Back in peak DeFi farming days you’d think about APY as how many days it would take to recoup principal with the yield.
That’s how risky it felt and that same rigor should be applied when thinking about interacting onchain today.
Yes, not all farms/vaults are created equal. Not all strategies carry the same risk. But you have to factor in the nonzero chance of total or meaningful principal loss.
I get asked constantly what is enough yield to come onchain. I think it’s at least 18% today. Anything below that is not worth the hassle or the risk.
I won’t do formal math but here is my mental model: smart contract risk + opsec risk (yours) + opsec risk (any protocol you’re interacting with).
These are ever-evolving risks. The more protocols you interact with the more risk blows out exponentially.
There are also things you can’t control. Anthropic’s latest model just found vulnerabilities in codebases from some of the most resourced and established companies in the world. If you think DeFi protocols (written faster and audited lighter) are a harder target, think again.
I say this as a DeFi bull.
Onchain rates are low because there is no demand for assets. That is not the same as risk being priced correctly.
In 1965, Georg Nees hung a dozen plotter drawings on a wall in Stuttgart. The audience walked out the room before they understood what they were looking at.
Sixty years later, @museumartlight is showing twelve on-chain generative works from the @artblocks_io 500 collection and this time, the work gets the deserved attention of the visitors. 🎆EMULATION: Selections from the Art Blocks 500. On view until August 16 with artists like @pointline_@Licia_He@harvey_rayner@kellymilligannz
A lot of this movement started with the vision of @snowfro. What he built with Art Blocks gave generative art a home it always deserved.
Museums are finally catching up. If you care about where generative art came from and where it's going, this one is really worth the trip.
#ArtBlocks #GenerativeArt #DigitalArt
There are 23 different color palettes in Anticyclone by William Mapan, each one named after an important woman in Mapan's life.
This palette, Camillie, is named after Mapan's partner:
"She continues to inspire me day after day, like a very bright sun."
-@williamapan
Construction Token #264 & #473 by @jeffgdavis
"Each Construction Token is unique and contains a randomized seed that determines the composition of the artwork"
Overjoyed with our Keith Haring 1987 digital drawings exhibition @MartosGallery designed by renowned architect KULAPAT! If you’re in NYC, don’t miss seeing how surprising NFT’s can be when installed like this!!
@nftnow@graceb_art
Really enjoyed my conversation today with @michaelh_0g from @0G_labs.
We spoke about his background, what first pulled him deeper into technology, how that developed into a broader Web3 thesis, and the story behind building 0G. It was a meaningful conversation with a lot of depth, clarity, and real perspective.
Grateful to Michael for the time and openness.
And special thanks to the @BeaconLayer team, as well as to @SachiMiyasaki and @DikshaWells, for making it all possible.
In crypto, liquidity dominates—shaping token prices, staking dynamics, and beyond.
Now, following DTCC's recent SEC approval for federally compliant tokenization services, institutional-scale real-world asset (RWA) distribution onchain is poised to accelerate.
Where will they go? Where the liquidity is... 👇
~~ Analysis by @kenzimori ~~
DTCC's Requirements
In its no-action request to the SEC, the DTCC lays out the technical requirements for key components of tokenization systems, including the underlying blockchain and supporting token-tracking software.
This letter forms the basis for the SEC's no-action exemption, and specifies that any systems used to perform key functions of the tokenization service must comply with the DTCC's internal "Tier 2" systems requirements.
Among other things, this criteria will require critical systems components to have "the ability to operate from a primary and a secondary location, a maximum four-hour recovery time objective, a maximum two minutes of data loss from an outage, and annual out-of-region disaster recovery and resumption testing."
While the DTCC is opting to remain tech-neutral, it will not forcibly prescribe a particular blockchain or tokenization protocol. Eligible solution combinations must support compliance controls and achieve a Tier 2 systems rating.
Liquidity Dominates
The DTCC processed $3.8Q (quadrillion) of securities transactions in 2024. It is the highest value financial processor in the world and surpassed $100T in assets under custody this summer. If one was asked to identify the single most important lynchpin in the global financial system, "DTCC" would be a safe answer.
