🚀 From Terra’s Ashes to $50M in Fees: How Based Found PMF on Hyperliquid

Why this matters

Across crypto’s most turbulent cycle, one team’s journey tracks the industry’s tectonic shifts: consensus wars (2018), DeFi summer (2020), Terra’s collapse (2022), USDC’s depeg and FTX (2023), and the rise of on-chain perps marketplaces (2024–2025). The result is a case study in resilience, ruthless focus, and business-model selection — culminating in a specialized trading wallet on Hyperliquid with real traction.

“In the last eight months we did around close to 40 billion in volume, which is roughly around $50 million in revenue.”

Origin story: from throughput to use-cases

• Engineering roots: Edison, co-founder of Based, entered crypto in 2018 as one of the first engineers at Zilliqa, working from consensus R&D to applications. The era obsessed over engines (Byzantine fault tolerance, Proof of Stake/Work) before the market asked why the engine should be used at all.
• Bear market apprenticeship: Singapore’s 2018 cohort (Kyber, Zilliqa) seeded a network of founders who later built Jupiter, Pendle, Defiance Capital, and more — a “mini PayPal mafia” effect.

DeFi, Terra, and a hard reset

• DeFi pull: 2020’s wave (Yearn, Sushi) drew the team deeper into on-chain strategy.
• Terra thesis: Luna-UST design, fast UX, and ~20% APR on Anchor made the ecosystem feel like the future. Arbitrage strategies (including the bLUNA-LUNA 21-day arb) were ubiquitous. As one anecdote put it: “we were making like 1.5 or 2% ... every 21 days.”
• Product #1 (Subra): a self-repaying model to DCA into BTC, buy Mirror stocks, or subscribe to services while Anchor’s yield covered principal — “we went live in March, for 40 days, and then the whole Terra collapse happened.”

“The collapse of Luna resulted in me losing 95% of my net worth.”

Host context framed the scale: “$40 billion in Luna and $20 billion in UST gone overnight.”

Survival pivots: payments, cards — and hard lessons

• Payment gateway (Suba): “Stripe for Web3” with on-chain subscriptions and Shopify stablecoin checkout. The team raised ~$2 million (led by Spartan and Hash) and extracted company funds at ~70 cents on the dollar during the chaos — enough runway to build.
• Macro headwinds: “USDC had a depeg… [SVB] …USDC pegs to 85 cents… very difficult time to build any stablecoin business.” Merchant adoption stalled.
• Visa x self-custody card (2023): smart-contract wallet (ERC-4337) with a non-custodial Visa card; Singapore-focused launch with an orange card, SSG (a SGD stablecoin) and later USDC. Economics were viable via interchange: “around 0.5% all the way to 2%.”
• Unit-economics reality: card businesses carry non-zero CAC and ops costs (shipping, activation). It turned profitable with scale — until an April 2025 fraud hit: “someone used our cards… charged us more than authorized… the company absorbed that bad debt.”

The decisive turn: specialize on Hyperliquid

Rather than another generic, multi-chain wallet, Based chose depth over breadth: build a high-performance, single-chain trading experience for Hyperliquid, aimed squarely at the “retail pro.”

“What if this whole wallet only works very well on one chain — and that chain is Hyperliquid.”

• Monetization via Builder Codes: The front-end captures spread/fee economics while Hyperliquid provides the execution layer. Quote: “If people pay 4.5 basis point for taker fees on Hyperliquid, we charge like 2.5 basis point on top — around seven.”
• Competitive stack: Binance is cited at “around six [bps]”; Coinbase is “very competitive,” making Based’s ~7 bps in line with major CEX pricing for the target user.
• Distribution engine: A referral-led growth loop (inspired by Bybit/FX CRM playbooks) substituted for an expensive field sales org and bootstrapped network effects.

Product-market fit: the numbers

• Eight-month clip: “close to 40 billion in volume,” generating “roughly around $50 million in revenue.”
• Positioning: Built for the retail pro — traders comfortable with five- to six-figure sizes who demand order books, funding, and robust mobile UX. Not the UI-minimalist retail (Phantom, MetaMask), nor fee-obsessed institutional pros who go direct.

“A platform that allows you to trade everything and spend everywhere.”

Ecosystem build-outs: Hyena Trade with Ethena

Based partnered with Ethena on Hyena Trade, a Hyperliquid H3 deployment with a USDe-quoted market design.

• Roles: Ethena procures what’s required to operate the H3 exchange and bears market deployment costs; Based built the entire front end and UX to scale the venue.

Strategy doctrine: revenue first, then scope

“Some things are cool, but it may not generate the revenue at the end of the day.”

• Learned constraint: Focus on solvable distribution, durable fee capture, and chain-aligned incentives (Builder Codes) over speculative multi-chain sprawl.
• Last-shot clarity: With “200k in the bank” and memoed closure plans, the team made one final, concentrated bet. Quote: “This was our last shot.”

AI agents are coming: build the financial OS now 🤖

Based’s next arc moves beyond humans. The wallet becomes a programmable control plane — identity, permissions, payments — for autonomous agents.

• Agent permissions: Smart accounts (ERC-4337) to grant scoped trading/spending rights; potential for virtual cards controlled by agents.
• Standards and settlement: Reference to “S42” as a payments standard for agent-to-agent settlement.
• Inference payments live: The team introduced a feature for wallets to pay for AI inference using USDH or USDC on Hyperliquid.

“Every human is going to have at least one personal assistant that is an AI agent.”

Contextual quote bank

• On fee model and parity: “We charge like 2.5 basis point on top… around seven… Binance is around six [bps].”
• On the pivot thesis: “What if this whole wallet only works very well on one chain — and that chain is Hyperliquid.”
• On growth mechanics: “Social virality is very important… empower a referral model.”
• On survival: “We managed to get [company funds] out around like 70 cents on the dollars.”
• On card economics: “You also get a revenue stream… around 0.5% all the way to 2% on interchange fees.”
• On the Terra shock: “We went live for 40 days and then the whole Terra collapse happened.”

Risks and watch items

• Concentration risk: Heavy reliance on Hyperliquid’s growth, stability, and Builder Code sustainability.
• Fraud and credit risk in payments: The April 2025 incident underscores non-trivial downside tails in card programs.
• Referral loop durability: Growth hinges on ongoing incentivization versus organic retention.
• AI timing risk: Agent capabilities and standards (e.g., “S42”) must mature to support the operating-system vision.

The takeaway

This is a blueprint for crypto founders operating in hostile cycles: pick the right business model, let incentives do distribution, and specialize for one ecosystem until the economics work. As Hyperliquid pushes an “AWS of liquidity” paradigm, expect more front-end empires to emerge — but few will match Based’s combination of velocity, fee capture, and a credible path from “trade everything” to an agent-native financial OS.