GRTiQ Podcast: 49 David Lutterkort

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.

David Lutterkort (00:21):

I think The Graph is to be very, very modest, is indispensable for something like web3 and I envision it as the database for web3.

Nick (00:30):

Welcome to the GRTiQ Podcast. Today I’m speaking with David Lutterkort, a software developer at Gin Node, one of the core dev teams working on The Graph. As you’re about to hear for yourself, David is an incredible talent who’s often credited with a lot of the work that takes place behind the scenes, especially with the hosted service. During our discussion, David talks about his education, his move from Germany to the United States, his entry into tech and web2.0, along with his subsequent move into web3. David also shares a lot of his experiences and observations related to going to work on The Graph, including some discussion about the hosted service in main net and his opinions about web3. As always, I started the discussion by asking David about his background.

David Lutterkort (01:47):

Yeah, I’ve been a software engineer for a long time. I think I’ve picked up the computer bug in high school or even middle school in eighth grade, one of my classmates came back with green and white paper that was really wide and had these weird words on it, go soup and gotor. This was in Germany, so to me, yeah, [inaudible 00:02:10] back then and yeah, he was part of the computer club at the school and I started doing that and decided that I wanted to do it professionally. Went to school, got a CS degree and then during my degree I did an exchange program with my home state and Massachusetts and spent a year and a half at the University of Massachusetts in Amherst. And then I decided that I wanted to go to grad school in the US and then went to Purdue because I figured that would be a great way to spend three, four years in the US and that’s been over 20 years ago. So yeah, I stuck around.

(02:48):

Professionally, I started at a really small web startup. We were doing database back web applications back when that was the hot new thing. We were initially using Oracle and then later tried to move on to Postgres, but it was really hard migrating an existing application to a new database and that startup sort of went down the drains. And literally at the last minute, Red Hat stepped in and bought that startup mostly because of the client list, I think not so much because of the tech. And yeah, so I stayed at Red Hat for a really long time. At some point at Red Hat, I became really interested in things like configuration management, infrastructure as a service and was one of the early contributors to Puppet and because of my contacts there, I then decided to change ship and go to Puppet. Worked there for a while and that’s where I was when I applied at Edge & Node, back then still called The Graph at Edge & Node. And that was like three years ago and it’s been a fantastic time. Yeah, it’s blown my mind how much growth we’ve seen, how much support from the community, how great the team is, everything is really amazing.

Nick (04:00):

I’d like to talk about your role at Edge & Node. What can you share with us about what your role at Edge & Node is?

David Lutterkort (04:05):

I would say a software developer mostly on The Graph Node project, so the centerpiece that takes data from blockchains and runs what you deploy as a subgraph over it and then stores the results in the database and lets you run graphical queries against that stored data. So a lot of that has been decidedly on web3. It’s a lot of sort of fairly classical systems database work on that end of the spectrum. But yeah, it’s super interesting. I also get to play with hosted service quite a bit, which is the largest system I’ve ever dealt with. And yeah, as I said, it’s been incredibly rewarding even though at times about a year ago it was really, really hard. We saw enormous growth and it was hard to keep up with that growth, but yeah, it’s fun.

Nick (04:55):

David, for listeners that aren’t familiar or don’t fully understand what a node is, can you help us understand how would you define or describe what a node is?

David Lutterkort (05:03):

Really generically, a node is just some service running somewhere and then depending on how people qualify it means different things. Like when somebody says archive node, they clearly mean an Ethereum client, something that’s connected to the Ethereum network, pulls down blocks, verifies them, and you can run these Ethereum clients in different configurations. Depending on the configuration, you call it a full node or an archive node, where an archive node basically keeps the history of everything that’s happened on the chain and you can have it do some more things that we actually take advantage of.

(05:35):

On the other hand, graph node is something that is part of The Graph protocol, something that we supply and it’s sort of the centerpiece of what we do in that it transforms data from the blockchains into a form that can then be queried with GraphQL and we often make the distinction between an index node and a query node, but that’s really just a different configuration of the same piece of software where because the workloads are very, very different and the demands it puts on the system that’s running graph node are very different on whether you use it mostly for indexing, pulling data off the chain or whether you use it for responding to queries like taking your GraphQL query, going to the database, finding data for it and sending it back to you.

