.jpg)
"The Edge of Show" dives into the bleeding edge of Web3 infrastructure innovation. Co-host Josh Kriger brings you inside conversations with visionaries shaping the next generation of blockchain technology, including Jon Jones of Supra, Rachit Agarwal from Wormhole, Igor Lessio of Eliza, and Gaurav Sharma of io.net. From native oracles and system-level automation to cross-chain communication and AI-enabled agent tooling, this episode unpacks how these tech leaders are redefining scalability, security, and speed in decentralized environments. For crypto developers, institutional investors, and builders exploring high-throughput chains and real-world blockchain integrations, this is your roadmap to what’s actually happening on the ground—and what’s next.
Key Topics Covered:
- Supra’s Native Oracle Integration: Jon Jones discusses how Supra’s fully vertically integrated L1 differs from modular approaches by embedding oracles and automation at the system level for lower latency and better execution in DeFi.
- Automation via "Autofy": Supra's launch of Autofy introduces automated smart contract logic using AI—bringing a new level of user accessibility for automated DeFi strategies directly into the protocol.
- Cross-Chain Comms & Supernova: Supra’s Supernova enables native cross-chain communication, facilitating unified execution layers for developers and users working across different virtual machines.
- Wormhole’s Institutional Bridge Strategy: Rachit Agarwal explains Wormhole’s pivotal role in tokenizing funds from BlackRock and Apollo, emphasizing value accrual, regional growth in APAC, and multi-chain messaging.
- Eliza’s AI Agent Swarm Systems: Igor Lessio showcases Eliza’s open-source AI agent platform with modular plugin support and mobile-first user interaction, aimed at empowering solo builders with smart assistants.
- io.net’s GPU Aggregation Model: Gaurav Sharma illustrates how io.net is democratizing compute access by aggregating unused GPU power from gamers, data centers, and idle machines—driving down AI costs for startups.
Episode Highlights:
"We're the opposite of modular. Supra embeds oracles, price feeds, and automation at the system level." — Jon Jones
"If your automation logic has latency, you risk MEB attacks—integrating it into L1 is crucial." — Jon Jones
"APAC is buzzing. The moment we mentioned BlackRock and Apollo tokenization, the credibility clicked." — Rachit Agarwal
"I built an agent that creates data for my other agents while I’m at the beach." — Igor Lessio
"AI companies spend up to 50% of their capital on compute. We cut that by 70%." — Gaurav Sharma
Transcript:
Josh Kriger: Hi, everyone. Josh Krueger, co-host of The Edge of Show, live at Hack Season in Dubai. It's a bustling week. I'm here with John Jones, the CBO and co-founder of Supra. Good to have you on the show, John. Thanks, Josh. Always great. Yeah, yeah. So it's been a minute. Catch us up on what's going on with Supra, of course. But just for background information, those that don't know, what is Supra all about? There's a lot of L1s out there. I think over 75 now. What makes Supra different?
Jon Jones : Sure. So Supra, we went to market as an Oracle years ago, but Supra is actually a fully vertically integrated L1 with native Oracle. So we have built-in system level Oracle, like for example, price feeds, automation, as well as cross-chain communications. And I can talk about why that makes sense, but we're like, We're the polar opposite of the modular world and thesis where we are fully vertically integrating these stack of Oracle services to make better applications on Super. Specifically, we're spearheading DeFi in the beginning and we're focusing on what we call execution quality.
Josh Kriger: and will have better execution quality when you vertically integrate Oracle services at a system level on the L1 at C. Yeah, so I guess that's like counter thesis to the idea of modularity and sort of giving folks more flexibility with what they need and what they don't need. Does this all-in-one solution service a particular segment of the population, or is it just one of those things where everyone can benefit from it, it's just what do you choose to use?
Jon Jones : Sure. I would say, first and foremost, a lot of these high-throughput chains, including Supra, Apto, Sui, are all-purpose chains. But when you think about it, blockchain is all about the transfer of value. And so from Supra's perspective, we're really trying to spearhead DeFi first. We think that if you can do DeFi well, and you can focus on having the best execution in DeFi, there are other models and other applications that can now also build on Supra that can benefit from the same execution quality and performance for DeFi.
