The New PMF Playbook (6 minute read)
Three product-market fit strategies are working in crypto right now: co-building with major financial institutions, positioning infrastructure for the AI agent economy, and dogfooding your own technology before seeking external adoption.
What: Jason Rosenthal outlines three PMF patterns succeeding in crypto: partnering with sophisticated TradFi customers whose requirements define your product, building infrastructure ahead of emerging curves like AI agents as economic actors (exemplified by AgentCash's x402-based payment rails for agents to pay for API access), and being your own first customer like ZKsync did with Prividium to power Cari Network for banks including Huntington, First Horizon, M&T Bank, KeyCorp, and Old National moving deposits on-chain.
Why it matters: These patterns show how crypto companies can avoid the trap of mistaking token mechanics and airdrops for real product-market fit, instead finding validation through institutional adoption and solving concrete problems ahead of market consensus.
Takeaway: Evaluate which of these three patterns fits your product and commit to it rather than iterating without clear direction.
Deep dive
- Pattern one involves partnering with elite customers like major financial institutions whose requirements become your product specification, trading speed for adoption quality since one customer handling trillions in daily volume provides more validation than retail attention
- Pattern two focuses on identifying exponential curves early and positioning infrastructure before the market fully understands the opportunity, specifically targeting AI agents becoming autonomous economic actors
- AgentCash exemplifies the second pattern by building payment infrastructure on x402 protocol that lets AI agents pay for API access with crypto, enabling programmatic transactions without human-managed billing systems
- Pattern three follows Amazon's AWS playbook of being your own first customer by building applications on your own infrastructure to prove capabilities before asking external developers to adopt
- ZKsync demonstrates this approach with Prividium anchored to tokenized deposits via Cari Network, allowing major U.S. regional banks to move customer deposits instantly across institutions on blockchain while funds remain in the regulated banking system
- The article warns that more capital just extends runway to bad outcomes and that growth hacks and continuous airdrops disconnected from strategy substitute for admitting you haven't found PMF
- Token mechanics and network effects that make crypto powerful can actually mislead companies on their path to product-market fit
- Strong crypto teams are finding PMF faster now due to killer apps like stablecoins and widespread TradFi adoption
- The human-in-the-loop assumption for AI systems is breaking down faster than expected, creating opportunities for infrastructure serving autonomous agents
- The core thesis is that the fastest path to PMF involves choosing the right strategic game and executing with conviction before consensus forms
Decoder
- PMF: Product-Market Fit, when a product resonates with a broad set of customers and solves a real need
- TradFi: Traditional Finance, referring to established banks and financial institutions outside of crypto
- x402: A payment protocol designed for programmatic API access and agent-to-agent transactions
- TVL: Total Value Locked, a metric measuring the total value of assets deposited in a crypto protocol
- ZKsync: A Layer 2 blockchain scaling solution using zero-knowledge proofs
- Prividium: ZKsync's enterprise privacy and infrastructure product for institutional clients
- On-chain: Transactions or operations executed and recorded on a blockchain
- Agentic economy: An emerging economic model where AI agents autonomously transact and deploy capital
Original article
Three PMF patterns that are currently working: co-building with elite TradFi institutions whose requirements define the product spec, positioning infrastructure ahead of the AI agent economy, and dogfooding your own rails before seeking external adoption. AgentCash illustrates the second pattern by building x402-based payment infrastructure so AI agents can pay for API access in crypto. ZKsync's Prividium demonstrates the third, with Huntington, First Horizon, M&T Bank, KeyCorp, and Old National moving customer deposits on-chain via Cari Network.