(06:19):

So in a lot of deployments it makes a lot of sense to separate these two roles onto different machines to make sure you get the right performance and one process that the somewhat slow running index node doesn’t get in the way of responding to queries quickly.

Nick (06:35):

Again, David, for listeners that might not be familiar with some of the terminology we’re using here, how would you describe the difference between the hosted service and the main net?

David Lutterkort (06:45):

In a nutshell, the hosted service for us was a stepping stone towards the network. The vision of The Graph has always been to be a fully decentralized, you can say in a way, database for blockchain data. The hosted service was sort of the stepping stone in that we tackled one part of the problem, which is getting data out of blockchains and sticking into a database for querying that we tackled that without having to worry about economics in a network and things like that. And having these nodes runs sort autonomously, responding to signals they get from blockchains to decide which data to look at and which data to prepare for queries.

(07:24):

So because the hosted service launched about three years ago, it’s gotten pretty big. It’s where most of the people today run their subgraphs. It’s very convenient because it’s, as the name says, hosted service software as a service offering, but with the launch of the network, the focus is definitely shifting towards that. So over time, more of especially the production workload, the things that people depend on for making important decisions about trading and what’s going on in networks will shift to The Graph Network, which is decentralized and which has an economy behind it, which has various mechanisms for the various roles that are involved and the hosted service will become much more of a development platform, something that’s convenient to use but not really for production traffic.

Nick (08:44):

Thank you so much for that overview of your background, your education in computer science and this move from Germany to the US. I would like to now shift our attention a little bit to crypto and how you became involved. Can you take us back in time and tell us when you first heard about crypto and what some of your initial thoughts were?

David Lutterkort (09:00):

I think I heard about Bitcoin in particular a long time ago and many, many people even today sort of dismissed it as this weird fairy mechanism that doesn’t really make any sense. Yeah, I was affirmed in my beliefs, I think it was during the big bull market of 2017 where I met some people who were telling me everything about these weird coins that they were mining for and it didn’t make much sense to me what they were doing. So I put anything crypto aside and looked at it a bit suspiciously, and then I think in the fall of 2018 I figured that maybe I should have a more serious look at it and not just go by what you read in the headlines on Hacker News or something like that. I started to watch some talks and YouTube, one that sticks out in particular was one by [inaudible 00:09:54], the CEO of Parody, and what struck me about that was that she was clearly still angry about all the Snowden regulations, NSA, all those things and that her answer to the threat that the centralized information brings with it because it’s a big target for anybody who wants to know things about people that they probably shouldn’t.

(10:16):

Her answer to that was decentralization, and looking more into it that was what really struck me that crypto is still used as a get rich quick scheme and sort of the scammier parts of the world, but there’s an earnest and challenging and interesting effort to build the technical foundation for something that is inherently not centralized, that inherently avoids a lot of the use features of the internet as we know it today. And that was the thing that really got me interested in crypto in general. There’s also really interesting book by a guy called, I think, Kevin Werbach, where he talks about how crypto is sort of a completely new way to establish trust amongst parties and he goes through the traditional ways to establish trust, a lot of which just comes down to state power, that the state has a monopoly on violence, which in a lot of cases means it can take your freedom away in the extreme.

(11:16):

And that crypto is a very different mechanism where ultimately you don’t have to rely on state power and you can establish trust by trusting economic incentives or economic penalties if you know that an attacker will lose more money doing the attack than they can gain for it. That sort of gives you reassurance that things might work out. And in the book, he goes through a few examples of where this is really beneficial, this different way of establishing trust amongst parties that might not know each other or usually don’t know each other. And so all that theoretical background led me to look at what companies exist in this space and I’ve known Nena, she was on your podcast a little while ago, a long time and yeah, I knew that she was doing something with crypto, so I talked to her a little more about what exactly she’s doing there and found the problem very interesting and agreed that what The Graph protocol is working on is a really, really important building block for web3 decentralization and building this sort of new platform, the devoids, all the misfeatures of the current internet.

Nick (12:29):

Well, I want to stop and focus on something you said there for a minute here and which is this dichotomy between viewing crypto and projects in the space as speculative investments versus this idea that this is technology, this is infrastructure and the future of technology. Can you take listeners through, given your background and your shifting perspective, how you were able to see that difference and why you think that’s an important distinction for somebody to make?