Josh Kriger: It makes a lot of sense. And now you have, I guess, one of your exciting new initiatives is automating some of the stacks. Talk to us a little bit about that.
Jon Jones : Yeah, so that's called Autofy. So it's a new term. Of course, you have automation, DeFi, you combine it, it's Autofy. Something new to add to our crypto lexicon. Yes, and hopefully this becomes something that is ubiquitous. You heard it here first, Autofy. OK, what's that all about? So automation is something that traditionally is in the blockchain space. You have external protocols like Chainlink or Gelato that can do automated services and automation is basically if this then that. So you create conditional logic and that's really powerful for blockchains. Are you using AI to assist with this? Yes, I'll get to that later. And that's interesting because being able to use your intent, like I want to do something and have AI craft and automation, that's actually where retail I think will benefit and use this the most. But what Supra does is we use automation at the system level. So our validators are not only running consensus on transactions, but they're also performing these automated tasks. So you can imagine that if you have one unified system that is ordering blocks, creating a transaction, seeing Oracle data, but also automating a task, it can be done with very low latency. So that alludes to the fact that we have great execution quality. Whereas if you use an external protocol to get that data, there's latency there. Every external protocol you use, you're adding latency and worse execution quality. And why would users care? Well, if you have bad execution quality, you have latency. Someone could, what we call MEB, right? They can MEB you, sandwich attack. Like I think a lot of, like some people in Web3, a lot of people on X might have heard that this guy was transferring like $250,000 in E. And he only got like 12,000 in value because someone front ran his transaction and he lost a lot of value. And that's like an example of bad execution quality. And that happens with added latency. And when other, you know, traders see what you're doing, you know, and see what you want to do, and they basically take advantage of you. So you can have better execution quality when you kind of integrate that into one stack.
Josh Kriger: Nice. So the potential for front-running is heavily mitigated? It is, yeah. Cool. That's exciting. Well, congrats on that. And I guess anything else on the roadmap that you want to make sure we cover?
Jon Jones : Well, so automation is going live on Testnet today. So we've been live on our mainnet. We were the third Movebase L1, like after Aptos and Sui, and Movelang and MoveVM to go to mainnet. That was in November. But now, six months later, we have the biggest testnet upgrade since our token generation event, and that's having automation live on testnet. So automation and also our cross-chain communication protocol called Supernova, that together is just really huge. And so we have already a lot of teams that are starting to test and build it out. But I'm really excited to see the things that they built. One thing, and you did talk about AI, right? And how that might kind of interact with automation, is that we have one project called Dexlin who built this chatbot where you can basically craft a strategy. Like, let's just say I want... hey, I want to be able to, if Bitcoin goes below this price and sentiment analysis is here, I wanna be able to put 50% of my assets into stables and I wanna automate this. And you basically speak your intent to AI and then the AI will then craft using blockchain and logic, will craft your automation and you can decide to hit yes Or you can then speak and say, no, I actually want it a little bit different. But this allows the general user to use these automated tasks that traditionally have been used only by high frequency trading firms, you know, to actually code this login.
Josh Kriger: Well, so it's right. It's basically built into L1. You don't need a separate DAP for that. That's right. Very cool. So. I guess what is the sort of volume of trading where super is at right now and what what coins can folks access through super and for your broader ecosystem? What are you integrated with? What are you looking to integrate with next?
Jon Jones : Yeah, so first of all, like, so we launched six months ago, and our defy ecosystem is really booming. I don't necessarily I did say one name, I don't want to do some too many name drops at the moment. But we started with Move, Move VM. And we were kind of limited to the amount of builders and apps that actually build in the Move ecosystem. That's AptoSuite, now Supra. But we got like about 20, 30 teams to actually commit to build, which is awesome. But we have a whole backlog. of like 100 plus MOUs of companies and builders that want to build on the EVM on Supra. So we're, you know, in the next, you know, little while, soon, we also have EVM, Ethereum Virtual Machine, also live on our network. So on one shared security validator set, we'll allow different virtual machines. So move first and then EVM will be on testnet shortly. So that's when things get really exciting because there's a whole new group of developers and teams that want to deploy specific DeFi protocol on Supra or a DeFi protocol that's already multi-chain or NFT marketplaces that want to integrate with Supra, but maybe they want the EVM integration and not the move integration. So we're giving developers a lot more access to deploy onto Supra without, you know, it's very developer friendly to be able to support many VNs.
