Inside Humidify: How Proprietary Market Makers Are Reshaping Solana's Trading Landscape

The rise of proprietary automated market makers (Prop AMMs) represents one of the most significant shifts in decentralized trading infrastructure on Solana over the past year. What began as an obscure category that crypto Twitter once called "dark MMs" has evolved into the dominant force for on-chain trading, routing the majority of volume for many of Solana's most active trading pairs. At the forefront of this revolution sits Humidify, a prop AMM that emerged in June 2024 and rapidly ascended to become not just the leading prop AMM, but the top decentralized exchange on all of Solana.

In a recent Lightspeed livestream, Kevin, one of the key figures behind Humidify, pulled back the curtain on how prop AMMs operate, what makes them fundamentally different from traditional passive AMMs, and why they've been able to capture market share so quickly. With Humidify preparing to launch its WET token via Jupiter's new Decentralized Token Formation (DTF) platform, the timing offered a rare opportunity to understand the mechanics behind these sophisticated trading systems and their vision for making Solana the primary home of global token trading.

The Genesis of Prop AMMs on Solana

Prop AMMs didn't emerge from nowhere—they built upon earlier innovations in the Solana ecosystem. Kevin was careful to acknowledge the pioneers who created this next-generation platform, naming Solphi, Obrick, and Zerophi as the initial cohort that established the category. Even before them, Infinity had attempted something similar, utilizing off-chain oracles like Pyth to update prices in real-time. What distinguished prop AMMs from traditional passive AMMs was their approach to liquidity provision and price discovery.

Traditional AMMs like Raydium, Meteora, and Orca are built to programmatically allow users to deposit liquidity and generate trading pool fees from those deposits. The liquidity sits passively in the pool, waiting for traders to interact with it. Prop AMMs operate on an entirely different principle. Rather than taking customer deposits, Humidify uses its own proprietary capital, employing sophisticated trading algorithms that update quotes in real time based on predictive price models.

The key difference lies in the frequency and sophistication of these updates. While passive AMMs require users to set liquidity parameters and then leave them, prop AMMs send multiple updates within each block—effectively every 50 milliseconds on Solana's current infrastructure. This constant streaming of price updates allows prop AMMs to maintain accurate pricing even in highly volatile conditions, offering significantly tighter spreads than would otherwise be possible.

Kevin's Journey from Traditional Finance to Crypto Trading

Kevin's background provides important context for understanding Humidify's approach to market making. He spent approximately nine years at Jump Trading, one of the most well-known high-frequency trading firms in traditional finance. This experience immersed him in the sophisticated world of quantitative trading, where milliseconds matter and price prediction models determine success or failure. After leaving Jump, he spent time at Paradigm and later at Symbolic Capital Partners, gradually transitioning deeper into the crypto ecosystem.

About five years ago, Kevin fully committed to cryptocurrency trading. When he began exploring Solana, he was doing automated trading but struggling to find the success he was looking for. A connection with the Temporal team changed everything. They recognized complementary skill sets and decided to meet in person to see what they could build together. After just a weekend of work, the first MVP of Humidify was born.

The rapid evolution from that initial prototype to becoming the top DEX on Solana was not something Kevin expected. Humidify executed its first trade in June 2024, became the number one prop AMM by August, and by October had claimed the position of top decentralized exchange on all of Solana. Kevin attributes this success to simply building something that people wanted, even if those users didn't fully understand what was happening under the hood.

How Prop AMMs Achieve Superior Capital Efficiency

One of the most striking aspects of Humidify's operation is the extreme capital efficiency it achieves. With approximately $8 million in total inventory, Humidify regularly routes $500 million to over $1 billion in daily trading volume. This represents a level of capital efficiency that would be impossible with traditional AMM designs.

Kevin explained this efficiency through a comparison with concentrated liquidity AMMs like Uniswap v3, which allowed liquidity providers to place their capital just at the top ticks near the top of book. Prop AMMs take this concept to its logical extreme: all liquidity is placed exclusively for the next trade. There's no resting liquidity waiting for improbable price scenarios.