David Lutterkort (12:54):

Yeah, that’s a really good question. I think part of it is just a distortion that you get from looking at what gets reported about crypto. Of course somebody is stealing $50 million somewhere or a scammy ICO makes much, much better headlines than mechanism design where somebody comes up with a clever way of doing roll ups. I think it’s also the fact that it’s so entangled with money and value and all that sort of distracts from the mechanisms and the ideas around basic collaboration, establishing trust and all that, which at the end of the day can get pretty boring and there’s a lot of people I’ve talked to whose eyes glaze over when I go into details, can’t fault them for that. There’s also the effect that I’m old enough to remember when the internet was all about porn and stealing music. There’s also this natural tendency to dismiss new technology as illegitimate and take the worst excesses and say that’s the typical thing that this is good for. And in the case of the internet, we’ve seen legitimate uses and it’s like an integral part of our everyday life now.

Nick (14:03):

I’m always interested in asking individuals like you who have moved from web2.0 To web3, how they characterize that experience. Was it a huge leap? Was there stuff that you recognized in systems or architecture that maybe you were already familiar with? How was that experience moving from web2.0 to web3 for you?

David Lutterkort (14:21):

In my case, it’s a little funny because when I started at The Graph, I was convinced within six months or so I’d have this whole blockchain, web3, whatever thing internalized, I would know this really, really well. Turned out that what we needed was a lot of really almost old school database work, systems work. There’s definitely a big component around not just decentralization but also distributed systems that creeps into everything that you don’t necessarily find as much in other areas of CS or software. That has been the biggest change in my day-to-day work thinking about it’s not just doing this thing once in some system, it’s that you have to build a system that will do the same thing no matter who runs it. That in the case of graph node, that means if I start indexing a subgraph and you do the same thing, we need to come to the exact same results bit by bit. That’s definitely a very unique way of approaching software development and approaching the design of a system.

Nick (15:26):

Given what now having successfully made the move from web2.0 To web3, if you could go back in time and give your younger self any advice about crypto, especially when you first heard about it and you were a little suspicious of it, I guess, what would your advice be?

David Lutterkort (15:39):

The biggest thing would be to be more open-minded and look more into the details of this than rather than just dismissing it because you read the headlines. I think a lot of these cryptocurrencies have been around for a while, but it took that much time for the underlying technology, but also sort of the community and people to start really thinking about what the possibilities are of this. And I think the shift from, this will probably get me haters, but the shift from Bitcoin to Ethereum is very indicative of this. And Bitcoin saw the placed the trail for just digital money, which was a gigantic achievement. And then Ethereum sort of took this all a step further and said, “What if we could build a world computer, a computational device that does computations independent of where it runs of any individual provider?” And in the process added a lot of utility with smart contracts, which is sort of fundamentally novel idea.

(16:42):

Having little programs or big programs in some cases that just run, that don’t depend on any individual party keeping them running as long as there’s Ethereum nodes around the smart contracts will continue working and you can trade money on those exchanges that the smart contracts enable, feed your crypto kitties, whatever it is. And then I would argue we’re still in a phase where we are trying to fully understand how this technology can be used well and the opportunities this technology opens up, but it will take a lot more effort by a lot of people to get there. But I’m really, really hopeful and the development over the last 10 years or so has been incredibly impressive.

Nick (18:44):

I’d like to ask you about that conversation you had with Nena and as you said, Nena’s been a guest on the podcast before. She’s a friend of yours and you mentioned she introduced you to The Graph, so if we can go back in time a little bit here, do you remember what she said about The Graph or Edge & Node and the problem they were working on and what it was that sparked your interest?

David Lutterkort (19:03):

What struck me from that conversation I had with Nena was that the founders of Edge & Node, Yaniv Tal, Brandon Ramirez and Jannis Pohlmann had clearly identified a piece of the crypto ecosystem that was totally underserved. They had tested that hypothesis a bit by just doing consulting with different projects where they were basically building the same system over and over, which was sort of a protograph node or at least in spirit, similar to a graph node. And I found it fascinating to see how not having this piece, to not fill that gap that The Graph protocol is filling is an impediment to the whole ecosystem because it makes developing decentralized applications so much harder because you need to get into things that are so far away from the actual purpose you you’re trying to serve. Imagine if every exchange besides deploying smart contracts, also had to figure out how to extract data about those smart contracts from the chain and then how to serve queries for that so that people would actually have a front end to interact with those smart contracts.