Josh Kriger: That's exciting, man. Well, appreciate the update. Always great to see you. If folks want to learn more about Supra, maybe follow you guys on XFOLU. Where do they go?
Jon Jones : Yeah, yeah. Well, Supra underscore labs on X. And you can follow me, John Evans Jones. That's J.O.N. Evans. I'm sorry, I'm not a part of the Josh Club.
Josh Kriger: No, and you're not a UFC fighter, but that's OK. Yeah, but I do have the John Jones. All right. The other John Jones, the John Jones of Web3. You heard it here. Good to have you in the show.
SPEAKER_06: What's up, everybody?
Richard Carthon: Richard Carthon here from the Edge of Show live here at Hack Seasons. And I just got off a really awesome panel with Rajit. We talked about decentralization and why it's important and what's going to be happening next. But we had such a good conversation. He gave such great responses. I was like, man, we have to have a further conversation. Make sure that the world knows more about what you're doing over at Wormhole. So tell us a little bit about yourself and a little bit about Wormhole.
Rachit Agarwal: I appreciate it. It was a great panel and a lot of fun. So, you know, Wormhole is an interoperability platform. Those are many complex words. Let me try to break that down for you. So think about it in a way where there's so many different chains, right? Like you have Ethereum, Solana, Base, Arbitrum. From a user perspective, how do you move funds from one place to another? That's where Wormhole comes in and Wormhole technology comes in, where you can transfer assets from one place to another. That's probably the most basic use case for Wormhole. Over the years, we've built up tech stack to support developers as well. A very simple, basic instance of this is messaging. So if you're on one chain and you want to go to another chain, you need to make sure that both of those protocols or those instances are communicating with each other. So that's one of the things that we've built. Fun fact, Worm was actually the first bridge from Solana to Ethereum. And that's sort of the backstory and how we've grown over the years. On me, I'm the APAC lead at the foundation. I joined about five months ago. I used to work at a marketing agency before that. I was leading the growth team. And I started off in crypto in the VC scene. So I've been around.
Richard Carthon: You have, man. So a well-rounded background and being able to be the very first bridge from Solana to Ethereum obviously is a very big bridge. But I want to lean more into what you're doing in the APAC. So as you're seeing opportunities that are happening in this space, there's been a lot of emphasis in the APAC region. So what's happening over there and why are you strategically focused on that specific region?
Rachit Agarwal: The world out there is wild. You're based in the States? I am. Like, if you ever go to Hong Kong, the crypto scene is so different. Like, OK, so one of the basic differences that I recognize is, you know, in the States or in India as well, like people love to back things that are doing multiple things like, OK, this thing's doing four different narratives and they fit into everything. In Hong Kong or in like APAC, they want to focus on one thing. So in Hong Kong, what I realized was they recognized Wormhole with Interop. And they actually understand what Interop is, which was pretty surprising to me because, you know, like with the Indian accent, it's really hard to say interoperability. So I thought, you know, it would be really hard, but they actually get it. So what we're doing right now is since we're like in this post TGE phase, Wormhole is an interesting part because we have a ton of users. What we realized in APAC is a problem for a lot of other protocols as well is our communication rails there are not very strong. We focus a lot on Twitter. We have really strong Western comms and English comms, but our Chinese comms, for instance, or our Korean comms or our Vietnamese comms, not up there. We've naturally had a community that's really engaged in all these countries. My primary job is to focus on each region individually. Each region will have a different focus as well. You really create incentive structures, community structures, and communication structures so that Wormhole's mission is then portrayed effectively to these things or these community members.
Richard Carthon: I really like that, like it's really important the communication and how it is presented and bridging those gaps and understanding and even like what could make sense in the West may not translate over into the East and everything else that's going. So as you're kind of bridging that gap from a communication standpoint and from a focus standpoint, what's the messaging kind of been like so far?
