SQD Stream Podcast Ep. 1: Crypto's Data Problem, SQD Strategy & What's Next

01 Welcome to the SQD Stream

Cold open / intro

Joe: We're here today with Dan, the CPO of SQD. I'm Mojo, Head of Marketing at SQD, and we're in Istanbul on the SQD team offsite. We run these every quarter or so, the whole team gets together in one place. Today we're going to talk about Dan's journey, what SQD is, where we're headed, recent events in the industry, and a few sneak previews of what's coming. Enjoy, and come and join the Telegram.

In the room

Joe: Hi Dan. How are you?

Dan: Good, mate. You?

Joe: Very well, thank you. We're here for the SQD Stream, Squidstream, some may call it. We're going to cover a few things today: your background, where SQD is headed, the KelpDAO situation, and what the community can look forward to. First, we're in Istanbul. Tbilisi was last time, now Istanbul. How are you enjoying it?

Dan: It's amazing to get the team together. We're a fully remote, distributed team based out of Dubai, so being able to get everyone in one room, really get into it, and do things like this is incredible.

Joe: We've been meaning to do this for a while. We run community calls on Telegram, but sitting down in person and getting it on film is great.

02 From Founder to CPO, Dan's Journey to SQD

Joe: How did you become CPO of SQD, and what in your background really shaped you for the role?

Dan: The short answer: I spent most of my career as a founder, from leaving university through to around 2020. Building companies, founding companies. Exited one around 2018 and had the chance to choose what I wanted to do next, and I fell down the Web3 rabbit hole. I dove into communities, became very interested in Polkadot. It was clear to me that one chain wasn't going to do everything. I liked the layer-zero idea, and I've always been drawn to infrastructure.

Every company I built was an infrastructure play of some kind, one was a logging infrastructure company for Fortune 500. So I'm always curious about the infrastructure of Web3. People say Web3 is decentralised, but that doesn't mean you don't need great infrastructure. You still need people to stand things up and run them. It's the intersection of security and infrastructure that I find most interesting.

I became involved with a project called Phala, Polkadot ecosystem at the time. Advised them, spent quite a long time there, then took a break. I was looking for another project to advise or get heavily involved in, and ended up speaking to Dmitry, who, as you know, is a real product visionary. He took me to the core problem SQD is solving: it's very, very difficult to get data out of blockchain. Everything is optimised for consensus, for transactions per second, for getting data in. SQD is solving getting data out. I thought: that's a project I'd like to be involved in.

Joe: I didn't realise you had a 2018 exit and then fell down the rabbit hole. I assumed you'd been Web3 the whole time.

Dan: I've been involved in Web3 since I was buying XRP in 2012.

Joe: Really?

Dan: Yeah. Always interested in it, but like most of us I didn't fully understand where it was going. I was always watching it. So when I could choose what I wanted to do, I became more and more attracted to this idea of a bankless, or trustless-first, future. The analogy I always give people: most of us have one bank account. The fraud algorithms are all automated. If you get caught as a false positive, you can just lose your account. And if you lost your account tomorrow, how would you pay your bills, your mortgage, anything? I'm really attracted to the idea of a trustless-first world. By 2018 I had the room to go deeper and deeper. So yeah, I've been watching it for a long time.

Joe: Great pedigree. A Ripple stan from 2012, who would have thought. Mine was Litecoin, actually.

Dan: How did you get into that?

Joe: The Coinbase app when it first came out. My father was right into it. We were big HBAR fans at one point too, had some Hedera merch. Fell down the rabbit hole well and truly.

03 What Is SQD?

Joe: Even people who know of SQD, investors, my best friends in Web3, they often ask, "Joe, what actually is SQD? I know you do data, I know you've got SDKs and tools, but how do you monetise that?" You've always been good at explaining this in a nutshell. How would you describe SQD to someone first coming across it?

Dan: Fundamentally, blockchain is about transactions per second and consensus. It's decentralised by nature, so we have to agree on a set of transactions that go into the next block. And we're pushing ever-faster chains, you see it every day, every week. The Monads of the world. I was at a conference in Cannes recently; that's the thing everyone wants. The main use case of Web3 is finance, we want crypto payments. So we need a block to happen in the time someone is patient enough to wait at a point-of-sale terminal.