"If people want to be able to sell all their Solana down to a price of $1, we won't be there. We have no liquidity waiting to buy Solana at $1," Kevin explained. "However, if you want to sell Solana right now for as much as you want, we're going to provide you the absolute best price. If you want to sell one SOL, ten SOL, a hundred, or a thousand—we're going to give you the best price. We instantaneously give you all of our liquidity for just that one trade."

This design philosophy means Humidify never has unused liquidity sitting idle. Every dollar of capital is actively deployed to service the current trade, maximizing returns while minimizing the capital required to provide deep liquidity.

The Oracle Update: Core of Prop AMM Technology

At the heart of Humidify's operation is its off-chain oracle system, which Kevin described as a predictive price model similar to what firms like Jump or Tower would use in traditional finance. This system continuously listens to prices from major centralized exchanges—Binance, OKX, Coinbase—across both spot and perpetual markets. Crucially, it also incorporates data from decentralized exchange venues, attempting to capture information that no other market participant has access to.

The result of this continuous data processing is what Humidify believes to be the fair value of any asset at any given moment. The challenge lies in getting that information onto the Solana blockchain as efficiently and effectively as possible. Once the oracle price lands on-chain, Humidify quotes a bid below that price and an offer above it, for any size trade a user wants to execute across any asset they quote.

Profitability is measured by the difference between the oracle price update and the actual transaction price. In theory, every trade where Humidify successfully lands an oracle update is profitable. The complication arises from competition—other sophisticated actors who are also trying to land their transactions on-chain with better price models or better execution.

Compute Unit Optimization: The Blockchain Equivalent of Latency

One of the most revealing technical details Kevin shared was the importance of compute unit (CU) optimization for prop AMM success. According to data on the Blockworks Humidify dashboard, Humidify has reduced its CU usage from over 300 to just 47 over the past six months. This dramatic improvement has profound implications for trading performance.

Kevin drew a parallel to traditional markets, where different firms have their own views on price, but expressing that price requires execution logic. Some firms use FPGAs and ASICs to achieve faster execution in traditional markets. On Solana, the equivalent competitive advantage comes from sending transactions with the lowest possible compute units.

"Being able to reduce from 300 to 47 means that when the sequencer is deciding whether to put our oracle update ahead of some other prop AMM's oracle update, this will be one of the determining factors," Kevin explained. The Temporal team, particularly KV and Ben, treat this as an ongoing competition to keep driving the number down.

Lower CU usage translates directly into faster transaction landing, which means Humidify can quote safely and aggressively with minimal risk of being arbitraged. If they couldn't land oracle updates efficiently, they would have to quote wider spreads to protect against adverse selection, ultimately hurting all users.

Competing with Adversaries Rather Than Blacklisting Them

A fascinating aspect of Humidify's approach is their stance on toxic flow—sophisticated traders or arbitrage bots that consistently profit at the market maker's expense. Unlike some competitors who have chosen to blacklist or refuse to trade with these informed actors, Humidify has taken what Kevin described as a "bold stance": they try to improve their system to remain profitable regardless of who's asking for a quote.

"We're competitive and we don't like being beaten by anybody," Kevin admitted. "When we see that somebody's able to consistently profit off of us, we try to figure out why." So far, they've been able to combat every adversary, though Kevin acknowledged this may not last forever.

The rationale behind this approach goes beyond pure competitiveness. To offer the best service to Solana users—the tightest spreads of any DEX—you have to be able to accommodate any trader. This means being what Kevin called "the alpha on the chain," setting a standard of being able to handle any flow profitably.

This philosophy extends to arbitrage bots that profit by aligning Humidify's prices with other venues like Orca or Solphi. These trades are still profitable from Humidify's perspective, and they contribute to overall market efficiency. The current meta is to allow this activity, though Kevin indicated they may see advancements that better protect retail users in the future.