Rachit Agarwal: Yeah, love the puns by the way, love you using bridging like four times, that sentence. But basically, okay, so Wormhole recently started working with Securitize, well not so recently, we've been working with them for a while, but they recently tokenized the BlackRock fund and we helped them go multi-chain on that fund. Similarly, Apollo Credit, also a big institution, big private credit enabler, they also came on chain with Securitize and we helped them go multi-chain with our technology. That's been a really big narrative focus for these APAC countries. They instantly hear BlackRock, Apollo, they instantly associate credibility. So the crypto credibility problem is now starting to solve with these bigger institutions coming up. where family offices in Hong Kong are like, okay, if BlackRock's tokenizing their fund, maybe I should be thinking about it. And that's the shift that we're slowly seeing in APAC, where the Western legitimacy of like these bigger institutions is coming here. And then Hong Kong government is also pushing forward for that. They're actually going to launch their own stable coin as well. So you'll see a lot of development come out, but it's always been a reputation and credibility thing. I think that's going away. And, you know, platforms like Wormhole are trying to portray that message forward.
Richard Carthon: I think it's awesome that you're seeing so many of the institutional players, whether it be banking, whether it be these big financial institutions who are embracing and being like, hey, no, we're going in on crypto. We're really embracing this. The fact that you're trying to create like Hong Kong's creating stable points and these on and off ramps for people who are trying to enter this space, I think is very impressive. And obviously, Wormhole has been doing that a lot for not just people within the crypto space, but for those who are coming in and trying to go cross chain. More generally, outside of what's happening in the APAC, when you look at what Wormhole is doing at the time they're supporting it, it's 2025. Over the rest of this year, what are some of the things that Wormhole is focusing on?
Rachit Agarwal: For sure. One of the bigger narratives that we're focusing on this year, and this is something that Dan, our co-founder, mentioned in one of his blogs recently, we're thinking about activating P-Switch. And we're thinking a lot about value accrual to the Wormhole protocol. That's one of the focuses. But if I talk about the overall vision, the 10, 20-year vision for Wormhole, we really want to get into this institution narrative. That's the number one focus for us. We're working with BlackRock. We're working with Apollo Credit. We're working with Securitize really closely. We see a lot of the banks, also we recently announced with Mercado Bitcoin, it's like the biggest exchange in Brazil. So crypto is making its rounds around the world. And we're most definitely going to make rounds alongside with it. So the FeeSwitch side, definitely interesting for our community members, the institutional side, interesting for everyone else.
Richard Carthon: So with emphasis going more into the institutional piece, for someone who's listening to this, they're like, hey, actually, I'm trying to break into that space as well. And it sounds like Wormhole could be a great way to potentially help to get into that. So how are these kind of like partnerships and alignments with institutions being facilitated?
Rachit Agarwal: Yes. So. Technically, it's very simple. We have this entity, it's called Native Token Transfer. It's like a pretty simple UI. You and I could launch a token and go multi-chain in like 10 minutes. But for institutions, they need a lot more credibility than that, right? So for them, the first thing that they want to figure out is, What chain should I go on first? All right. Traditionally, it's been Ethereum. We're seeing a lot more conversations where people want to come on Solana and do a lot more of those things. But the first thing that institutions need to worry about is what chain are you actually going on? After you focus on that, that's where a partner like Wormhole comes in where, OK, you've built a base, you've built a home, and then you can go to 10, 15, 20 different chains. The biggest unlock for a you know, like a platform like Wormhole to work with an institution is that liquidity is really fragmented in crypto. And as an institution, why do you want to be married to one chain when you can get liquidity from every blockchain? Right. So it is still recommended for an institution to think about one blockchain first, test the rails, understand what's going on, work with a partner like Wormhole or the Gereon Foundation or the Solana Foundation. get into it, you know, understand how it helps the 24-7 market thing is something that institutions like it's been mind-blowing for them, like how crypto is 24-7. Stress for them. But yeah, pick your base and then think about multi-chain. Think about how you can access liquidity to your products.
Richard Carthon: Awesome. I think that's a really great explanation. Appreciate you breaking that down. For those who want to learn more about Wormhole and potentially connect with you, what are ways that they can do that?