Assuming superior liquidity is bound to win out in tokenization markets, the DTCC's near-limitless, preexisting supply of assets arms it with a durable advantage that should ensure everlasting dominance in tokenized securities markets.
Which Chain Wins?
Just as the DTCC is destined to become the dominant tokenization service by virtue of its enormous scale, liquidity dynamics dictate that a single blockchain will inevitably become the default choice for DTCC-sponsored tokenization offerings.
Given the DTCC's technical requirements, several candidates can likely already be dismissed:
@solana — Extended network outages over four hours are incompatible with the DTCC's uptime requirements. Solana has experienced multiple disqualifying outages, the most recent occurring last year.
@Ripple — Despite being touted for banking integrations, it too has suffered outages; the network last halted in February.
L2s — Contemporary designs use single sequencers that are liable to outages. It's also questionable whether they comply with DTCC requirements to "operate" from both a primary and secondary location.
Bitcoin — The ecosystem lacks the requisite smart contracts functionality to support sophisticated financial applications and DTCC-required transfer restrictions.
In this environment, @ethereum could very well become the default. The globally decentralized blockchain has recorded over ten years of consecutive uptime, making it one of the few blockchains that can achieve the DTCC's uptime requirements without a doubt.
Even so, the DTCC also requires its critical systems to pass "annual out-of-region disaster recovery and resumption testing." How can Ethereum (a blockchain designed not to fail) possibly conduct any type of outage recovery testing? And might stratospheric, market-determined gas fees during high-use periods constitute an untenable outage?
Circle's Arc could be another solution. The blockchain's permissioned proof-of-authority validator will be both decentralized enough to operate from multiple locations and centralized enough to accommodate testing.
Another contender is @CantonNetwork, a public blockchain created by Digital Assets that claims to be the only network capable of offering configurable privacy and institutional-grade compliance. Canton purports to support over $6T in onchain real-world assets and process $280B in daily transactions through its network of 500 validators.
Still, nothing prevents the adoption of proprietary database technology that can operate from more than one location, including some kind of Fedwire-styled, government-sponsored permissioned ledger.
Conclusion
Tokenization markets are a winner-take-most game.
Whether DTCC tokenization ultimately settles on an existing general-purpose blockchain like Ethereum, a purpose-built RWA solution like Circle's Arc, or an unreleased Fedwire-styled government ledger, a future defined by multichain fragmentation appears an unlikely outcome.
Crypto history is unambiguous on this point: liquidity crowns kings. Whichever chain manages to underpin DTCC tokenization efforts stands to inherit a gravitational pull multiple orders of magnitude greater.
From a monster airdrop to ending the year as the #4 revenue machine in crypto ($650M+ generated and 70% of all perps volume at peak),
Hyperliquid didn’t just grow in 2025 — it conquered.
Here’s exactly how they pulled it off (and why 2026 will be the real test) 👇
~~ Analysis by @kenzimori ~~
Q1 2025: The Crypto-Native Advantage
Hyperliquid's year of breakneck growth kicked off with a reminder of what it means to truly have your finger on the pulse.
When TRUMP launched in January, Hyperliquid had perps go live almost immediately, beating other exchanges to the punch and beginning its streak as the place for pre-launch tokens. It was able to move quickly because it stands somewhat unimpeded by corporate guardrails, but a significant element was being emphatically "in the know," spotting opportunities due to its team being tightly interwoven with the goings-on onchain.
February brought the HyperEVM launch, the general-purpose smart contract layer built on top of HyperCore. While it took time to find its footing, it did so without any top-down incentive programs, building a core user base who believed in the chain's vision rather than just extracting incentives.
Q2 2025: Breaking Out
Traction came faster than most expected. Beyond HYPE climbing nearly 4x off its April lows, by May Hyperliquid commanded 70% of all onchain perps volume, a staggering figure for a platform with zero VC backing and no token incentives.
As the market roared back, Hyperliquid's smooth UX and deep liquidity captured the order flow, with total volume climbing to $1.5T. The HyperEVM hit its stride alongside, growing TVL from $350M in April to $1.8B by mid-June as projects launched through @kinetiq_xyz,
@felixprotocol, and @liminalmoney, all while burning HYPE in the background.
