The Good Tech Companies - One Company Runs 95% of Solana's Validators. Flowra's Harry Hwang Is Building the Open Alternative

Episode Date: July 6, 2026

This story was originally published on HackerNoon at: https://hackernoon.com/one-company-runs-95percent-of-solanas-validators-flowras-harry-hwang-is-building-the-open-alternative. Flowra CEO Harry Hwang on loosening Jito's 95% grip on Solana orderflow, why open competition pays validators more not less, and compliance-grade MEV rails. Check more stories related to undefined at: https://hackernoon.com/c/undefined. You can also check exclusive content about #flowra, #ai-and-ml, #cybersecurity, #jito, #good-company, #solana, #web3, #technology, and more. This story was written by: @ishanpandey. Learn more about this writer by checking @ishanpandey's about page, and for more stories, please visit hackernoon.com. Jito's client runs on 95%+ of Solana's active stake; one provider operates the validator client, the relayer, and the block engine where ordering rules live. Flowra — led by Harry Hwang (ex-TradFi/CeFi quant; founder of block-builder Mevity) — is building an Open Orderflow Auction (OOA) plus Programmable Block Policy (PBP) as the open alternative. The counterintuitive core: exposing orderflow to many searchers compresses searcher margins, not validator revenue. Ethereum's PBS produced a measured +261% in proposer revenue. Compliance is built in: PBP enforces sanctions/AML screening and blocks single-bundle sandwich attacks at the infrastructure layer. The honest risk: open markets can re-concentrate — on Ethereum, the top three builders now win >95% of auctions.

This audio is presented by Hacker Noon, where anyone can learn anything about any technology. One company runs 95% of Solana's validators. Flores Harry Wong is building the open alternative by Ashan Pondi. On Solana, almost every transaction that needs to land during congestion runs through infrastructure one company built. Jita's client sits under more than 95% of active stake, and its tips have grown to the majority of the network's real economic value. The closest thing the chain has to a monopoly.

on who gets into a block and what that access costs. That cost isn't abstract. Priority fees can spike as much as 100x in a volatile session, and a U.S. class action now alleges billions in retail losses tied to how Solana orders transactions. One provider controls the order flow layer. One set of rules governs who wins. Flora wants to prize that market open on a claim that sounds backwards, expose order flow to more competitors, and validators earn more, not less. It's the Bet-Etherium already ran and won. Today, Ashan Pondy sits down with Harry Wong, CEO and co-founder of Flora for Hackernunes,

Behind the Startup, series. The through line. From trading on top to building underneath. Ashan Pondi. Harry, welcome. You have had an unusual path. Derivatives prop trading at a traditional finance desk, then Cepheque, Quint trading, then Defi, then building meviti,

a block building company focused on helping validators increase their rewards, and now founding flora. Each of those moves represents a fairly sharp pivot. What was the through line in your thinking? And at what point did you decide that building infrastructure was more interesting than trading on top of it? Harry Wong. Thanks for having me today. I understand it sounds like a series of random pivots, but it has always been about chasing risk-neutral opportunities. At the prop trading firm, I was primarily running pure statistical arbitrate strategies, not so much in the Defy space. Even after transitioning into Defy, Iha had the privilege of having a front row seat to understanding what validators and the broader ecosystem actually wanted and needed.

It became clear to me that the infrastructure layer underneath held bigger and more fundamental unsolved problems than the trading layer on top. In a way, my experience at large financial institutions also gave me an appreciation for just how important compliance as if traditional financial institutions are going top participate in this space. That perspective heavily influenced how we built Flora as a decentralized infrastructure with compliance at its core. I genuinely felt we were the right team to solve what the Solana ecosystem needed next, and that conviction is what ultimately led to Flora. Ashan Pondi Mavati onboarded roughly 4 million Sol of validator stake as a block building operation focused on increasing validator rewards. Most founders would have

kept building on that. What did you observe from inside a top tire block building operation that convinced you the deeper opportunity was the infrastructure layer itself rather than operating on top of it? Harry Wong, I'd frame it less as hitting a ceiling and more as seeing the next chapter clearly. Building Meveti was an incredible experience for us. We we re-fortunate to have major Cephy exchanges and institutional validators on board and or early results were very promising. But having such a close view of what validators actually needed from block building revealed a much larger and more scalable opportunity. That realization led us to pivot toward building the underlying infrastructure. Creating a block engine that is more auditable, competitive,