Rachit Agarwal: For sure. You can message us on Twitter. So, AdreWormhole, message me on Twitter, 0xAgarwal. You can actually message anyone from the Wormhole team. Like, you know, internally, we have this joke, like, if someone's not replying, we'll sort of call them out. Like, hey, what's up, man? Yeah. So what happens is if you text someone and then you text me, I'm like, who did you text first? And then we'll call them out. Yeah. So the culture is pretty strong. We're always helping our ecosystem projects and whoever wants to integrate.
Richard Carthon: Excellent, man. Well, appreciate your time and appreciate you being here at AXIS, man.
Josh Kriger: Welcome back to the Edge of Show, live at Hack Season in Dubai at the Museum of the Future, where we're talking about the future. And I'm here with Igor Alessia, one of the core developers on Eliza. It's great to have you on the show. So I've been so impressed with the development of Eliza over a relatively rapid period of time.
Igor Lessio : It was very fast.
Josh Kriger: Let's just start with a little bit of your background before Eliza. And what got you excited about that project?
Igor Lessio : So I am an OG, 2011, but I am an AI engineer, and I am a Web 2.0 AI engineer. So our interface sucks. There is no way to present to a customer an incredible agent that is this ugly. So I was looking for something that helped me fast track the interface for my own agents. And I started in Eliza that was just come out from show. That is our founder. And I start to use it. And then I start to build on it, plugins. And all my plugins, everything I build on Eliza is going to go for the open source community. So I start to collaborate with them. It was very fast. From the first use to the complete collaboration took me just two months. And now I developed a part of the core, not all of it, of course, we are multiple developers. And we are now working on version two. It is better, faster. You have multi, real multi-agent system. Let's call it swarm, but it's a little more complex. Okay. And for the first time, agents, multiple agents in your stack can collaborate between each other just because they exist in the same space and you can control all of them from the same, from one single point.
Josh Kriger: So for those that aren't familiar with this idea of these agents that can do different things, what's the scope of possibilities for these agents? And is this in a technical capacity, or are these consumer-friendly agents too?
Igor Lessio : No, we want to build consumer stuff, so it's B2C. The best use case is that any task that is, let's say boring, okay, but that also is part of our production of content, any content, you can automate it with an agent. and the agent should be your private assistant.
Josh Kriger: Your friend, your concierge.
Igor Lessio : Yeah, it can do basically anything that have a connection. So any API, even the groceries.
Josh Kriger: Well, what are some of the most creative examples of agents that have been built with Eliza that you're just like, wow.
Igor Lessio : Okay, okay, everybody gonna start to build a trading agent because it's crypto. But my own case, I have agent that create data for me, for my other agents. Meanwhile, I go to the beach. So my agent is taking my job, thanks God, okay? And I can control it from the phone. That's why I'm probably the only developer here without a laptop, because everything I can do directly from Telegram at this moment. So it's important that this technology get to the last mile and not stay only in the corporate situation, because every person can have advantage of this. Imagine you are in another country to speak the language, but you need a lawyer, you need a doctor. If your agent can find it for you, you don't have to have problem of communication with anybody.
Josh Kriger: Makes a lot of sense. That's really cool. Yeah, I mean, I use AI to like actually give me a training schedule and recommended like food plans. I mean, there's so many things that you can do, but I think the shift here for folks is from independent query and prompting to automated ongoing tests, right? Like things that you do on a regular basis that it would be great to just outsource to someone else so you don't have to think about it.
Igor Lessio : but it's not someone else, it's your own agent. So your data are going to be more private. You're not going to distribute your things to anybody. Slowly, we're going to go, sorry, to a more modular system where you're going to have a very small entity that live in your phone. So your data stay there and communicate with other agents outside for the task that is not capable of.
Josh Kriger: Pretty amazing. And next, you guys are working on a launchpad.
Igor Lessio : Oh, OK. So the launchpad is out. It's beta still. So please don't get mad if you try to do something that still don't work 100%. But we was in a rush because the community wanted to have the launchpad. That is a very different and peculiar way to launch the tokens. Because now you can attach an agent to that. So you're going to have already your agent working on your social media straight out of the box, together with the launch of the token. This is very, very cool. And it's going to be progressing together with version 2. It's going to add even more specific agents for tasks like, I don't know, community management, marketing, project management, all this kind of stuff. It's going to come out of the box.