Amid this growth, Hyperliquid seemed to be everywhere. Publicized on national TV. Profiled by Bloomberg. At the center of policy conversations with the CFTC.
Q3 2025: Peak Momentum & Splintering Begins
Q3 opened with a signal that Hyperliquid's infra was becoming essential outside its own ecosystem.
@phantom Wallet integrated Hyperliquid via builder codes, Hyperliquid's mechanism for letting external platforms earn fees on trades they route to HyperCore.
@Rabby_io followed. Then @MetaMask. A myriad of mobile trading apps went live on builder codes. "Partners" have earned nearly $50M in fees through these integrations, routing $158B in volume.
Then, in September, came the USDH bidding war, revealing just how valuable Hyperliquid had become.
The problem was simple: Hyperliquid held ~8% of Circle's USDC supply in its bridge, leaking roughly $100M annually to Coinbase while seeing none of that yield recycled into its own ecosystem. A native stablecoin would fix that, potentially redirecting $200M in annual revenue back to Hyperliquid.
Heavyweights threw their names in. Ethena offered $75M in growth commitments. Paxos dangled PayPal and Venmo integrations. But @nativemarkets won, a team led by HYPE contributor @fiege_max, former Uniswap Labs COO @Mclader, and Paradigm researcher
@anishagnihotri. Why? They fit the ethos: bootstrapped, aligned, and ready to build something organic.
The ripple effects extended beyond Hyperliquid. MegaETH announced its own native stablecoin initiative shortly after. Sui followed suit in November.
Yet USDH also marked HYPE's peak in mid-September, and the moment competition began to bite. Aster and Lighter both launched with aggressive airdrop campaigns. Hyperliquid's market share splintered, sitting at 17.1% at time of writing.
Q4: Maturation & Growing Pains
In October, HIP-3 went live, opening permissionless listings on HyperCore and advancing both the exchange's expansion and its decentralization.
Anyone who stakes 500K HYPE could now deploy custom markets such as equity perpetuals, markets using yield-bearing collateral from Ethena, or markets for synthetic exposure to private companies from @ventuals.
Yet, despite HIP-3's launch, HYPE price has dropped nearly 50% from its September peak. Besides market conditions and competition, two developments stand out.
First, the quarter brought Hyperliquid's first ADL (Automatic De-Leveraging) event in over two years. During October 10's market breakdown, over-leveraged positions ran out of margin faster than the liquidation system could absorb. The protocol triggered auto-deleveraging over 40 times in a 12-minute span. While the system stayed solvent, Hyperliquid will likely need time to recover from the event.
Then, in November, team token unlocks began. Despite lower-than-expected totals, this vesting is likely contributing to HYPE's underperformance. Selling was minimal, only 23% went to OTC desks while 40% was re-staked, but the pace of future unlocks remains unclear. From a protocol that's stood out by being transparent, this lack of clarity is likely causing market unease.
The Perps Proving Ground
While the market and trading activity are down, it's important not to discount how much the perps landscape has evolved alongside Hyperliquid itself.
@Lighter_xyz and @Aster_DEX offer real alternatives. Offchain, Coinbase's perps offerings will soon be joined by Robinhood. More competitors will emerge as perpetuals continue going mainstream.
Hyperliquid is in the midst of its proving ground and will continue to be in 2026. The question isn't whether it had a remarkable 2025, it certainly did. The question is whether the exchange can demonstrate that its model remains superior as the field gets crowded.
What got them here was building a better product and ecosystem without shortcuts. What keeps them there will be doing it again.
Introducing Uninvited.
starting a weekly read with long reads, interviews w the most interesting people in private markets, plus a read of the room
subscribe if you:
- want to break in (VC, PE, growth)
- already in, want culture on top
you're invited -> https://t.co/TJclz7dElJ
In a volatile year for DeFi, Coinbase’s L2 kept compounding builder momentum while other chains stalled...
Finishing #1 in L2 revenue with $82.6M earned, $4.3B in DeFi TVL, and $4.8B in stablecoins.