The Day One Venue Vision: Monad as a Case Study

One of Humidify's most ambitious goals is to make Solana the primary venue for trading new assets from the moment they launch—what Kevin called being a "day one venue." The recent Monad token launch provided a compelling case study of how this works in practice.

Wormhole had created Project Sunrise and approached Humidify on a Friday night about potentially providing liquidity for the Monad launch. Kevin jumped on a call Sunday night, and the team revealed they were launching in just eight hours. Despite the tight timeline, Humidify agreed to participate.

Thanks to their stable setup for adding new pairs, they were able to start quoting Monad within about an hour of when it started trading on the Monad network itself and on Coinbase. The results exceeded expectations. "By the end of the day, we were one of the top venues in the entire crypto space. Within a couple of hours, we were offering the best prices of any venue, including versus CeFi exchanges and on some DEXs on Monad themselves," Kevin recalled.

According to data cited in the discussion, approximately 60-80% of the total trading volume on the MON token flowed through prop AMM venues. This demonstrated the potential for Solana to become the primary trading venue for new token launches, regardless of where the underlying asset originates.

Trading Without CeFi: The "Abstinence Trading" Philosophy

Kevin introduced a compelling concept that Humidify calls "abstinence trading"—trading without any centralized exchange involvement. The underlying idea is that there's no fundamental reason why a price model can't be built using only decentralized trading activity.

The ORE token represents their first attempt at this approach. ORE has no CeFi listings and limited DeFi venues, making it exceptionally challenging to price. Success requires creativity and innovation in building price models from scratch using only on-chain data.

"All the other HFT firms have done it. There's always some source of price discovery, and we want Humidify to become that for Solana for as many assets as we can support," Kevin explained. This vision aligns with the broader goal of making Solana "internet capital markets"—a platform that can support any asset that traders want to access, whether it has CeFi listings or not.

Handling High Volatility: Lessons from Market Stress Events

The discussion addressed a notable incident during a period of extreme market volatility (referred to as the "10-10 event") when Humidify's oracle sender and price prediction model crashed. Kevin was away at lunch interviewing a candidate when the messages started flooding in, and he had to rush back to restart the model.

Kevin was transparent about what happened: their system simply couldn't handle the unprecedented volatility. By missing that hour or so of trading, Humidify estimated they lost more than $500,000 in potential profit. Those high-volatility conditions are precisely when prop AMMs can thrive due to their superior price models and execution capabilities.

"I consider it a skill issue," Kevin said bluntly. "Our system wasn't ready. Now it's much better suited for that. But that doesn't mean it will never go down ever again." The incident opened up a huge opportunity for competitors, who performed very well during that period.

Kevin pushed back against some of the theories that circulated on crypto Twitter, clarifying that Humidify wasn't intentionally stopping quotes. The expected behavior during high volatility should be wider spreads and thinner liquidity—not complete cessation of trading. His view is that multiple venues, including passive AMMs and other prop AMMs, are necessary for ecosystem resilience.

The Relationship with Aggregators: Jupiter, DFlow, and Titan

A significant portion of Humidify's volume flows through aggregators like Jupiter, DFlow, and Titan. Kevin acknowledged that this reliance on aggregators helped propel prop AMM market share growth, particularly early on when Jupiter's dominance meant that connecting to their flow was sufficient to justify the work of building a prop AMM.

The relationship has evolved over time. Initially, Humidify had to conform to requirements set by Jupiter. Other aggregators adopted similar requirements to support as many prop AMMs as quickly as possible, ensuring they could also access the best liquidity and pricing.

Kevin expects this dynamic to shift. "I expect Humidify to put in features that we require the aggregators to support. I expect the aggregators to continue launching new features that we'll need to support." There's no specific preference for any aggregator—Humidify's customers are all Solana users—but aggregators that ship new features faster will win more flow.

