Navigating the Permanent Cloud with AR.IO

GM GM. Welcome to what's going to be another fantastic episode of DevNTell. So if you didn't know, DevNTell is a 30-minute podcast held every Friday—or Wednesday in this case—allowing founders, hackers, and anyone in between the opportunity to come on the show and showcase what they built. And today, I'm ecstatic to welcome Dylan Fiedler from ar.io. ar.io seeks to create a decentralized and incentivized cloud network aimed at attracting more gateways to the Arweave network, therefore making the permanent web more accessible to all. So if you stick around for today's episode, you'll see Dylan show us how WayFinder, a client-side routing and verification protocol, makes accessing Arweave content fast, reliable, and verifiable. Alright, let's get into it.

GM GM. Welcome to the show, Dylan. Pleasure to have you on, man.

Hey, how's it going?

Doing well, doing well. Even better now that we're going to be talking all things ar.io, our friends of the DAO and really love what they're building there. And yeah, I'm just really excited to learn all about WayFinder. I've been digging into it a little bit the last couple days, and it looks fairly interesting and a long time coming, in my opinion. Something like that's been needed for a bit. But before we get into all that good stuff, did you just want to give a brief introduction about yourself? Yeah.

Yeah, so I'm Dylan Fiedler, I'm part of the ar.io team working on decentralized access to Arweave is kind of like our broad billboard caption. I've been with the team for just over three years now, so starting from just kind of ideation and what a decentralized gateway network looks like to building out the smart contract and the incentive layer and the observation protocol, and then helping David and the rest of the team actually building the gateway and node software. So we recently launched Mainnet back in February of this early this year. I think we're at 150 epochs of Mainnet. An epoch is just kind of a day of both gateways serving the network and observing each other. And now we're on that next iteration of, like, post-Mainnet: how do we get people to leverage and use the network to unlock all that it can do? So it's been quite the whirlwind. We've evolved as just as much as Arweave and the rest of the ecosystem has evolved. We think of days like SmartWeave all the way into AO. So it's been a lot of fun, learning a lot, met a lot of cool people, including yourself. So excited to be here today and talk a little bit about WayFinder.

Amazing, amazing. Yeah, it sounds like you've been deep in the weeds of all things ar.io, which is really awesome and it's come quite a long way. There's a bunch of different product offerings now versus when, like, I just a few years back, I guess, like you were mentioning. And just curious, like, how did you start in tech? Like how did you get interested in tech? Was your first footing in the door with ar.io?

So I had... I've been like an engineer, software engineer, since college. And I had gotten into it in college. I was actually originally a finance major. I had worked at a few high-frequency trading firms and one head trader that I was working for interning at the time, I was doing a lot of VBA scripting and stuff for their models and their Excel sheets and their trade positions and stuff. And the head trader at the time gave some—what I thought was—interesting insight. He's like, 'Well, I can teach you finance in three months, but what I can't teach you is how these tools allow us to execute and do all these things that we need to do.' So he really implored me to pursue or look into potentially adding on a computer science degree just to complement the finance background that I had. So I ended up doing that, stayed in school an extra year and got a computer science degree. And that kind of really took me down the tech rabbit hole. I was always interested in technology before that, you know, like building computers and just kind of, you know, when there's a new operating system or whatever came out, I was always fiddling with it. So that really cemented like I'm going to actually pursue this and try and find a job in the computer science realm. Post-college, I went to GE Healthcare, large enterprise, and did some interesting stuff with them, kind of like hardware, firmware type programming, embedded systems. That was interesting, but you know, the state of that industry, healthcare in general, at the time was very slow and it's hard to get things done. It was also a large enterprise. So the rate at which I was trying to satiate my interest, I was doing a lot of side projects and things like that, so building mobile apps and just exploring different spaces. After that, I kind of went to a few smaller startups, most in the healthcare tech world. But again, I was continuing to do these side projects. And that's really what led me to like the crypto kind of sphere. I was doing some side projects, stumbled across Arweave while I was working on a project of just like, man, I want to store large amounts of... at the time it was like workout data. I like working out quite a bit and I was using Strava and I wanted to own my own data. So I was looking for a platform in which I could upload that data and like build my own app around it. And I found Arweave to be a solution that provided that. So I'm not sure exactly how I got connected with Phil and Ariel and David at ar.io, but I was kind of interested in the space in general. They had connected and reached out and it's funny, we did an interview early in the process and I think one of the early questions they had, just kind of vetting my experience, was, 'Explain to me how DNS propagation works' or like when you type in a URL in the browser, what happens? That was one of their interview questions. And I look back because that's so kind of relevant towards what we're trying to do with ArNS and understanding those pieces of DNS resolution in our Web2 world and how we're kind of perceiving ArNS can help solve some of those problems, differentiate ourselves from some of those problems. So it's been very full circle since I joined PDS and ar.io of just decentralizing gateway networks, but now this new layer of ArNS on top of decentralized domain names is really kind of where we're striving. So, yeah, it's been a long journey, a lot of fun, kind of have really grown to appreciate the ecosystem and the smart and incredible people working in it.