Here’s a closer look at the products, programs, and upgrades that made Base's year one for the books👇
~~ Analysis by @punk7954 ~~
1. Growing the DeFi Mullet
While @base has grown alongside Coinbase, Coinbase has also leaned on Base, using it to offer a unique suite of hybrid onchain/offchain products (commonly referred to as the DeFi mullet) that apply centralized finance convenience to DeFi flexibility.
Here are some of the top efforts:
➢ Onchain Loans — Through Coinbase, U.S. users can borrow up to $5M in USDC against BTC or up to $1M against ETH, enabled by the exchange's direct integration with the Morpho lending protocol on Base.
Your BTC auto-converts to cbBTC, deposits into Morpho, and your USDC arrives in under a minute. Compare this against the week-long process of applying for a bank loan and it becomes clear how superior crypto's systems are. So far, this feature has seen $1.5B borrowed and $1.6B in collateralized BTC, with nearly 19K borrowers to date.
➢ Yield Opportunities — Beyond borrowing, Coinbase users have access to a series of yield options. They can deposit USDC directly into @Morpho vaults to earn yields higher than typical 4% savings rates, timed nicely as rate cuts reduce traditional yields. For those seeking maximally compliant liquidity venues, Verified Pools offer specialized @Uniswap v4 pools open only to Coinbase-KYC'd users, optimized by risk-management firm @gauntlet_xyz.
➢ DEX Integration — Coinbase also integrated DEX trading directly into its exchange, allowing U.S. users (minus New York...) to trade Base tokens without leaving the app. When you make your first trade, the app creates a self-custody wallet — you hold the keys, but the UX looks just like buying tokens normally on the exchange, with all swap fees sponsored. For traders, it's onchain access without the usual friction. For Base projects, it's expanded distribution to millions of more users.
Taken together, the combo of accessible loans, competitive yields, and DEX integration positions Coinbase to function as much as a neobank as it is an exchange, a bold vision enabled by Base.
2. Scaling Builder Tools & Programs
Base stands out for its builder-first growth ethos. Instead of elaborate governance structures, the chain has developed tools and programs that reward and support existing builders and lower the barrier for new ones.
➢ Builder Programs — Base organizes a global builder program called Base Batches for pre-accelerator talent, with dedicated tracks for AI, stablecoins, and consumer apps. Structured around a series of hackathons, the program aims to provide early-stage support and possible pathways to incubators and venture funding. Meanwhile, Base Build offers a development dashboard with real-time user analytics from launch, along with a monetization option through builder rewards.
➢ Toolkits — For plug-and-play infrastructure, Base provides a set of tools, including Embedded Wallets for account abstraction, Base Pay for USDC checkout within applications, and Sign on with Base, which streamlines account creation and onboarding.
➢ x402 — Perhaps the most forward-looking addition has been x402, an open payments standard that allows payments to be attached to web requests. The protocol solves the hassle of traditional API access by letting you pay automatically based on usage, per-call or per-inference, with payment happening instantly as part of the request. This enables AI agents to act as true, autonomous service providers, managing all the expenses and access they need to run operations independently.
The through line here is accessibility. By equipping builders to ship faster, Base stimulates its onchain economy, makes it easy for people to experiment onchain, and positions their L2 as a positive-sum environment.
3. Shipping New Technical Upgrades
Beyond brand power and product launches, Base kept making under-the-hood improvements, strengthening security, slashing latency, and expanding connectivity.
➢ Decentralization — In April, Base hit Stage 1 decentralization with the launch of permissionless fault proofs and the implementation of a security council. This strengthened the chain's security and reduced its trust assumptions, particularly important given the chain's close relationship with a shareholder-owned public entity.
➢ Scaling — July brought the debut of Flashblocks, which slashed block times from 2 seconds to 200 milliseconds, moving transactions into "instant" territory. In December, Base launched a native Solana bridge using @chainlink CCIP, enabling SPL token support within Base applications and cross-chain actions where transactions on one chain can trigger transactions on the other. It's the first non-Ethereum chain connected to Base, positioning it as a hub where every asset can exist across every network.
These upgrades lay the groundwork for Base to compete on performance, not just distribution, while reducing the trust assumptions that come with being tied to a centralized exchange.