Nick (20:13):

David, this idea has come up on the podcast several times before and I always like to take a minute and just explore it a little bit and it’s this topic of how the blockchain is fundamentally difficult to index and query, that it wasn’t necessarily designed or created in a way that you can easily extract or query data from it or information from it. So can you help listeners better understand this and walk us through why this is a case?

David Lutterkort (20:37):

There’s a few factors that make this hard. For one, the data you’re interested in is intermingled with all kinds of other data. Just because you’re interested in say some defi contract doesn’t mean that all the data you’re going to see from the blockchain is related to the defi data. There’s lots of other things in there. Also, because storage on blockchains is pretty expensive, the data usually doesn’t come with any explanation as to what it means to just get a bunch of buys and text strengths and things like that, but nobody really tells you that this is the sending address and this is the receiving address and this is the amount was exchanged, and so you need to add something to just the raw data from the blockchain to give the data some meaning, to put it in the context of your application, whatever it is you’re interested in.

(21:28):

And then there’s lots of more pedestrian problems. blockchains are huge, there’s terabytes of data to sift through, so there’s definitely a scale problem in there too. And then all these things take together just make it hard to just query blockchain data. The blockchain data is also not in a form that’s very good for querying because that’s not the job of the blockchain. The job of the blockchain is to record things that happen and the data is optimized for that. So all these things tend together make it hard to an application that quickly displays the things that are of interest to you and ignores all the rest that’s not of interest to you.

Nick (22:11):

Going back then to that conversation you had with Nena, you obviously decided to make the move, you moved to Edge & Node, which is a core dev team working at The Graph. What can you tell us about your experience in those early days after just joining The Graph? What are some of the things you observed and what was that experience like for you?

David Lutterkort (22:27):

As it happened, and I had no idea what that even meant when I started, I started two or three weeks before Graph Day, which was our one-day big conference in San Francisco where the team, everybody showed what they’ve been working on and the vision for the whole project at a theater here and it was really, really interesting to meet so many people even though I had very little context being really new. But one of the things that was sort of launched at that Graph Day was the hosted service and at that point I also had very little idea of what that meant other than it’s some system running somewhere in Google Cloud. I very quickly learned what that meant because we kept having all kinds of performance problems. People were complaining that this or that is slow and that was really just an outgrowth of how fast things were taking off.

(23:24):

They were luckily not taking off so fast that we couldn’t fix the plane while we were flying it, but it was definitely a case of fixing the plane while we were flying it. We were probably on our second, I think actually third iteration of sort of the storage back end, how we store things. It’s always been Postgres, but we’ve changed how we use Postgres very dramatically over the last two years or so based on what we’ve learned from the demands of the hosted service. So I would look from a technical point of view, the hosted service was invaluable in learning about the performance characteristics of graph node and I’m confident that it’s given us a head start on the demands that indexes will face just by the virtue of being very, very big and forcing us to address a lot of those performance problems that we’re actually in really good shape to serve traffic fast on the network.

Nick (24:22):

David, you mentioned that the web3 stack, and I always like to take the opportunity to ask this question, which is how do you envision The Graph fitting into the web3 stack and how important do you think a solution like The Graph is for web3?

David Lutterkort (24:35):

I think The Graph is, to be very, very modest, it’s indispensable for something like web3 and I envision it as the database for web3. Where right now we’re very focused on making just the data that’s stored on-chain itself accessible with graph node and we started with Ethereum, we’ve recently launched NEAR, there are other non Ethereum networks in the works and I think there’s, at least around some of them there’s been announcements and somewhat talk, but we’re also looking very, very seriously at indexing data that’s not directly on-chain, data that’s stored say in IPFS or [inaudible 00:25:15] some of the storage networks. And I think with that, The Graph will grow into this decentralized database for all the needs that decentralized applications have.

Nick (25:26):

Another theme that’s coming up on the podcast and something I want to explore with you, which is this relationship between web2.0 And web3, and it’s probably reflective of my own education, but there was a time where I framed the discussion and I think a lot of people still do, that it’s a zero-sum game, web2.0 wins, web3 loses or vice versa. But the more I learn about it, the more the conversations I have, I realize that there’s probably some coexistence that needs to be accounted for here. So how do you think about that relationship between web2.0 and web3?