That's amazing, man. That's quite the journey from finance to healthcare to decentralized storage and the perma-cloud. But yeah, I think you made the right choice. I mean, when I first started going into the weeds of the world of crypto and Web3, like I started with Bitcoin and Ethereum like a lot of other folks did, but then I was quickly attracted to decentralized storage and Arweave in particular because I was like, 'Oh, that's pretty neat.' Like I come from an engineering background as well, I was like, 'Oh, storing data permanently and making it accessible like... it really, really turned on the right switches, so to speak.'

I have a funny anecdote about my Bitcoin history. I think everyone who's been around in crypto has some version of this, so it's not to override anyone, but in college, I had worked a few jobs and I had to pay rent and cover tuition and stuff. And I had owned Bitcoin in whatever, 2012, 2013. You know, it was like... I don't know the exact price, it was a couple hundred bucks. But I had owned some Bitcoins, I was telling my friends about it, they didn't really grasp it, didn't understand it, didn't expect them to. It was like harder to buy then; Coinbase wasn't as, you know, integrated and prevalent. But I remember having to sell—and I actually went and did this a few months ago—I had to sell Bitcoin to pay my rent in like 2012. And it was like... it was like probably a Bitcoin, maybe more. So looking back at just the evolution of crypto, I think everyone kind of looks at Bitcoin as like the state of the world of crypto. And that was my story of like, damn, had I not done that, you know, that would have been worth probably a lot more and I would have been able to pay rent for years, or even buy a place with that money. So, yeah, I've been around long enough to have one of those stories, for sure.

Oh man, that's brutal. But I mean, it's not, I guess, it's not as bad as the 10,000 Bitcoin for a couple boxes of pizza. I think that guy, he needs to get a hug from everybody he meets. Yeah, 100%. 100%.

Well, yeah, I guess without further ado, let's get into WayFinder. So you touched on it briefly, I guess like you said, full circle when you were being interviewed about DNS and all that. I guess for folks who aren't familiar, what is the elevator pitch of WayFinder?

Yeah, it's... the elevator pitch is decentralized access and client-side verification for data stored on Arweave. And that's... there's a lot to unpack in that just top-level statement. But really, what WayFinder does is it's an abstraction layer to everything that gateways do on top of the Arweave network. If you think of Arweave as a storage protocol, its main responsibility, the incentive layers for operators there, are to store data perpetually. You know, there's the 200 years plus of storage guarantees with Arweave. And that's great, that's what the protocol promises. What it doesn't make any obligations to is the... you know, the ability to access and retrieve that data within relatively good UX experiences. You have to do a little bit of work to find that data if you're going directly to those nodes. And that's not to say there's a gap in Arweave, it's just that was never kind of the main promise of Arweave. And that's where gateways kind of introduced themselves as that caching layer and that latency kind of problem-solver, that UX to Arweave, that kind of gateway to Arweave in that sense. So gateways do a lot of things to just help create a UX about data stored on Arweave. And then what WayFinder does is: you can't rely on just a single gateway. You do all this work to be decentralized, you've decided to upload your data to Arweave, so you have some interest in decentralization already. You want that perpetual access. But you can't get that perpetual access straight by going to the nodes; you need a gateway to help. So there's two ways you can do that: you can run a gateway yourself and rely on your own kind of infrastructure, or you can use the ar.io network, which is this incentivized collection of gateways to provide that access to that data. And then where WayFinder sits... so that's kind of like the building bridges to what WayFinder does. WayFinder intends to be that really easy client-side interface for interacting with these gateways. You know, there's 700 plus, I think, on the network right now. You don't want to rely on just one because that creates a point, like a centralized point of failure. You know, arweave.net has stood for a very long time as kind of like the initial gateway that everyone used, but there's just risk in that. So these gateways as a network give that access, and then what WayFinder does is leverage that network in a way that's really easy on the client side, where ultimately the experience for the application, the tool, the NFT, whatever it might be, can still be stored on Arweave, but then you use WayFinder to kind of dynamically route it through the ar.io gateway network. And then it's packed with a bunch of other features around like verification and things like that, because ultimately, you know, you're relying on a decentralized network. As a user, as a client, you should always be conscious of, 'Well, how do I verify that this gateway gave me what I expected?' You know, I got routed here, I don't know anything about this gateway, I want to be able to verify those things. So WayFinder also provides that interface as well. So that's kind of the building block: you have Arweave as the base storage layer, gateways as that intermediate access caching, latency solution, and then WayFinder allows you to unlock decentralized access across the ar.io network.