Humidify doesn't operate a frontend interface. They have a smart contract that routers can access to get quotes, which then get incorporated into the aggregator's routing algorithm alongside quotes from other venues. Kevin noted that while Humidify has some experience with routing algorithms, it's not an area where they believe they're best in the world, so they focus on what they do best: providing tight spreads and optimal trading outcomes.

Why DTF for the WET Token Launch

The decision to launch Humidify's WET token via Jupiter's new Decentralized Token Formation (DTF) platform was driven by several factors. First and foremost was alignment in principles around transparency. Despite Humidify's historically secretive nature as a proprietary trading operation, Kevin emphasized that reputation matters enormously to the team. They want to do things the right way, with full transparency about token dissemination, wallet locations, and future plans.

Jupiter's position as one of the top aggregators in the space, having evolved the market microstructure alongside prop AMMs since the beginning, made them a natural partner. Their user base aligns well with Humidify's target audience, and being the first project on DTF offered an opportunity to kick things off successfully for both parties.

The token launch represented Humidify's shift toward a more educational mindset. While there's typically no reason for proprietary traders to share alpha, Kevin and the team want people to understand how the microstructure works and to contribute to broader ecosystem knowledge.

Solana vs. Ethereum: Why Prop AMMs Thrive on Solana

Kevin provided detailed analysis of why Solana is uniquely suited for prop AMM activity compared to other chains. Ethereum's 12-second block times create a fundamentally different trading environment where the ability to price assets matters enormously, but latency is less critical because you only have to transact once per block.

Solana's architecture, with effectively 15-millisecond sub-slots within 400-millisecond slots, creates a much more high-frequency environment. Every 50 milliseconds represents a potential transaction opportunity. Success requires the ability to consume data and respond to it very quickly, and landing transactions efficiently remains "more of an art than a science," according to Kevin.

The microstructure differences are significant. On Ethereum, the supply chain is well-documented with predictable actor behavior. On Solana, leadership rotates around the world, requiring good connectivity everywhere. Kevin revealed that dark fiber—dedicated fiber optic connections used by sophisticated HFT firms—is a critical advantage for moving data around quickly. Double Zero has "blown the spot" on this, making it common knowledge that better networking and bigger pipes are critical for chain performance.

Base represents the closest L2 comparison, and Kevin noted that Tessera, one of Humidify's Solana competitors, appears to have launched a prop AMM on Base. However, the single sequencer architecture on Base creates a different dynamic where the target for data delivery is better understood.

The Role of the Temporal Partnership

The Temporal team plays a crucial role in Humidify's technical capabilities. Kevin describes Humidify as a joint venture of sorts, comprising himself, one other close collaborator, and engineers from the Temporal team. They use Temporal's Zomi product for transaction landing because it's the best available product for their needs.

However, Kevin was clear that he doesn't have insider knowledge of Temporal's broader plans, including their work with Harmonic. He gets the same news as everyone else and suggested that Ben and Connor at Temporal should reach out if they want him involved in an advisory capacity.

What Humidify does do is study Jito, currently the primary method for landing transactions, and propose improvements to make the microstructure more efficient. This kind of collaboration between prop AMMs and infrastructure providers helps drive the entire ecosystem forward.

Profiling Traders and Managing Flow

Humidify maintains extensive analytics and metrics on different market participants, profiling arbitrage bots, CeFi-DEX traders, retail traders, and others. The goal is to understand how each participant type interacts with the system and whether they contribute positively or negatively to profitability.

Kevin emphasized that distinguishing between sophisticated toxic flow and retail users trying to get the best price is not black and white. Humidify naturally provides better prices when they know they're transacting with retail. But they've taken the stance that arbitraging between Humidify and other venues like Orca or Solphi still creates profitable trades, so they're willing to accommodate this activity.