Josh Kriger: So what's sort of the ideal project for launching a token with an agent? Like, what are the types of things that you see coming out?
Igor Lessio : Oh, we start to see everything from the shrink of a doctor to a pharmacist to any task that in this moment are covered by a KOL, for example.
Josh Kriger: So is the token to use the product?
Igor Lessio : Not really. It's more like to finance the project for the moment. Support the project. To have a token utility, we're going to need a little bit more time because it needs to be embedded in the agent. The agent needs to know this.
Josh Kriger: Do you foresee projects returning revenue to financiers? What's the incentive to finance one of these new agent projects?
Igor Lessio : The incentive is that the company can coexist in a unicorn way. So you can be a solo person, you have a great idea, but you don't have the funds to assess it. And if you are solo, it's going to be very difficult for you to access a VC. So the token for you is the best way to finance your project. This is the easy way. But in general, attaching an agent to a token, even if it's an L1 token, it's going to be necessary because it's what is already coming, but it's what is happening everywhere. So if you don't have an agent that handles your community on Twitter, You're going to be a dinosaur.
Josh Kriger: Makes sense. Pretty exciting stuff. And I know you guys just keep building. How big is the team now and what is next on the road map?
Igor Lessio : The full team, I think we are around 25 because I changed depending on the load. We are working now mostly on the launchpad because we want to make it stable. Yeah, make it stable and then go back to the version 2. Me as an integrator, my main integrator, I'm already preparing new plugins, so team come to me to prepare the plugin for their chain. We're going to be fully integrated with version 2 in probably 60% of the chain that count between now and the end of the day.
Josh Kriger: Exciting stuff, my friend.
Igor Lessio : And if people want to learn more about Eliza, maybe also follow you on X. Or you can follow us anywhere. So Eliza West on X, where you're going to find all of us. And if you want to know something about Eliza, go to Eliza How. All right. Thanks, my friend.
SPEAKER_06: Hi everyone, welcome back to the EDGES show.
Josh Kriger: I'm Josh Krieger, your co-host for today. We are live at Token 2049 at the Hack Season Dubai event. And I'm here with Gaurav, the CEO of Ionet. It's great to have you on the show. Pleasure meeting you. So you guys have made big waves in the Web3 space over the last six months. I think I see articles about you all the time. But let's start with the basics for those that haven't been catching all the news. What is Ionet all about? And how are you guys approaching decentralized compute differently?
Gaurav Sharma: I think first of all, we'll tell you like we're solving something which has already been done in a lot of other industries. And we're basically carrying on those fundamentals. And one of the most critical areas in the world, which is AI and AI space, right? So AI needs GPUs. So irrespective of whatever AI project you're doing, compute is very important.
Josh Kriger: More and more than ever before.
Gaurav Sharma: And I see one pattern in industries that whenever the power is too much in the hand of few limited players, then they take advantage of it. So I'll give you an example of a different industry altogether, and then it'll be very, very easy to map it out to this industry also, which is compute, right? So I have a background with Amazon, Booking.com, eBay, HP, R&D and all, right? I've seen this thing happening in the past, right? So, in a hotel industry long time back, let's say you come in, token 2049, and imagine Booking.com doesn't exist, all the local pairs don't exist. What do you do? You go to one popular chain like Marriott, you ask for a room, it's a popular event, everything is booked. You go to Hyatt, a different chain, everything is good. Let's say you now land up in the third chain. Let's say Skipinski out here, right? You ask them and they say only one room off. What do you do? You panic and you say everything is gone. Limited inventory. When you ask the price, they charge you thrice the price. Yeah. And that used to happen in hotel industry. Right? Does it mean that when this event is happening, there's no empty hotels here? No, there are plenty. But you are only used to going to these three and they charge you a bomb for it. Then what happened? Airbnb came, Booking.com came and US Priceline came and what did they do? They aggregated all this inventory, gave it to the users, created a fair marketplace both for supplier of hotels and the users. It's exactly the same problem here in the compute world. AI needs compute and today compute lies in the hand of three players, and they share 65% of the market share.
Josh Kriger: So there's pockets of availability that the consumers are not learning about.