Interesting, interesting. So it's kind of similar to the IPFS construct, in that they also used to have dedicated gateways people would access the data from, but then in the browser, you could do like `ipfs://` like whatever the CID is. Is it something similar?

Yeah, so IPFS, I think like, you know, you have to pay to pin something in IPFS and you have to kind of... the expectations there is the CID will persist so long as someone has paid for that to be hosted by one of those gateways. ar.io differs because, you know, the transaction will always exist on Arweave. You can then... you can run your own gateway and cache any transactions that you care about yourself. I think that's kind of like an orthogonal but similar situation to IPFS. Where I think, you know, Arweave kind of extends itself beyond that is these utilities that we've built on top of Arweave, things like ArNS. You know, if you want to just pin a transaction ID and make it really easily discoverable, you just give it an ArNS name. And what ArNS names are is just those friendly kind of pointers to those TXIDs. So that's where ArNS can kind of solve some of those like, 'Well, how do I find what I care about?' It's like a simple, you know, ArNS name to point to that transaction ID. And then gateways will just always serve those ArNS names; that's part of their incentive structure is to serve those ArNS names. Gateways also always serve the transaction IDs that are uploaded to Arweave. So it's a little bit different because like everything in Arweave is accessible via a gateway, whereas IPFS, you know, you have to do that kind of pay-to-pin and trust that that's going to persist. So they solve different problems for sure, they're not necessarily straight-up head-to-head kind of competitors, but Arweave is a little bit more if you think long-term or persistence in access—that's where you can get some guarantees that I don't think IPFS has the equivalent. And I'm not an expert on IPFS, you know, I use it to a degree, but I've always found like well, why would I use IPFS if like Arweave is just like what I would consider like a more robust, longer-term solution.

Fair enough, fair enough. Yeah, yeah, definitely. And I guess in terms of the routing itself, is there a particular... are certain gateways preferred when it comes to the routing algo? Because I know you guys have a leaderboard and whatnot, so just curious if you guys have gamified it a bit or not.

Yeah, so if you go to like the network portal and you look at the stats around gateways, you can see some important metrics. You know, there's just like obvious ones of like well, how much is the operator itself staked? You know, it's like a proof-of-stake mechanism where you yourself stake it. You could also have delegates stake to you, and I think several gateways have, you know, like millions of RIO staked because they either have a team kind of invested in the success of that gateway, or it's become critical for their piece of infrastructure or whatever. So there's, like I said, there's 700 gateways, there's a minimum kind of amount of RIO you have to stake in order to participate in the network, just because, you know, you need to be financially incentivized to participate in good faith. And then what WayFinder does is it allows you to say like, 'You know what? I only want to route requests to the top 10 staked gateways because I believe, you know, based on the experience I care about and my expectations for access and retrieval, those top gateways are ones that I think will consistently be high performers for my WayFinder data retrieval requests.' There's other ways you can route within WayFinder. You can say, 'You know what? I really don't care, I actually just want to distribute load across the network so I'm not super reliant on a given gateway to do anything all the time.' So that's where you can say like, 'I just want like a random strategy. I'll confine it to active gateways that have the minimum stake met and, you know, meet some criteria.' And then anytime you make a request via WayFinder, it will say, 'Okay, cool, I'll route you based on this or that or that.' So WayFinder gives you that flexibility of like, when you have 700 gateways, there's a lot of different ways you can choose to route, pick, and choose, and WayFinder gives that expressiveness based on like what you're building and what you care about. And there are some, like, you know, if you just want to set a static list of gateways that you run yourself or you and partnering teams or whatever run, WayFinder allows that too. So it's really the intent of WayFinder is to give you as much control as possible, whereas the inverse is like it's always just been arweave.net. So we're actually like we're on two total ends of the spectrum, and what we're trying to help people understand is like you're over here, you're relying on arweave.net. WayFinder can give you not only just like one increment closer towards decentralization, but also, you know, you can work towards that entire decentralized access layer if you do just want to let WayFinder manage all of the routing and retrieval and things like that.