David Lutterkort (26:00):

Yeah, my gut feeling is I think that two of them will coexist and it’s more a matter of emphasis and sort of eyeballs how one grows versus the other. I’m definitely thinking that web3 will over time become very big without web2.0 ever going completely away. There’s nobody who’s going to turn off the switch for web2.0 at some point, but just as there’s social media platforms that we all joke about, they still exist but nobody really cares. I think that that’s more the trajectory. I had a colleague at a previous job who worked for the British government and he liked to tell the anecdote how in one day he went to the 20th birthday party of a mainframe and then to a talk on Kubernetes and I think that’s the model for a lot of technology development.

(26:53):

Mainframe business is still big and still going strong. Not a lot of people focus on it anymore, but it’s still exists. Kubernetes is in sort of the systems operations space. The big thing right now, and I feel that that’s going to be a similar trajectory for web2.0 and web3 and I think the criterion to measure web3 by is really how much does it help address privacy concerns of people? How much does it really fulfill that promise of decentralization for people? Even if some people, or for some things people still choose to stay in the web2.0 world, do they have reasonable tools available to say, “No, this is too personal, this is not something I want to have in a centralized system. I’m going to use a decentralized system for that.” Even though in some cases they might still have friends on Facebook that they every once in a while might message or something like that.

Nick (27:48):

Another thing I’ve done on the podcast when talking about web3 is I ask guests how they think or envision web3 will impact the lives of everyday people, people that aren’t necessarily involved in the debate, they’re not working in technology, but they get up each day and they interact or engage with the internet and in some cases it’ll be web3. How is their life any different?

David Lutterkort (28:10):

I think for those people web3 is something that should give them peace of mind that what they read about the dangers of centralization or they just find it icky what some of those companies are doing when they collect data about their personal lives in greater detail than anybody can imagine, that it gives them peace of mind that by using this other tool to achieve similar purposes, that they know that these other things that are important to them are taken care of so that they have a way to stay in touch with friends, to post pictures of birthday parties or share their thoughts on politics or whatever with the world that they know they can do that, but also have a good understanding that this is not something that will be mined for nefarious purposes.

Speaker 5 (29:01):

Hey [inaudible 00:29:09] Stefan from The Graph Academy announcing The Graph Academy Grants Program, a program that is intended to foster the growth of our community driven platform, The Graph Academy. We are offering grants to contributors to incentivize the creation of high quality educational material. If you want to apply simply visit thegraph.academy/grants. That is thegraph.academy/grants. You can also find more information in the show notes.

Nick (29:51):

Coming back to your role at Edge & Node, are there any upcoming projects or initiatives that you’re working on that you can share with us?

David Lutterkort (29:56):

Yes, absolutely. We have the big focus right now on increasing indexing performance. We’re actually seeing that query performance is pretty good as of course with a system that’s so open to let you formulate your own queries in a lot of ways it’s always possible to create slow queries, but we can see from our metrics that what people are typically doing, we’re doing pretty fast on the query side. On the indexing side, things are a lot more complicated because there’s a lot more factors in there, but it’s also crucial that you are able to deploy a subgraph and then have data for it that’s current as quickly as possible. So there’s a big focus on improving indexing performance and a lot of it is pretty technical nitty-gritty work. For example, right now when a subgraph executes, it does three fairly distinct steps where it first looks for blocks on the blockchain, that are interesting to the subgraph, then it extracts data from that block and runs the user’s mappings over it and then it takes whatever those mappings output and stores it in the database.

(31:06):

So one thing we’re working on right now is to pipeline these things so that we’re already looking for the next block while processing the current block and then while we are at the same time storing the previous block so that all these three steps interweave, which will help a lot in especially speeding up how long it takes to get to the head of the chain to get you to the point where you have current data. There’s also a lot that we’re going to improve in how we use the database, which at the same time will give subgraph offers much better tools for when they write their subgraphs. Like one thing I’ve in particular I’ve been looking into quite a bit is a lot of subgraphs do a lot of aggregations, like you could calculate daily and weekly totals of trading volumes or average prices or things like that.

(31:56):

And right now that has to be done, it’s fairly pedestrian, you have to code that all yourself. Besides being harder to code than it needs to be. It also obscures from the system. What you’re doing there like graph node doesn’t really understand that what you’re doing is a daily aggregation of some values. So we’re going to add facilities to graph node that let you express much more clearly that you’re aggregating things, which will then also allow us to handle the data much more intelligently and speed up how we’re indexing some graphs and it will also bring some benefits to the query side.