4. Laying Out the Next Phase of Base
Beyond infrastructure and product releases, two big developments stood out: the first showcasing a new method for traversing the chain, and the second (soft) confirming something many had hoped for from the beginning.
➢ Base App — Base launched the Base App in July and has expanded it to 140+ countries, replacing Coinbase Wallet as the ecosystem's primary interface and signaling a shift toward social-first crypto. The app integrates chat, trading, social features, and mini-app discovery all wrapped up in one familiar feed. It currently has 169K registered users, and while many are certainly there to farm the chain's upcoming token, there is a core group of users embracing its call to "just coin it," and tokenize their media onchain.
➢ BASE Token — At Base Camp in September, the team publicly confirmed they're "exploring" a native token. Given the soft framing, it's likely it will be some time before a BASE token goes live. In the meantime, airdrop hunters are getting active as it's almost certain that onchain activity will be rewarded if they pursue an airdrop as part of a possible token launch.
Together, these moves sketch a vision of Base as a living ecosystem with its own culture and incentives, one where participation, especially as a builder, can pay off handsomely.
Looking Ahead
Base's 2025 was defined by stacking advantages.
The symbiosis with Coinbase has proven mutually beneficial: Base gets a built-in distribution mechanism and trusted brand for onboarding newcomers, while Coinbase gets the infrastructure to offer hybrid products that meaningfully distinguish it from competitors. At the same time, the chain's builder-first ethos has cultivated a positive-sum environment that stood out during a year that was all over the place. And with Stage 1 decentralization achieved, Base has made meaningful progress in reducing the trust assumptions that naturally accompany an exchange-backed L2.
The result is the most successful exchange-backed L2 to date, built on a foundation that positions it well for the next phase of growth. Its most credible near-term challenger may be Robinhood's L2, which, if launched, could prove to be an ultimate test for the chain's staying power.
It may still be day one for the chain, but the momentum suggests it is building toward something much larger.
.@VitalikButerin just made a bold declaration: ETHEREUM HAS SOLVED THE BLOCKCHAIN TRILEMMA
He credits a new combination of technical upgrades that are working together to push the network’s scaling limit – without sacrificing decentralization or security.
Today, we’re exploring where Ethereum is headed next. 👇
~~ Article by @punk6583 ~~
🔍 Ethereum’s Scaling Future
All networks make tradeoffs between decentralization, security, and scalability.
While peer-to-peer file sharing platform BitTorrent is highly decentralized and very scalable, it has no single source of truth that forms network consensus, providing weak security guarantees.
Although Bitcoin achieves strong decentralization and security guarantees, the network has an extremely slow throughput of approximately 7 transactions per second.
And while the Solana network delivers Bitcoin-like security at much faster speeds, the network’s hefty hardware requirements restrict participation to commercial-quality node operators, limiting decentralization.
The blockchain trilemma constrained the development of a balanced blockchain that simultaneously optimizes for decentralization, security, and scalability in each of these examples.
Yet, Vitalik Buterin believes Ethereum can achieve where others have failed using zkEVMs and PeerDAS.
🔒 What are zkEVMs?
Zero-Knowledge Ethereum Virtual Machines (zkEVMs) are a type of scaling solution that executes Ethereum transactions offchain and proves their correctness back onchain using cryptographic zero-knowledge proofs.
Instead of individually submitting transactions to the Ethereum L1, zkEVMs bundle thousands of transactions together in a proof, which accounts for the results of transactions while circumventing the need for blockchain computation. This design increases scalability by using offchain execution, while preserving Ethereum-grade security via onchain proof verification.
Vitalik characterizes contemporary zkEVMs as “alpha stage,” claiming they enjoy production-quality performance, but require safety work. He expects their full benefits to emerge over the next 4 years, with the first opportunities to run zkEVM nodes occurring sometime in the new year.
📡 What is PeerDAS?
Peer-to-Peer Data Availability Sampling (PeerDAS) is Ethereum’s latest leap in data scaling. It was adopted to mainnet on December 3, 2025, and served as the headline improvement of Ethereum’s recent Fusaka upgrade.