Gaurav Sharma: Not learning. So there are data centers which have a lot of inventory. There are software companies who have GPUs sitting idle there. They have inventory. And there are a lot of gamers who have inventory also, right? There are a lot of miners who have a lot of GPUs. The beginning should be that we should use this inventory as much as possible. So practically giving an example, the rate at which adoption is happening of AI and the rate at which the GPU are getting created is two to three times of a difference. I see. And this gap is not going to go lower because in GPU business, there are only limited players who can make it. To be honest, the most powerful one right now is only with one player, NVIDIA. they have the whole market share, right? Secondly, there's trade wars happening, the raw material issue happening. So this gap of AI adoption and the raw material, not going anywhere.
Josh Kriger: So your product's deployed. Yes. And, you know, what's been the traction so far?
Gaurav Sharma: It's been phenomenal because we did something that customers were looking for. So we had a product market fit right from day one. So we have already made more than 11 million in T&E just in past five months. We are a one and a half year old project. We knew what we were doing. We had builders who have done it in the past. In the first one year, we built a platform. Last five months, we focused on the revenue. 11 million T&E already done. We captured all the contracts, 35 plus million already there. So we're exponentially growing in this business. And honestly, this is not stopping because this is a business where People come, first of all, they give you small orders. They see you're working well, and they come with the big orders.
Josh Kriger: With that sort of early traction, what keeps you up at night?
Gaurav Sharma: Me? To be honest, I've been in crypto for the last three years. I was part of the Binance team. And I'm fascinated by what decentralization and all that everything can bring to the customers, right? And I see the real value and finally something other than Bitcoin, which can actually help people. The real adoption which can happen in Web 3, from the Web 2 world. Like this is the actual product which will bring it, right? And I know it. So what keeps me at night is, how can I migrate these Web 2 people to Web 3 fastest? So this traction is there and we just utilize it as much as possible at the speed where It's going to catch everything.
Josh Kriger: Is it like paid advertising campaigns using Google Ads? Is it doing traditional, like we're at a Web3 event, are you guys also going to traditional events like more traditional tech conferences? What are your thoughts on sort of the fastest ways to onboard more users?
Gaurav Sharma: We're doing a combination of multiple things. First of all, we had a good name in the Web3 side of things, so almost all the big AI agents and the AI players by default came to us, so we captured that market pretty well. In the Web2 side of things, we have a BD team, they knew a lot of customers from their past background, they captured it, and now the flywheel has already started. Now what we're doing is, because we know that this can exponentially grow, We are beefing up our team, we are doing this LinkedIn, Google and all these ads obviously. Now we are actually doubling it down. What we are doing is, now we are using these channel partners. We are going to channel partner, partnering with them and they are subject matter experts. They already have a big, big, big team which are already trained. So we don't need to hire people in-house, train them. We are doing that though, but the faster way is, with the channel partners also who are doing it from years and years. So we are doing that. We are hiring our BD team, beefing it up. All the popular channels, we are utilizing that.
Josh Kriger: And I would assume that the regulatory climate changes both in the U.S. and this region and Hong Kong, it's just creating more opportunities, right? It's sort of reducing the barriers to entry.
Gaurav Sharma: Absolutely, yes. That's exactly what's happening.
Josh Kriger: So I guess at that point, it's all about advancing the product, right? Is there various levels of sophistication? How expansive is your technical roadmap for the feature sets that you want to do? And how are you thinking about the customer discovery process, given you already have product marketing?
Gaurav Sharma: To be honest, that's what the fascinating thing about this industry is. Because we have a product, and now to bring customers in, We have the blueprint also. We are not imagining something that if we do it, the customer would come. We already know the blueprint. Why? Because these builders are using a bunch of products in Web3, and they're saying, please build the same products from the Web2 side to the Web3 side, and we are ready to come in and master it, right? For example, simple, simple concept like Kubernetes, VM, queuing some systems, some managed databases. All these things have already been done in Web2. So we know the blueprint. And the customer is saying, you have a compute, which is at extremely cheap price. We're ready to consume. But we have already gotten married to some of these products which are there in the Web2 side. The moment you bring them in your ecosystem, we can bring in a lot of business in. And we're just doubling down on that. And we have a bunch of engineers who have 10 or more years of experience just building this in the past 10 years. And we're just following the same blueprint, building it. Boring stuff. But that's what keeps us excited and that's what brings the revenue in.
