The five-handoff framework covers a remarkable share of the current market. Not all of it.
Three live or near-live roles reveal buyer anxieties that have nothing to do with human-agent coherence. Lead with the Trust essay for these and you will sound relevant but imprecise. At Director+, imprecise is the same as generic. Generic loses.
Here is how to detect each exception from role language, and what changes in your lead proof when you do.
Exception 1: Org-Model Transformation
Detection signal: The posting describes how the design organization works, not how the product works.
Capital One's Director, Product Design — AI in Experience Design is the clearest current example. The role sits on the XD top-of-house team, working with the SVP and Chief Design Officer and senior leadership team. It owns roadmaps across "AI tooling, AI-enabled design process evolution, and AI-centric talent upskilling across Design at Capital One." The mission: "establishing a future vision for AI integration and accelerating AI adoption across the Experience Design organization."
Read that again. This is a workforce transformation role with a design title. The buyer's anxiety is whether the design organization itself can evolve its processes, tools, and talent to work with AI. The posting names "workforce transformation," "executive alignment," and "integration into a highly matrixed operating model." No handoff framework addresses this. The problem is upstream of the product surface entirely.
What changes: The Trust essay works as philosophy but cannot lead. Lead with the organizational architecture of what you built at scale. Alibaba's cross-functional restructuring. Thermo Fisher's five-module system redesign. Both demonstrate operating-model change, not just interface change. Your TinyFish context gives you current AI-tooling fluency for conversation, but published proof must come from cases where you changed how teams work, not just what they ship. The essay becomes supporting evidence. Not the opener.
Watch for these phrases: "Across the design organization" or "design process evolution" rather than "across the product surface." That shift from product to org is the signal.
Exception 2: Builder-Stage Prototyping Capacity
Detection signal: The posting emphasizes shipping and building over leading and scaling. No mention of team growth. Explicit absence of people management.
Anthropic's Design Engineer, Education Labs is a pure builder role wearing a design title. "Part researcher, part product builder, part interaction designer." Second technical builder on a small team. Will "operate as a one-person technical shop," "prototype new ideas, establish technical direction, and ship production-quality features to millions of users." The posting explicitly states the role does not involve people management "off the bat."
The buyer wants to know one thing: can you build. Prototype. Ship code. Move from research insight to production feature without a team of engineers translating your intent. A five-handoff framework presented to a hiring manager looking for a one-person technical shop will read as over-indexed on strategy for a role that needs hands.
What changes: Lead with Agentic Labs work. Brand Pulse and Carrier IQ as shipped AI products demonstrate builder credibility in written proof. TinyFish demonstrates current builder capacity but lives in conversation, not outreach, because it is current-role context rather than published portfolio evidence. The essay can surface later as evidence of systematic thinking behind the building. Not the door opener here.
Watch for these phrases: "Ship production-quality features" combined with "no people management." Builder roles sometimes carry senior titles. The absence of team language tells you more than the title does.
Exception 3: Design-System Infrastructure for AI Readability
Detection signal: The posting or company signal focuses on making design systems machine-readable, not on designing human-agent interactions.
Atlassian has no public design-system leadership posting naming this work right now. Confidence: speculative. Monitor. The signal comes from their blog on the design system as context engine. Their argument: fragmented design-language context across docs, Confluence, Figma, Looms, and code makes AI-generated Atlassian UI "unpredictable and expensive." Their solution requires "every component, token, and element" to carry "consistent, lightweight, machine-readable details." They report measurable results: 52% accuracy improvement on specific AI queries, 34% faster completion, 26% fewer tool calls.
This is a design-infrastructure problem. The Trust essay's handoff framework assumes a human in the loop. This work is about giving the AI enough structured context to participate coherently in the first place. Different layer. Different proof.
What changes: When a role appears, lead with the structural rigor visible across your case studies and the Trust essay's consistency argument applied at the infrastructure layer. Be direct about the gap: your published portfolio demonstrates platform-scale information architecture and systems thinking, but not machine-readable design-token infrastructure specifically. Alibaba shows structural redesign of a complex platform. It does not show design-system work in the Atlassian sense. Name that honestly if asked, and bridge with the principle: the consistency problem at the token level is the same consistency problem your essay describes at the interaction level, applied one layer down.
Watch for these phrases: "Machine-readable," "design tokens," "context engine," "AI-generated UI consistency." Any posting that treats the design system as an AI input rather than a human reference.
The Decision Protocol
New role description lands. Run this before you draft outreach:
- Does the posting describe product-surface interactions between humans and agents? Lead with the Trust essay. Route to the matching handoff for portfolio proof.
- Does the posting describe how the design organization should change? Org-model exception. Lead with operating-model cases.
- Does the posting emphasize building and shipping over leading and scaling? Builder exception. Lead with Agentic Labs. Keep strategy conversational, not written.
- Does the posting treat the design system as AI infrastructure? Design-system exception. Lead with platform-scale systems work. Monitor for the posting to materialize.
Most roles in the current market are Door 1. The essay opens them. The three exceptions will cost you precision if you treat them as handoff problems. Detect the difference before you write the first line.