Nick (32:31):

David, that’s a lot of work and a lot of important work and I get the sense that a lot of people in the ecosystem care deeply about performance.

David Lutterkort (32:38):

It’s certainly not just me or just Edge & Node that’s looking at indexing performance. There’s a whole task force of the various core devs that are looking at it from all kinds of different angles. There’s a lot of opportunity to improve indexing performance on a lot of fronts, and so it’s really an integrated effort and I think it speaks to how well the core devs work together, that we actually have this group of people who are all working towards the same goal by different means depending on the specialization that everybody has.

Nick (33:10):

I want to ask you a question about the relationship between the core dev teams working at The Graph, it’s been mentioned several times now you are at Edge & Node, which is one of the core dev teams at The Graph, but there are others. There’s StreamingFast, there’s Figment, Semiotic, and now The Guild. So five core dev teams working on The Graph. What can you tell us about the relationship between all these different teams and these different kinds of expertise?

David Lutterkort (33:35):

In a nutshell, the relationship is very, very close and there’s almost in my day-to-day life, there’s no big distinction between people at Edge & Node or people at one of the other core devs that I’m working with that’s solely governed by who’s the expert in what topic, not so much by who they work for. What I find fascinating is that each of those core devs has different backgrounds and with those different backgrounds also different strengths and areas of focus and it’s incredibly complementary. If you look at StreamingFast, they have a lot of experience in big data and making, especially the ingestion process, the process of taking blockchain data and bringing it into a form that’s easy for graph node to digest. They have a lot of experience there. The Guild has a lot of experience in just GraphQL development. They’re experts in The GraphQL world.

(34:28):

Figment has a lot of experience with a lot of the chains that I haven’t even looked at where they’ve been running clients for them and writing applications on top of them. And Semiotic, their expertise is really in machine learning and economic modeling and all that. And so all these teams are very complementary and I find it very gratifying to have such a wide array of people to work with and to know that no matter what the problem is. There’s somewhere there’s an expert on that specific problem that you can turn to and talk to and bounce ideas off or get advice on something that you’re stuck on. In a lot of ways it doesn’t feel so much like different companies working. It feels more like larger engineering organization with different teams. That’s how close the relationship is to me between all these core devs.

Nick (35:15):

David, I want to thank you so much for your time. This has been a lot of fun to get to know you and better understand a lot of the work you’re doing. For listeners that don’t know, David is an individual within The Graph ecosystem that is highly respected and there was a time where I was asking, “Who should I talk to next, who should be the next interview?” And your name came up many, many times. I’m curious in your role, David, what it is that you enjoy most? What makes you excited about the work you’re doing?

David Lutterkort (35:40):

That’s an excellent question and I think that answer comes in several layers, maybe from smallest to biggest layer, the technical challenge, a lot of really interesting, fairly hard technical problems to solve and we’re not going to run out of them anytime soon. There’s a long list of features and improvements and all that we want to make and all of them are fascinating from a technical point of view. I think that the next layer is probably just the team. The people I work with on an everyday basis are really amazing, very driven, very smart. But at the same time, the thing that struck me very early on about Edge & Node and I think it translates to the wider community, is that there’s very little ego, very little, “I’m the such and such, and that’s why we’re doing it.” Technical problems are always discussed on the technical merits no matter who you’re discussing them with, which makes for a great work environment.

(36:38):

And then finally, of course, there’s the larger journey of where the whole crypto, web3 space will go and all the things that can be built there. And as I said before, I think The Graph is an integral part of this web3 platform in bridging the gap between what’s recorded on blockchains and what people need to see and interact with in the applications in the middle. In my mind sits The Graph and we’re getting pulled in from both directions. Blockchains want to do things differently for all kinds of reasons. Applications want to do things differently from how they’re done today for all kinds of reasons. And as the space grows, these demands will only become bigger and more fascinating, and I find it incredibly gratifying to just be part of that journey and having a very small part in giving birth to something that I think will be amazing in the end.

Nick (37:32):

Thank you. David. If listeners want to learn more about you or follow the work you’re doing, what’s the best way to stay in touch?

David Lutterkort (37:37):

I Think the best way would probably be through Discord. My handle in The Graph Discord is Lutter , L-U-T-T-E-R. That’s probably the best way to get in touch with me. But if it’s graph node specific issues, pull requests, those are all great ways to start the conversation.