Interesting, so it's completely customizable from like a client perspective in the requests. Okay, awesome, awesome. And like, there's a great quote, I think Jason Fried from 37signals had like a real long-form post on LinkedIn or X or whatever, where he talks about like the defaults matter, because when you have people or users coming to use your tool or system, you know, 99% of them or 95% of them are just going to use the defaults. So we feel, you know, that's what we're iterating on is like: what are the defaults of WayFinder? So the onboarding process is one, you have trust in it, it works in the way of like you swapped out arweave.net or you're resolving dynamic TXIDs via WayFinder. What are those defaults where you yourself don't really have to think that much and you get decentralized access without having to think about all those customizations? So we've done a lot of iterating on those defaults and we think we have kind of a good, you know, the implementation of WayFinder is like you don't have to think about much, you just install it, you start using it, and you see what it can give you in terms of that decentralized access and that routing mechanisms, and then you start to think about like 'Well, actually I kind of like doing this and doing that.' So we're really focusing on getting those defaults right so that then the customization becomes more of like your taste and preference about how to access things in Arweave.

Awesome, awesome, that's awesome. Like I really love it when you have that flexibility given to you. And just curious, you touched on it a bit, but in terms of client libraries, like how can people start using the WayFinder protocol?

Yeah, and I'll try and share my screen just so I can show the repo. Sure. And folks for folks who are tuning in, as Dylan is bringing up his screen, if you have any questions, please post them in the chat and we'll get them answered for you. So you can see it now, right? So this is just... cool. So I know, you know, I don't want to dumb people down by showing a GitHub repo, but this is like we're still... we have docs in our formal docs portal, but I'll just go over this quickly. So, you know, the initial library is WayFinder core. That's like all the routing logic, that's everything that you would need in terms of just setting up and starting to use WayFinder. What we've added to, and this is feedback from our previous SDKs, is like the WayFinder React library is like: how do I start if I'm just building a client-side React app, you know, like an NFT gallery or something like that? And then the extension is something interesting too, because what we want to see is people understand this routing and how it works. And so if I... I have the extension installed now, and I don't know if you'll be able to see it when I share my screen here. Does it pop up? Probably doesn't. But what I could do is in the toolbar, I'm typing `ar://` and I'm just going to go, you know, to `arns`. So the URL was `ar://arns`. And what WayFinder actually did is it's routing me to a custom gateway. So the URL that I got routed to—I'm going to put it in this search bar just because for display purposes—is `arns.ar.skyeagle.com.tr`. A random gateway in the network. That's the gateway I got routed to when I loaded `ar://`. And I could do that over and over again. And my current settings on WayFinder are just random. So every time I load an ArNS name with the WayFinder extension installed, I get... that routing logic happens and WayFinder says, 'Okay, I'm going to route you here based on... I check to make sure that they can serve that name, they have, you know, a certain amount of stake, so they're going to serve this request for you.' So that's what the WayFinder extension is doing. I can't really show it well in this little screen share, but that's... that's part of what we want people to understand: WayFinder extension is the easiest way to just start playing around with WayFinder. If you're not a builder, you don't want to use it or trust it yet, look at the extension, you know, route to a few TXIDs, see what gateway it takes you to. Look at the UX and the experience and understand like, 'Oh, this is actually interesting, it's helping me understand from a client perspective what WayFinder can do.' And then I get to see all these like this real decentralized access happen because those top-level domain names are changing, but the data being served back is the same and it's verifiable. So that's where the WayFinder extension is kind of like the initial implementation that we like to think is like: how do we just demo this to people? So I'd say that to say like if you're interested in building with it, look at the WayFinder core and WayFinder React. And if you're just interested in like learning more about how it works, look at the WayFinder extension. It's Chromium, so it works on Brave and Chrome and things like that. And you can do some cool things with the settings there on like your routing logic. So that's how people would get started with WayFinder.

