bases that they’ve now built on Audius.

Nick (01:01:19):

Well, there’s going to be a bunch of links in the show notes, so I encourage listeners that are new to Audius and want to figure out how they can get involved and learn more to visit the show notes and explore it. Like I said, I’m a user and it’s a really great web3 artifact of what the future looks like and what people are building. So as I said, now it’s time for the GRTiQ 10. These are 10 questions I ask each guest of the podcast every week to help listeners learn something new, try something different, or achieve more. So are you ready for the GRTiQ 10?

Roneil Rumburg (01:01:48):

I am.

Nick (01:02:00):

What book or articles had the most impact on your life?

Roneil Rumburg (01:02:03):

Oh wow. That is a good question. Books, I would say, and I think after hearing me describe a lot of this, if any of you read this book, probably, well, you’ll understand why I like it is The Alchemist by it’s Paulo Coelho. I don’t know how to say his name, but it’s just a great, great book.

Nick (01:02:26):

Is there a TV show or a movie that you recommend everybody should watch?

Roneil Rumburg (01:02:29):

There are a lot. Right now, I’ve been slowly working my way through this series, The Expanse, that’s on Amazon Prime Video, which it’s fun. It’s based on a set of sci-fi books that I really enjoyed, but I had never gotten around to watching the show too. And it’s surprisingly a well done adaptation. Movies, what I recommend everyone watch. I think one of my favorite movies is probably Office Space from the ’90s, and I would definitely recommend everyone watch that if you haven’t already.

Nick (01:03:01):

This is a great question for someone like yourself. If you could only listen to one music album for the rest of your life, which one would you choose?

Roneil Rumburg (01:03:12):

If my past listening history is any guide, that would probably be Enema of the State by Blink-182. That’s the album I think I’ve listened to far more than any other album. I don’t know if that’s the thing I’d want to listen to for the rest of my life though. Yeah, we’ll go with that. I’ll think. Maybe I come up with something better.

Nick (01:03:34):

What’s the best advice someone’s ever given to you?

Roneil Rumburg (01:03:37):

I think not to sweat the small stuff, which is an oft repeated piece of advice. But you really have to pick your battles and decide what you care about. You can’t care about everything.

Nick (01:03:54):

What’s one thing you’ve learned in your life that you don’t think most other people have learned or know yet?

Roneil Rumburg (01:03:59):

I don’t know. I don’t think there’s really anything I know. I feel like all of the lessons that I have to share and things that I tend to find myself repeating to folks are pretty obvious. Maybe they’re just obvious to me, I don’t know. But yeah, maybe it’s find that thing you enjoy doing and then don’t stop doing it. Yeah. I think there are too many people I come across who don’t really like what they’re doing, and it’s in your power to change that. You don’t have to just accept that.

Nick (01:04:40):

What’s the best life hack you’ve discovered for yourself?

Roneil Rumburg (01:04:43):

We were actually talking about this before we hopped on. So I don’t usually eat during the day, I just eat dinner. And there’s differing sciences about this, but that’s a form of intermittent fasting that I found to be sustainable compared to doing the two day, three day fasts that some people do. Just keeps my energy levels more consistent. And yeah, I feel a lot healthier because of it.

Nick (01:05:16):

Based on your own life experiences and observation, what’s the one habit or characteristic that you think best explains people finding success in life?

Roneil Rumburg (01:05:25):

I think it’s being flexible with that definition of success. And by that, I mean, if you have your own definition of success, you will very likely be successful. If your definition of success is predicated on the world’s perception of you, you don’t control that. So I know that’s a very meta answer, but the people who I tend to perceive as most successful, it’s not because they are playing the same game that everyone else is playing. It’s because they’re inventing their own game and then they’re winning it.

Nick (01:06:03):

And then the final three questions are complete the sentence type question. So the first one is, complete the sentence, the thing that most excites me about web3 is?

Roneil Rumburg (01:06:13):

Onboarding the masses to a worldview that’s not predicated on this sort of surf relationship that people have with technology generally. We are not beholden to technology. Technology should be beholden to us. And I think web3 is the implementation tool set for, or these crypto native mechanisms is the implementation tool set for achieving that.

Nick (01:06:46):

And how about this one? Complete the sentence. If you’re on Twitter, then you should be following?

Roneil Rumburg (01:06:52):

You should be following Audius Project. It’s the best Twitter account. Can’t recommend it enough. And I’m obviously not biased in that recommendation.

Nick (01:07:03):

And lastly, complete this sentence. I’m happiest when?

Roneil Rumburg (01:07:07):

When I get to build stuff.

Nick (01:07:16):

Roneil Rumburg, thank you so much for your time. It’s been a total joy to speak with you and you’ve been very generous with your time. If listeners want to learn more about you and follow things you’re working on, what’s the best way for them to stay in touch?

Roneil Rumburg (01:07:29):

So I’m not super active on social platforms. People keep telling me I should be. I don’t know, it feels like it takes away from the things that I actually enjoy doing. But yeah, the best way to keep up with what Audius is doing is, like I said, the Audius Project twitter. There’s a Discord community. Feel free to find me there. If you ever want to chat or find me on Telegram, I’m just my first name in most of these things. I wouldn’t recommend following me on Twitter because all I do is repost Audius content. So if you just follow the Audius account, you’ll get it straight from the source instead of just seeing my retweets.