Josh Kriger: How do you relate your approach to sort of this problem to what they're doing? And is there an intersection there? I mean, they're working with NVIDIA, OpenAI, a bunch of these projects. Are you looking at big partnerships like that? Or is it more about the broader market of potential users?
Gaurav Sharma: We were actually looking at a more broader market. And honestly, we were looking at more of the startups, the early startup, the mid-level startups.
Josh Kriger: These are the lost soldiers. These are a lot of cases. And that's probably similar to you and Aether as well. They're going a little bit for the bigger consumers. The smaller guys are getting left to sort of get the crumbs, right?
Gaurav Sharma: We always talk about mid-level startups are the ones which need help, honestly. A lot, right? They're being taken advantage in every way possible. I normally give an example to people that all these AI companies normally spend like 50% of their money on just compute. Imagine you're a $20 million funded company and $10 million you're just spending on compute. If you can save 70% of that money, you can hire a bunch of engineers, have a longer runway. And secondly, people don't talk about it. And I've been in that industry, I know. So a lot of the roles in these web 2.0 players, which is like architects, which are supposed to help these early builders who are sometimes doing their first projects. They're working in countries in Asia and Africa where there's not a lot of software ecosystem. They don't know where they're headed. So they need the guidance. So they have a role like architects. who are supposed to help them. And now these Web2 players use these roles to pull in more money from these startups. They basically guide them to use the product which they don't even need. So these companies just don't need this early compute, cheap compute. They also need the guidance of people who are there. So we're doing that as well.
Josh Kriger: It's a fascinating industry you're in and some very unique dynamics. We could keep chatting for a while. I had one more question for you in terms of ecosystem growth. Typical programs are grant programs, subsidies, credits, incentives. Is there anything in that area that you're doing a little differently?
Gaurav Sharma: We're doing that. So we are giving credit points to people where they can use it for GPUs. We're also doing grants for the early startups, depending on their product. We just launched a project on a launchpad where the early builders who have built a good product, but they don't know how to show their product to the market, how to basically bring it in front of the people, we do that also. So we have a launchpad also where we just not guide these early builders, but bring them in the forefront also where they have eyes on the product and the whole ecosystem can grow. So we do that as well.
Josh Kriger: That's great. I'm really excited to see you guys continue to flourish and have you back on the show at some point. In the meantime, where can people go to learn more about io.net? I think the answer's on your shirt, io.net.
Gaurav Sharma: Go to the website, and honestly, just try the product. Just try the product. You'll get your answers. If you need to know and learn more, you can reach out to us on LinkedIn. We have a good presence on X also. Are you on X? I'm on X. What's your X handle? Gaurav underscore, I own it.
Josh Kriger: All right. Well, thank you so much for hanging out with us. That's a wrap for the Edge of Shove.
People and Resources Mentioned:
About Our Guests:
Jon Jones is the Co-Founder and CBO at Supra, a vertically integrated L1 blockchain focused on high-speed, low-latency execution for DeFi applications. With a deep background in decentralized infrastructure, Jon is pioneering on-chain automation and Oracle services that streamline app development and system performance.
Rachit Agarwal serves as APAC Lead at Wormhole Foundation. Formerly in VC and marketing, Rachit is helping shape Wormhole's multi-chain messaging protocols and institutional adoption strategies, especially in high-growth regions like Hong Kong, Korea, and Vietnam.
Igor Lessio is a core developer at Eliza and a seasoned AI engineer since 2011. Focused on simplifying AI agent integration, Igor builds decentralized tools that empower solo creators through mobile-first, Telegram-accessible AI assistants.
Gaurav Sharma is the CEO of io.net and a former Binance executive with experience at Amazon and eBay. He’s leading the charge to decentralize compute infrastructure, bringing idle GPU power into AI workloads via a scalable, Web3-native architecture.
Guest Contact Info:
Jon Jones (Supra)
Rachit Agarwal (Wormhole)
Igor Lessio (Eliza)
Gaurav Sharma (io.net)
.jpg)
.jpg)
.jpg)
.jpg)
.svg)