Awesome. And how would people install it? Is it through the Chrome Extension store or do they have to download...

Yeah, it's just the Chrome extension store. I can let me see if I can get you a link here, we can share that with folks. You know, we're hyper-focused on defaults so like there's nothing out of the box that's crazy with WayFinder. You're not going to get like all this verification and all these things that we would consider like more advanced configuration. But you can install it just by going to the Chrome Web Store and looking at the RIO WayFinder extension. You know, it's got some kind of details about how it's set up and configured. And then you know, I talked about that like that routing logic. I'll zoom in here a little bit. But it's got fastest ping, so like the quickest gateway near me. Balanced just means kind of random, it goes around based on certain, just making sure that all gateways are serving the same thing. And then that static gateway—like you could even just install a WayFinder extension and plop in arweave.net if you want the exact same experience as using arweave.net, but it's powered by WayFinder as opposed to just relying on, you know, arweave.net URLs.

Cool. Cool. And is this something that you guys have recently launched or it's been around for a little bit?

I think the idea of WayFinder has always been like the culmination of like what a decentralized gateway network means. So now that we have the network, you know, we have like I said 700 gateway operators, what we want to do now is like: let's start leveraging it. Let's get people and clients and builders and users to understand: how do I start using the ar.io gateway network? And we, you know, there's been a lot of interesting conversations with teams who, like an NFT example is, like you know, they encoded the arweave.net URL/TXID URL in the NFT metadata. The concern there is like you know, we don't expect arweave.net to go anywhere anytime soon, but the problem is like that's dependent on a single TLD that for whatever reason could be compromised. What WayFinder could do and where we're kind of helping bridge that gap is like: well, you can keep that NFT metadata the same—obviously it's not going to change—but what you can do on client-side apps is you pop that arweave.net URL into WayFinder and you say, 'Give me an actual dynamic URL for a gateway that I know is still on the network and is meeting the criteria of the network.' Then you just take that URL that was arweave.net before and then you get this dynamic URL and you can perpetually serve that NFT metadata. So that's where with WayFinder we're trying to like just help people understand is like: the step to using WayFinder is actually very low, it just needs people's awareness. So when you say is it something new, it's like it's new in terms of the library and the extension and that's where we're kind of helping people understand like 'Well, how do I start using it?' and we're making those cases and working with those teams who are doing like a lot of uploading and a lot of retrieval from arweave.net. You know, even like a GitHub search of just arweave.net references is like: every single one of those people are potential users of WayFinder because there's just a hard-coded dependency on arweave.net in that implementation.

Gotcha, gotcha. Yeah, and it seems like it's well on its way to becoming a staple here. And we have a question from the crowd. Tolulope asks: 'As a developer, how can we use and extend ArNS and WayFinder code to build their app?' I think, perhaps, what are some example use cases that folks can use WayFinder in their apps?

Yeah, that's a great question. I think, you know, at any point if you're building some app that's referencing resources or data on Arweave, you can use WayFinder. Client-side or not, you know, it could be a backend system that's doing some pipeline ELT process for data. For like frontend apps, you know, if you have an app that's even if your application is uploaded as a perma-website, as what we're calling it, so like it's got a TXID with a manifest. That manifest has a TXID. You can use WayFinder to always load that app. Or what you can also do is you buy an ArNS name, you assign it to that manifest ID of that perma-web app, and then you use the WayFinder extension to just route to that ArNS name. So there's multiple layers there, but you know, upload your site via perma-web app, get the TXID of it. Then you can always use WayFinder to load that application. ArNS just makes it friendly so you can point to it, and you know, you could share that link of `ar://[your_arns_name]`. We're trying to get to a world where the `ar://` protocol is like standard. So that's how you would be able to reference your application and WayFinder would be able to resolve it against all the gateways in the network.

Gotcha, gotcha. Great question, great answer. We got another one from James, asking: 'Will there be a hosted/web-enabled WayFinder service that can link/import the script to pure HTML publishers?'