If everything is optimised for getting data in, then getting data out becomes a real challenge. Whenever you optimise for one thing, you trade something else off. The entire system is built for writing data onto the blockchain. So getting data out is painful, slow, and expensive.

And yet, if we're going to build a world that relies on Web3, it's not enough to just have really fast chains, we need to understand what's happened on those chains. We need to get the data out.

That's what SQD is solving. Of course we'll have faster and faster chains, but we'll solve the other side. How do we understand what's happening inside them? We either need to re-engineer blockchain entirely so it's optimised for reading, which we just established we can't do, or we build a new layer on top that reads information off the blockchain. That's what SQD is. We make reading data, understanding what happened, who did what, what moved from here to there, easy, scalable, and efficient. We've got some major Web3 players relying on us so they can focus on their products and leave the infrastructure problem to us.

The one-liners

Joe: You've got a few one-liners. I like "all of Web3's data in one place", that's what we recently changed the X bio to. What's your favourite?

Dan: If I'm talking to an engineer, I say "universal read layer." Engineers think in terms of write and read, you write information into a database, you read it back out. So I use that for a technical audience. For a non-technical audience, I'd use your line, all of Web3's data in one place. That's what we're building. Our flagship product, Portal, is a single API to get all the Web3 data you need at scale.

04 Portal, The Flagship Product

Joe: If you had to pick the proudest thing you've shipped, would it be Portal?

Dan: Yeah, I think that's the vision of the company. To get into the challenge for a second: you have what we call an RPC node, that's typically how you'd access data from a chain. That was essentially an afterthought. Something Satoshi added so people could see their wallet balances on Bitcoin. We still have that interface 16, 17 years later. It's very bare bones. It's raw. Not particularly efficient.

So the vision was: can we extend that? Can we build something more elegant that gets you a lot more data in one go, and that's optimised to run more complex queries? You can't run complex queries on an RPC node. That's the entire vision of the company. So yes, shipping Portal last year was a very proud moment.

Joe: I remember DZ, our CTO, saying it's the biggest change to blockchain data since Satoshi. And, "it's raw", that's one of your favourite captions, isn't it?

Dan: Yeah. It's raw.

Joe: Our very own Squid Ramsay.

05 The Web 2.5 Vision

Joe: This week we've had a lot of internal meetings about where we are, our challenges, and where we're going. There's been a lot of talk about Web 2.5 clients and attracting enterprise customers. Without giving too much away on the internal roadmap, could you give us a glimpse of the North Star?

Dan: Of course. One important thing first: if you go to any crypto or Web3 conference, you'll see it's a small world. There are only so many companies to sell to. If Subsquid is going to be really successful long term, genre-defining, we need to reach escape velocity from the Web3 bubble. There are fantastic customers inside Web3, but we need to reach what we call Web 2.5.

So, what is Web 2.5? It's a term we use internally for Web2 companies who are looking into Web3 and want to enter that world. These are some of the largest companies on the planet. They are the most serious and most profitable customers in Web3. A different league with a different set of requirements. They need absolutely rock-solid stability. And so you start to enter enterprise territory, where you need to deliver SLAs that are very, very competitive.

Joe: SLA, sorry, for the community?

Dan: Service level agreement. A contractual obligation to deliver a level of service the customer can rely on. So it's no longer just about the product, it's about delivering stability. We become a key partner. They can rely on us, and when something does go wrong at 2am, we've got an engineer on call to fix it.

It's become very clear in the last year that there's a handful of really significant players in Web3. The more of those we work with, the more obvious it is that we have to provide a very serious, very competitive product, as much about stability as features. As much about accreditation, proving we're as reliable as needed, but doing all of that in a very professional way. Going on that journey then opens the door to working with Fortune 500 banks.

It's a very different world serving enterprise. A different mindset. Several years ago we were working with Web3 startups who'd just raised a seed round, they weren't asking us about SOC 2 accreditation. That's where we see our future, because that's where we can really offer something a lot of people can't. A lot of our strengths start to compound.

06 Revenue Pools & Tokenomics

Joe: A theme at our event in East Denver was consolidation in the industry. There's been a lot of here-today-gone-tomorrow projects. We're now in our fifth year as a company and, from what I know, in a very strong position. Revenue is important and we're in a healthy place. Revenue Pools is something worth touching on alongside Portal.

