Framework
Prototype and Prune
AI is a proxy for skill, not taste. Generate multiple enterprise-appropriate prototypes using real design system context, then let a human with taste prune down to the right one. Coined by Jules at Meta, reinforced by Scott Belsky. The bottleneck moves from production to curation — and curation is the irreducible human job.
How it works
- 1The old comp-to-handoff waterfall is replaced by a generation-and-curation loop: AI generates many valid options, humans select.
- 2Brad Frost's cocktail napkin vision: everyone becomes a designer — but only if the generation is grounded in real design system context and rules.
- 3Taste is the scarce input. Skill (technical execution) is abundant and becoming more so. The designer's new job is judgment, not production.
- 4The bottleneck moves from 'can we build this?' to 'which of these options is right for this context?' — a fundamentally different constraint.
- 5Knapsack's context archetypes and rules engine make AI generation enterprise-appropriate: the system knows which prototypes are WCAG-compliant, on-brand, and legal-approved before a human ever sees them.
Hear it on the podcast
Stop Designing Screens, Start Designing PatternsListen to the episodeMore from this episode
The Intelligent Product EngineA three-part architecture that makes AI generation predictable and enterprise-appropriate: context archetypes (different rule sets for different modes), real-time artifact monitoring (Figma and git sync to structured artifacts for humans and AI), and a rules engine (WCAG, brand governance, legal gates, release criteria). Each part is necessary; together they turn generative AI from a prototype toy into a production tool.Design Systems ThinkingStop designing individual screens. Start designing systems of experiences. Every interface element is a pattern with explicit rules and constraints — not a one-off creative decision. This shift from instance-level to system-level abstraction is what enables consistent product experiences at scale and makes AI generation enterprise-appropriate.
Get the next framework first
One email a week with the frameworks, tools, and founder stories from the podcast. Free.
More frameworks for Building the Product
Sprint ZeroBefore writing code, conduct a discovery phase that audits pirate metrics, forces explicit assumption validation, and produces a 1–1.5 month validated roadmap. Sprint Zero converts founder conviction into actionable direction. Almost every founder arrives with a three-month roadmap and no data to support it. Sprint Zero surfaces what is real and what is assumption — and the roadmap that comes out is one the team can build against with confidence.Engine + NLP + AutomationModern product development converges three capabilities: an existing engine (platform, open-source framework, or no-code tool), natural language AI (NLP you can prompt rather than program), and trigger-based automation (API integration that has existed for 15 years and is now commercializing at scale). Build around the right engine and connect components rather than coding from scratch. In 2025, building from zero is immature.Requirements-First DevelopmentThe CTO's core job is not writing code — it's translating business objectives into technical requirements, then ensuring the code solves the right problem. Start there. Generating code before generating requirements produces something technically functional that solves the wrong problem. The 70% wall is the symptom; deferred requirements are the cause. Requirements elicitation — guided prompting to extract what you actually need — is what separates MVPs from prototypes.The 70/30 Software StackSeventy percent of every new software application is commodity: authentication, user management, CRUD operations, data pipelines, notifications. It was never innovative. Every dev shop had internal boilerplates for it. The competitive advantage in software has always come from the 30% that's unique. In 2026, this matters more: any decent CRM can be built in a day. Automate the 70%; focus all human effort and genuine IP development on the 30%.The Whole Person DiagnosticEffective diagnosis requires integrating three data streams that no specialist silo currently combines: longitudinal medical records (what the healthcare system has captured), wearable data (continuous biometrics between clinical encounters), and life story — the mind-body layer of trauma, stress, and relational context that the research literature documents as medically relevant but that EHRs structurally exclude. Consumer ownership of the record is the architectural enabler: the patient integrates across institutions because the institution won't.Meet Consumers Where They AreDo not build a new interface and ask customers to learn it. Identify the medium where your target customers already interact with this type of service — email, text, phone call — and build the product to operate entirely in that medium. The lower the adoption friction, the higher the trust, and the faster the referral loop. No app to download, no portal to navigate, no new UX to learn means less resistance at every stage of acquisition.