Um, web service... I'm not entirely sure what you mean by service, but like, you know, WayFinder it is... it will have just a direct JS script that you can inject in pure HTML for sure. So that's already available, it's a bundled minified script where you can, you know, just load that into your app and then use WayFinder. What I will say is like when you're using WayFinder, it is dynamic, it is asynchronous, because right it has to go find those things in the network and then route accordingly. So your request flow for you like, you know, images are a little bit different, especially when you're dealing with like raw HTML. So you'll have to kind of load those dynamic URLs and inject them into your HTML so it's a little bit more JavaScripty if you're doing just HTML. But it's definitely there, it's definitely available, and then Narb I can send you like a link to the actual like JS script that someone appears to inject for a raw HTML application.

Yeah, yeah, for sure. Yeah, and we can share that information in the description of the video. That would be... that would be amazing. And I guess... I guess back on the use case train, just curious: has ar.io seen and Arweave in general seen a big influx of people using the network to store training set models for AI agents? Or AI in general.

I know that's... I haven't looked at the numbers, so I don't have like exact things. I know that's been a bigger use case of like, you know, if you're going to train something and you want immutable data that can persist over time, and then you want to be able to like deterministically understand the inputs that created the actual like outputs of the model. I know that's much more of a use case that's become prevalent. I think AO, the AO Arweave team is having a hackathon in a few weeks around like agents on Arweave and AO. So that's I think kind of in parity with like well how can you build like these autonomous agents and these models that leverage data stored on Arweave to then act in accordance to certain behaviors. So that's a use case that I think is growing; it's kind of ripe for more exploration in builders that I think everyone in the ecosystem is excited to see.

Yeah, 100%. And the reason I bring it up is because we're kind of seeing a little bit of a renaissance in how basically people are starting to kind of trend towards more open-source AI, even like the big names like the OpenAI's and the Googles of the world. They're starting to release open-source versions of their models—not necessarily the top dog models, but some models, right? And then we have this very vibrant community of people who are uploading their models to like Hugging Face and whatnot. And it just seems like having those models kind of be permanently available to people and not being taken down if like Hugging Face goes down or like Docker goes down I think is like a very compelling use case. So it's great to see.

Yeah, and just to speak to the ar.io network too is like: if you think about like censorship or, you know, some country bans the use of a given model for whatever reason. You know, gateways in that country might not be able to serve that model, but gateways everywhere else would, and WayFinder would be able to kind of do that too. So I think that's another benefit of you know, you store these data on Arweave, then you have this decentralized way to access it, so you're not beholden to either the region you're in or, you know, the laws for a given thing because you can build in this really decentralized and perpetual way as the phrase I kind of like to put it. It's long-term.

Well said, well said. Yeah, and I totally agree. Yeah, unfortunately, we've come to time. Just a final question for you, Dylan. If people start playing around with WayFinder and they have questions or they need help with a particular thing, what's the best way to kind of join the ar.io community and get that help?

Yeah, I'd say first thing, you know, go get a primary name. You go to ArNS app—arns.app. Get a primary name, attach it to your wallet address. That's the first thing, because then you know, you have kind of your identity space within the ecosystem. Join the Discord—we have the ar.io Discord, we can send links to that. All of our repos, all of our code... talk about open-source, like we take a lot of pride in open-sourcing everything. So if you're playing with it, something's not working or you're confused or you just have a question, feel free to create an issue or throw a conversation on the discussion boards. And then yeah, like yeah, we're all over X and all the other things, so feel free to just ask us questions. And I'm a big believer in like those conversations in public are better than anything in private. So if there's something not working and you don't like the way you have to interact with something or it's confusing, ask us in a public forum that we can kind of help talk through because we'll take that feedback and like update documentation or, you know, create a sort of material to help other people onboard too. So we want to make it really enjoyable to join our ecosystem and then really delightful once you're in. All the tools and things available to you are really nice and fun and solve problems. Most importantly.

Superb, superb. Yeah, you heard it here, folks. Get on that WayFinder train. We will post the resources for you to do that in the description below. And Dylan, thank you so much for taking the time out of your day to chat with us today. This was an awesome talk and I know I enjoyed it and hope all the folks tuning in today enjoyed it as well. Great. Yeah, thanks a lot for the time, I appreciate being here and yeah, anyone, any questions at all, feel free to reach out. Amazing. Thanks, man, appreciate that. And with that, I just want to wish everybody a very happy Wednesday, happy rest of the week, and we will catch you back here for another great episode of DevNTell next week. Till then, have a good one, folks. See ya.