and compliant ultimately became the core mission behind Flora. The mechanism. Why More Competition Pays validators more as Sean Pondy? Walk us through the mechanism. How does making order flow visible to more searchers actually produce more revenue for validators rather than simply compressing margins through competition. Harry Wong. It seems counterintuitive, but increased competition compresses searcher margins, not validator revenue. On Solana, order flow is typically only visible to validators or private providers like GEDA, allowing a small number of searchers to capture most of the MEV while paying relatively low tips. FlowRAMakes order flow visible to multiple searches simultaneously, creating competitive bidding for the same opportunity. As a result, a much larger share of the

MEV is paid to validators as tips, while searcher profits are competed down toward operating costs. This dynamic has already been validated on Ethereum with MEV boost, and Solana's higher throughput and faster block times make the effect even more pronounced. Greater than the reporting behind this, Harry's mechanism isn't just theory, Ethereum ran greater than the experiment. Proposer builder separation, MEV boost, opened block greater than construction to a competitive market of builders, and a peer-reviewed greater than empirical study measured a plus 261% increase in proposer, validator, revenue greater than afterward. The logic is specialization plus competition. Builders bid away greater than most of the value to win the right to have their block proposed, so the greater

than proposer captures MEV indirectly, by auctioning block space rather than greater than extracting it personally. Ashan Pondi, Jita controls roughly 72% of state Saul and has raised $12.1 million in total funding. You are entering this market at launch with approximately 60 million Saul already committed to migrate. What is the real switching cost for a validator moving from Jita to Flora and why are institutional validators who have every reason to be conservative, willing to root stake through you before you have a live performance track record, Harry Wong, for several reasons, honestly. First, the experience we built through Meviti has been our most valuable asset.

We weren't entering the market as just another infrastructure vendor. We had already spent significant time working alongside validators on block building, and those relationships gave us a very different starting point for those conversations. Second, we see the evolving regulatory landscape around digital assets as Amahor opportunity. As more traditional financial institutions enter the space, compliance will be become increasingly important. Coming from a trad-fi background ourselves, we understood early on that institutional participants would need infrastructure with built-in compliance capabilities. That's exactly what Flores programmable block policy, PbP, is designed to provide.

Third, an open mempool doesn't just improve transparency, auditability, and user protection against excessive fees. It also creates the potential for significantly higher validator yields by enabling a more competitive transaction marketplace. We're encouraged that this thesis resonates across multiple dimensions, and we're grateful for the traction and support we've seen so far. A sandwich attack factory with better branding? A Sean Pondi, the open mempool framing is a double-edged sword. The first question any serious validator or searcher will ask is whether this IS essentially a sandwich attack factory with better branding. How does programmable block policy enforce ethical MEV at the infrastructure level in practice, and what prevents a validator from simply switching

those protections off? Harry Wong. I'm glad you brought that up because it's something we think about a lot. At Flora, we believe the right approach isn't to pretend MEV can be eliminated, but to shape it into something that benefits the ecosystem while protecting users. That's where our programmable block policy, PbP, comes in. It serves as the foundation of our compliance framework, allowing validators to enforce policies required by financial institutions, such as sanctions screening and AML requirements, while also mitigating harmful forms of MEV directly at the infrastructure layer. From day one, PbP is designed to prevent single bundle sandwich attacks,

and we're continuously expanding its capabilities to address more sophisticated forms of predatory MEV. For us, these protections aren't optional features, their baseline requirements for building infrastructure that institutions can trust. That said, we don't believe MEV itself is inherently bad or something that can be completely removed. MEV is a natural byproduct of blockchain design, and attempts to eliminate it entirely often just drive it into less transparent channels. The more important question is whether the incentives around MEV area aligned with the long-term health of the network. Many forms of MEV, such as atomic arbitrage, liquidations, and healthy back running, actually improve market efficiency and contribute to price discovery. Our objective is to preserve those beneficial mechanisms while systematically reducing extractive