This permissive approach contributes to higher volume, as Humidify effectively helps other venues align their prices with market consensus. Whether this continues indefinitely remains to be seen, but for now, the incentives align with being as accommodating as possible while remaining profitable.

Transaction Failures and Oracle Update Dynamics

The discussion clarified the relationship between oracle updates and user transaction failures. Humidify strives to land every oracle update, but Solana's microstructure—400-millisecond slots with complex transaction ordering through both TPU and Jito—means success isn't guaranteed.

Sometimes oracle updates revert due to rapid price movements. If an update is sent because prices moved up, but then prices reverse before the transaction lands, the second update might fail to land while the first one succeeds, leaving Humidify willing to buy at an elevated price. These scenarios can result in losses.

User transaction failures are a separate issue, typically caused by slippage tolerance settings. If a user sees a quote at $130 but prices move to $140 before they sign the transaction, and their slippage tolerance can't accommodate that move, the transaction will fail. This is independent of Humidify's oracle update success rate.

The Importance of Being First to Market

Kevin articulated a clear vision for what "day one" trading requires. The Monad launch demonstrated that success requires multiple entities pushing in the same direction: someone to bring the asset on-chain (Wormhole's Project Sunrise in that case), multiple exchanges willing to list and trade the asset, venues capable of pricing without CeFi references, and users who want to trade.

The chicken-and-egg problem is real, but Kevin believes you have to play for keeps. Half-measures won't work. If Solana wants to be the home of trading, it needs to support all these components from minute one, offering the deepest liquidity and best spreads for traders of all sizes.

The alternative creates fragmentation: users go to Coinbase for the presale, realize Humidify offers better prices, bridge their assets, trade on Humidify, then bridge back to stake. This friction undermines the user experience and prevents Solana from capturing full trading activity.

Looking Ahead: Firedancer and Beyond

When asked about what developments excite him most about Solana's future, Kevin focused on the vision of becoming the primary home of liquidity on day one for new assets. While technical improvements like MCP (Multi-Concurrent Processing, expected in 6-12 months depending on who you ask) are interesting, Kevin expressed appreciation for the overall discourse and competition forcing everyone to push harder.

His specific advice for Solana was to focus on doing a few things—or even one thing—exceptionally well, rather than trying to do everything in a mediocre way. The emphasis on trading as Solana's core strength aligns with Humidify's mission and makes the collaboration between prop AMMs and client teams particularly productive.

Kevin committed to solidifying Humidify's home on Solana regardless of what other chains evolve. While other ecosystems have their strengths and weaknesses, and Humidify continues to monitor them, Solana remains clearly the top chain for trading from their perspective.

Competition Drives Innovation

Throughout the discussion, Kevin repeatedly emphasized the positive role of competition in driving the ecosystem forward. Having multiple prop AMMs, aggregators, and trading venues creates healthy pressure that benefits end users through tighter spreads and better execution.

Kevin's commitment to avoiding blacklists and competing directly with sophisticated traders stems partly from this philosophy. When adversaries find ways to profit, the response should be to improve, not to exclude them. This creates a continuous improvement cycle that raises the bar for everyone.

The recent innovation in routing—DFlow's JIT routing that checks prices in real-time, followed quickly by Jupiter's similar implementation—exemplifies how competition accelerates development. Humidify welcomes this dynamic and expects to both respond to and drive feature development going forward.

The Bigger Picture: Decentralized Finance's Promise

Underlying all the technical discussion was a clear philosophical commitment to decentralization. Kevin explicitly stated that the ambition is to support every asset that CeFi exchanges can, and more. If prop AMMs can only quote tokens with CeFi listings, they're not really advancing crypto's purpose of creating decentralized markets.

The goal is to help create liquidity where people don't want to trade on centralized exchanges. Passive AMMs support this by always being ready to set a price, but they do so inefficiently. Prop AMMs marry the efficiency of sophisticated trading systems with the decentralized nature of on-chain execution.