For people who may not know, Revenue Pools involves locking SQD. Participants earn stablecoin yield that comes from real data demand from our clients. For example: Lambda, a big player in Solana DeFi. They have a portal and a revenue pool. The first pool, Lambda, saw around 5 million SQD locked in roughly four hours, with a 250,000 SQD cap per participant. We were genuinely shocked. So, are you excited about Revenue Pools? The community is asking for more.

Dan: You've been heavily involved, so you know how excited the team is. Personally, it's an important landmark. We launched the first pool, now we've launched another. From a product and technology point of view, the most important thing to realise is that these pools let us decouple the holder of the token from the capacity that's provided.

We can stand up a portal from a technology perspective, we can rent the server, take the codebase, run it. But we don't need to have the 5 or 10 million SQD ourselves to back it. The community provides that to us, and earns some of the revenue that portal generates. It's a key piece of technology, using the capacity of the network while decoupling the two components.

It was on our roadmap for around six months, so it's amazing to see it come out the other side. As we bring more enterprise customers on, and if we have a revenue pool for each one, it lets someone take part in our journey simply by buying the token. They don't need to know how to run a portal. They provide collateral, the customer accesses capacity, the holder earns yield. It's a harmonious, effective economic loop. It's a key part of what we have. As we add more enterprise customers, you can figure out where this is going.

Joe: I have access to our enterprise channels and the team's meetings. I'd be scolded if I started naming names. But I think if the community knew what we knew, they'd be pretty excited. Fair to say?

Dan: Yeah. And one thing we can say, back to consolidation, we've been here for five years, and that gives us an enormous amount of pedigree. When someone from that Web 2.5 world approaches us, the difference between being around for five years and five months is incomparable. We have a track record. We've seen the waves of Web3 and what comes and goes. We're laser-focused on making it easy to read data. Yes, now and again we've experimented outside of that. But we keep coming back to it.

Being around this long, you can start to imagine some of the people in our enterprise funnel. As we learn more about what those customers need, it compounds. It should be an interesting rest of the year.

07 Security, Why You Can't Trust RPC

Joe: A lot is going on in the industry right now. Some bear-market vibes, plus more hacks with the arrival of AI. As we record, there's a recent example I'll mention, Drift Protocol last week, around $200 million. And now the KelpDAO and LayerZero situation. There's a lot of fog of war and details are unclear, we're not pointing fingers. But it's a $290 million hack, and it looks as though an RPC was made to fall over, or a DVN / RPC list was leaked, or someone was compromised.

We're not the cure to this, I'm not saying that. But what can you say about how SQD validates data, and how that's a barrier to this kind of thing?

Dan: I can't comment on the specific case, I don't know it well enough. But we sit at the intersection of security and infrastructure, which is why I'm so passionate about this project. The fundamental takeaway is that you can't trust RPC. When you join a network as a full node or a light client, you cryptographically verify the data. There's a whole separate class of hacks that come from vulnerabilities in smart contract code, but staying on RPC for now: an RPC node is trusted. Nothing is being verified there. You have to trust whoever is running it.

We like to think we live in a decentralised, trustless world, but if you're getting data from any RPC node and you're not running it yourself, you're trusting that operator.

One of the things we're building over time is making sure you don't have to trust RPC nodes. We'll go and get the data for you. We deal with that interface, but we'll go to multiple RPCs. We'll make sure we're not relying on one person. And we run a lot of checks so we know the information we're getting matches what was onchain.

Every piece of information we pull in already goes through six cryptographic data checks, and we're building on that. We want to solve multiple issues: data quality (sometimes you pull from an RPC node and the block is empty because they're having issues), speed and freshness, and correctness. Between quality, speed, and correctness, we want to take those infrastructure problems off your hands.

That won't completely remove the possibility of a hack. But we want you not to have to trust RPC nodes, and we want to get data to you in a way that you don't even have to trust us. We can prove to you. The classic Web3 line: don't trust, verify. That's the long-term goal, and we keep pushing towards it.