behavior that comes at the expense of users. In other words, we want to build infrastructure where transparency, compliance, and healthy market dynamics reinforce one another instead of competing. Greater than the reporting behind this, the stakes here are large and now legal. A US greater than federal court has allowed a class action against pump. Fun, later expanded to greater than include Solana Labs, the Solana Foundation and GES. to Labs, to proceed, it greater than alleges $4 to $5.5B in retail losses tied to M.E.V. in transaction ordering, a greater than figure the court has not verified. Whatever the final number, it is why, fare greater than ordering, on Solana has become an economic and regulatory

question, not just a greater than technical one. The business, beyond the tip FEE is Sean Pondi. Price alone rarely wins a structural market. Where does Florese revenue model go beyond the tip fee, and what does the long-term business actually look like? The Mempool API, the validator client, or something else seen tirely, Harry Wong. Our core argument isn't about competing on fee rates. It's about growing the size of the pie itself. Because the total tip volume generated within Flora's O.O.A network is larger. Both validators and Flora earn more in absolute terms without having to compete solely on

fee rates. A smaller share of a much larger pie is still worth more. Looking further ahead, we see three primary revenue streams. The first is the protocol tip fee, which exists today. The second is the open order flow data stream itself. Real-time Mempool data has independent value for trading firms and institutions, making it a meaningful revenue layer in its own right. The third is institutional services built around programmable block policy, PBP.

Exchanges and financial institutions that want to operate Solana validators require compliance-grade block policy infrastructure, and that capability doesn't meaningfully exist today. Providing it represents a distinct business opportunity. At this stage, however, our primary focus is on growing the network and expanding the transaction marketplace. As that foundation scales, the revenue model naturally scales with it. The honest case. Why not simply the next Jita? Ashan Pondi. Final question. Validators chose Gita voluntarily because it paid better, not because they were forced. The dependency was economic, coercive. What is the honest case for why Flora is structurally different rather than simply the next dominant provider in the making, once the network affects Kikin? Harry Wong. That's a fair

question. I agree with the premise that validators chose G-da because it delivered better economics, not because they were forced to. If Flora builds a better product, we hope validators will choose us for the same reason. The difference we're trying to build is structural rather than purely commercial. We can't promise that Flora will never become large or successful, but we can design the infrastructure so that success doesn't create the same kind of dependency. The first difference is our open order flow architecture, OOA. Because the transaction marketplace is open, the information and access that competitors need to build alternative infrastructure are available from day one. If someone believes they can build a better block engine or offer a different value proposition,

they don't have to start from scratch or rely on privileged access. That's a fundamentally different foundation from a closed transaction marketplace. Greater than the reporting behind this. This is the right question to end on, because greater than Ethereum shows the precedent cuts both ways. Proposer builder separation greater than raised validator revenue sharply, plus 261%, but the same open market greater than reconcentrated. Today the top three builders win more than 95% of MEV boost greater than auctions,

and private order flow, approximately 12% of transactions, came to carry approximately 54, 6% of greater than block value. Open architecture lowers the barrier for challengers. Whether it greater than stays open depends on resisting the exclusive order flow deals that greater than re-centralized Ethereum. That is the bar flora has set for itself. The second difference is programmable block policy, PbP, validators retain control over their own block policies, whether that's compliance requirements, bundle restrictions, or other operational preferences. Those policies belong to the validator, not to flora. That gives validators real autonomy rather than requiring them to adopt a single provider's view of how blocks should be built. Finally, we think the market itself is evolving.

As digital asset regulation matures and more financial institutions enter the ecosystem, compliance will become a competitive requirement alongside economics. Our goal isn't simply to be the highest paying block engine. It's to provide infrastructure that is transparent, compliant, and economically competitive while remaining open enough for others to compete with us. If we've built the system correctly, our succession and come from locking others out. It should come from building the best infrastructure in an open marketplace. Don't forget to like and share the story, vested interest disclosure. Hacker Noon has reviewed the report for quality, But the claims herein belong to the author.

Hashtag D-Y-O-R. Thank you for listening to this Hackernoon story, read by artificial intelligence. Visit Hackernoon.com to read, write, learn and publish.