This vision extends to price discovery itself. Dan Smith's observation that price discovery may be moving on-chain due to prop AMM activity represents exactly what Humidify wants to achieve. The ultimate success metric isn't trading volume or market share—it's whether Solana becomes the primary venue where asset prices are discovered and liquidity concentrates.

Conclusion: A New Era for On-Chain Trading

Humidify's emergence and rapid growth represents a fundamental shift in how decentralized trading works. By combining sophisticated quantitative trading techniques from traditional finance with Solana's unique high-performance architecture, prop AMMs have demonstrated that on-chain venues can compete with—and potentially surpass—centralized exchanges for many use cases.

The technical innovations are impressive: compute unit optimization measured in single digits, oracle update systems that land transactions in milliseconds, and capital efficiency that allows billions of dollars in volume with single-digit millions in inventory. But the broader significance lies in what this enables: truly decentralized trading for any asset, from day one, with competitive pricing for traders of all sizes.

As Humidify launches its WET token and the prop AMM category continues to mature, the groundwork is being laid for a future where Solana serves as the internet's capital markets. Kevin and the Humidify team are betting that proprietary trading expertise, deployed transparently on a decentralized blockchain, can deliver better outcomes than any centralized alternative. If the past six months are any indication, that bet is paying off.

Facts + Figures

• Humidify executed its first trade in June 2024, became the #1 prop AMM by August 2024, and the #1 DEX on Solana by October 2024

• Humidify operates with approximately $8 million in total inventory while routing $500 million to over $1 billion in daily trading volume

• Compute unit usage has been reduced from over 300 CUs to just 47 CUs over the past six months through optimization

• Solana operates with effectively 15-millisecond sub-slots within 400-millisecond slots, compared to Ethereum's 12-second block times

• During the "10-10" volatility event, Humidify estimates they lost over $500,000 in potential profit due to system downtime

• Approximately 60-80% of total trading volume on the Monad (MON) token flowed through prop AMM venues

• Humidify quotes approximately 15 tokens beyond the SOL/USDC pair

• Kevin spent approximately 9 years at Jump Trading before entering crypto about 5 years ago

• Humidify was able to start quoting Monad prices within about one hour of the token launching on Monad network and Coinbase

• The first cohort of prop AMMs on Solana included Solphi, Obrick, and Zerophi, predating Humidify

• Tessera, a Humidify competitor, has reportedly launched a prop AMM on Base

• Humidify is the first project launching on Jupiter's new Decentralized Token Formation (DTF) platform

• Dark fiber networking has been identified as a critical competitive advantage for landing transactions on Solana

• The prop AMM oracle system listens to prices from Binance, OKX, Coinbase across both spot and perpetual markets, plus DEX venues

• Humidify's team consists of Kevin, one close collaborator, and engineers from the Temporal team

Questions Answered

What is a prop AMM and how does it differ from traditional AMMs?

A prop AMM (proprietary automated market maker) uses its own capital and sophisticated trading algorithms to provide liquidity, rather than accepting deposits from external liquidity providers. While traditional AMMs like Raydium or Orca allow users to deposit liquidity and then passively wait for traders, prop AMMs send multiple price updates within each block—effectively every 50 milliseconds on Solana. This constant streaming of updates based on predictive price models allows prop AMMs to maintain accurate pricing even in volatile conditions and offer significantly tighter spreads than passive alternatives.

How does Humidify achieve such high capital efficiency?

Humidify achieves extreme capital efficiency by placing all liquidity exclusively for the next trade rather than spreading it across a price range. Unlike concentrated liquidity AMMs that place liquidity at specific ticks, Humidify has no resting liquidity waiting for improbable price scenarios—there's no liquidity sitting ready to buy SOL at $1. Instead, all $8 million in inventory is instantly available for the current trade at competitive prices, allowing the platform to route billions in daily volume without requiring proportionally large capital reserves.

Why did prop AMMs gain market share so quickly on Solana?