Rollups need to publish transaction data to Ethereum so the network can independently verify state transitions, and PeerDAS dramatically reduces the cost of making this data available by allowing nodes to sample small pieces of data from peers, rather than downloading entire data blobs.
PeerDAS (live today on mainnet) allowed Ethereum to increase the dedicated “blob” storage space for data availability with no changes to the hardware requirements for nodes that want to validate the Ethereum L1.
🧭 Where to Next?
Over the next four years, Vitalik believes that Ethereum can become the first network to fully conquer the blockchain trilemma. Such a breakthrough is not anticipated to be the outcome of a single sudden achievement; rather it is expected to be the gradual result of sustained progress over time.
Ethereum has already expanded data availability through PeerDAS, and while developers expect initial zkEVM node experiments to begin this year, Buterin doesn’t anticipate zkEVMs will become a dominant method of block validation until later this decade.
Although execution risk remains, should Vitalik’s scaling vision materialize, Ethereum could very well redefine what’s possible for decentralized systems, offering a model where security, decentralization, and scalability reinforce one another instead of competing.
@kenzimori sits down with @nickemmons, Co-Founder of @AlloraNetwork, to unpack the early conviction that pulled him into crypto — not as a trend, but as a new way to rethink power, coordination, and how systems are built.
Before founding Allora, Nick led blockchain development at a major asset manager and insurance firm across North America, during a time when most enterprises were still focused on private chains and consortiums.
Instead, he pushed toward public Ethereum — building real enterprise use cases around decentralized insurance and trying to capture what actually made blockchain powerful.
From those early experiments to the AI x crypto frontier, this conversation traces how Nick’s view of decentralized systems kept expanding.
@AlloraNetwork has now raised $35M in total — but the latest $3M round wasn’t just about capital.
@sachimiyasaki speaks with @nickemmons about why this round was built around strategic alignment: bringing in AI-forward partners, funds, and individuals who understand where decentralized collective intelligence is heading.
The goal isn’t only to grow the team or move faster. It’s to surround Allora Network with people who can help shape the next phase of AI coordination.
AI doesn’t need another isolated model.
It needs a network where intelligence can compound.
@kenzimori in conversation with @nickemmons, Co-Founder of @AlloraNetwork, on breaking AI out of silos, turning fragmented models into collective intelligence, and building the coordination layer that could move AI beyond the control of a few monoliths.
Web3 and AI are both hitting an inflection point at the same time.
AI is redefining intelligence, automation, and how people interact with technology. Crypto is building the economic rails for that intelligence to coordinate, transact, and operate in open networks.
@dikshawells in conversation with @nickemmons (Co-Founder, @AlloraNetwork) on why the overlap between AI and crypto may become one of the biggest design spaces of the next decade — and why we’ve barely scratched the surface.
“Proof of Collaboration = how strong the swarm is. Proof of Contribution = what each agent actually moved, with permanent on-chain audit trails.”
@ronbodkin (Founder, @TheoriqAI) joins @sachimiyasaki to break down trusted performance in Theoriq: actions are committed on-chain as non-repudiable evidence, and evaluators use transparent scoring rules over the full history—while the system stays open for specialized eval agents.
In 2017, I stepped into Google Cloud’s CTO Office because I could feel the shift coming. AI wasn’t a feature — it was the next operating layer of the world. Google was leading that wave.
@kenzimori in conversation with @ronbodkin (Founder, @TheoriqAI) about the Google years that sharpened Theoriq’s vision — and the early signals that made the AI trajectory impossible to unsee.
“DeFAI = DeFi as an agent economy: set the strategy, let agents execute, watch feedback in real time.”
@sachimiyasaki x @ronbodkin (Founder, @TheoriqAI) on how AI-run DeFi could bring smart-money infrastructure to everyone — not only institutions.
“Responsibility means steering crypto + AI toward outcomes that benefit everyone — and giving the community real power to set the course.”
Our host @dikshawells in conversation with @ronbodkin (Founder, @TheoriqAI) on why responsibility in crypto + AI starts with governance from day one — so the future isn’t dictated by monopolies or closed-door incentives.