08 The SQD Edge, vs The Graph, Alchemy, and Allium

Joe: That touches on something I wanted to ask. Speed alone is difficult. Correctness alone is difficult. Doing both is some task. Is that our edge against The Graph, Alchemy, and Allium? Allium is structured data, so maybe a different conversation, but for Alchemy and The Graph: is speed-plus-correctness our edge?

Dan: Looking at specific players, The Graph is largely a legacy project now. It works well to get started. You can look at what they're building alongside or on top of subgraphs, Substreams is a notable example, because they're hitting their own scaling problems with subgraphs. Alchemy are essentially an RPC provider trying to do that as well as possible.

You're right, our edge is in providing raw data in a much more efficient, scalable, reliable way. We've already made huge strides, but we want to keep improving freshness, getting latency down as much as possible. And we don't just go to a single RPC provider. We'll hit two or three and give you whichever block arrives first. We might run an RPC node in a specific region to lower latency there. Our enterprise clients want sub-second latency. Quality is another big one. And correctness, all those checks.

So our edge: an incredibly reliable, stable product. Customers who themselves need to be reliable, a massive DeFi protocol, say, can rely on the latency, correctness, and quality of their data without thinking about it. Subsquid Network is a huge part of how we do that, because we have thousands of workers across which we distribute historical data. It's a mixture of all three, and depending on the client, one matters more.

But the core message is: you shouldn't have to trust an RPC node. There are some incredible RPC providers out there. But long term, we want to be the ones verifying. We're a Web3 company at heart. Let us do the verification, you focus on your specialty.

Joe: A strong message given current events.

Dan: That's our message all the time. Even in a bull market when no one's worrying about hacks, like the sign on the factory floor, "X days since the last accident." Whatever the number is, the goal is the same. We solve the read side of Web3 data on freshness, latency, and quality. There aren't many companies that build for the long term, but that's part of our DNA. We want to be here in five years still solving these problems.

09 What's Coming

Joe: What can the community, holders, stakeholders, and the team look forward to? We're coming up to May, with Q2, Q3, Q4 ahead. From my side: more revenue pools, more enterprise clients. There's one thing I want to say which I shouldn't, but it's very bullish and should be happening in the next few weeks. Anything else people can look forward to this year, or have I nailed it?

Dan: You've sat in all the meetings this week. We want to focus on doing one thing incredibly well, Web3 data. As I said earlier, we've sometimes stepped outside of that, and we want to come back to our core strength. Even within Web3 data, it's a big space, Allium with structured data, Dune with their platform for structured data. We want to compound and ultimately just be the best at offering raw data. That's what we built Portal for. That's the vision.

We always work to meet our customers where they are. We have fantastic SDKs. We help people build out structured datasets. But if we create an incredible experience for getting raw Web3 data at scale, reliably, that's our edge. This year is about doubling down on all of that. You've seen the roadmaps from SRE, marketing, sales, it's all focused there.

So, more revenue pools, more enterprise clients. It might sound a little dull, but we're doing one thing and we want to do it really well.

10 Community & Sign-Off

Joe: Right, I'm going to throw a few phrases at you. Long live the squid.

Dan: Long live the squid.

Joe: SQD forever.

Dan: Correct.

Joe: 1 SQD equals 1 USD.

Dan: Not financial advice.

Joe: Of course not. We love the community, we're in there every day chatting. If anyone listening wants to come and join: join our Telegram, @Subsquid. You can delegate. You can join a revenue pool, they fill up extremely quickly, so stay tuned on Twitter. Follow the Twitter. Strap in for the ride. This year is going to be one hell of a ride.

Dan: One thing, Joe. If you can get enough interest from the community, let's say some number of likes on the tweet of this episode, something like 10,000, I'll do a livestream and answer questions live. I'll jump on Twitch and people can ask me anything.

Joe: Oh, okay. Just to be fair, Mukam, that's his name, sent over a list as long as my arm. I've tried to cover quite a few of his questions in this. So just to be clear: I have taken community questions and put them into this interview.

Dan: I'll commit to that, if we hit the number of likes you decide on. A couple of hours on a livestream, showing some of what I'm working on and what the business is working on.

Joe: Love it. We'll find a good venue for that. Thank you so much, Dan.

Dan: Thanks, boss.

Joe: Cheers.

End of episode.