Prop AMMs gained market share rapidly because Solana's high-performance architecture creates ideal conditions for sophisticated trading systems. With 15-millisecond sub-slots enabling frequent transaction opportunities, firms with superior price models and execution capabilities can offer meaningfully better prices than passive alternatives. The dominance of Jupiter as an aggregator also helped, since connecting to their flow provided sufficient opportunity to justify building a prop AMM. Additionally, Solana's ecosystem explicitly focused on trading as a core use case, creating alignment between infrastructure development and prop AMM needs.

How does Humidify handle sophisticated traders who try to arbitrage them?

Unlike competitors who blacklist or refuse to trade with sophisticated arbitrageurs, Humidify attempts to improve their system to remain profitable against all traders. When they identify actors consistently profiting from their quotes, they analyze why and adapt their system accordingly. This approach stems from competitiveness and a belief that accommodating any trader—including arbitrage bots that help align prices across venues—is necessary to offer the best prices to all users. Kevin acknowledged this may not work forever, but currently views it as essential to maintaining the tightest spreads.

What makes Solana uniquely suited for prop AMMs compared to other chains?

Solana's 15-millisecond sub-slots create a high-frequency trading environment where rapid data consumption, response, and transaction landing provide competitive advantages. In contrast, Ethereum's 12-second block times make latency less critical since transactions only need to be processed once per block. Solana's rotating leadership across global validators requires good connectivity everywhere, making dark fiber networking a significant advantage. The chain's explicit focus on trading infrastructure and client team responsiveness to prop AMM needs also creates alignment that other ecosystems haven't matched.

How did Humidify support the Monad token launch as a "day one" venue?

Humidify was approached by Wormhole's Project Sunrise team on Friday night about the Monad launch. By Sunday night, they learned the launch was just eight hours away. Thanks to their stable setup for adding new pairs, they were quoting Monad prices within about one hour of the token going live on both Monad network and Coinbase. Within a couple of hours, they were offering better prices than any other venue including CeFi exchanges, demonstrating how prop AMMs can serve as primary liquidity sources from launch day.

Why is compute unit optimization so important for prop AMM success?

Compute unit usage functions as the blockchain equivalent of latency in traditional markets. When the sequencer decides which transactions to include first, lower CU transactions have an advantage. Humidify's reduction from over 300 CUs to 47 CUs means their oracle updates are more likely to land before competitors'. This faster landing enables them to quote safely and aggressively with minimal risk of being arbitraged. If they couldn't land updates efficiently, they would need to quote wider spreads to protect against adverse selection, ultimately hurting all users.

What caused Humidify to stop quoting during the high volatility event?

During an unprecedented volatility period, Humidify's oracle sender and price prediction model crashed due to inability to handle the extreme market conditions. Kevin was away interviewing a candidate and had to rush back to restart the system. This wasn't an intentional decision to stop quoting—the expected behavior during high volatility is wider spreads, not complete cessation. Kevin characterized it as "a skill issue" and noted the system has since been improved, though he acknowledged it may face challenges again in the future.

How does Humidify measure profitability on trades?

Humidify measures profit as the difference between their oracle price update (which represents their view of fair value) and the actual transaction price. The oracle system incorporates data from major CeFi exchanges like Binance, OKX, and Coinbase across spot and perpetual markets, plus DEX venue data. Every trade where they successfully land an oracle update is theoretically profitable. They also track broader metrics like Markouts and total fund accumulation in their pools to assess overall performance.

What is Humidify's vision for "abstinence trading"?

"Abstinence trading" is Humidify's term for creating markets without any centralized exchange involvement. The ORE token represents their first attempt, as it has no CeFi listings and limited DeFi venues. The concept relies on building price models using only decentralized trading activity—something Kevin notes traditional HFT firms have always done by finding some source of price discovery. The goal is for Humidify to become the price discovery venue for Solana across as many assets as possible, regardless of whether they have CeFi listings.