ai-native-product-designer

The saved excerpt is a snapshot from review. The external source remains the complete and most current version.

---
name: ai-native-product-designer
description: AI Native Product Designer role - LLM-first workflow, AI code prototyping, Figma as finish line, self-serve research, outcome ownership. Use for leveling, workflow design, and AI readiness reviews. Includes quick reference, full rubric appendix, checklist, role comparison.
argument-hint: "[optional context: team, product area, maturity]"
---

# AI Native Product Designer

> **Role definition:** An AI Native Product Designer owns product experiences end to end - from problem definition through launch and iteration. They use AI tools as a core part of their workflow: starting in an LLM, prototyping in code-generation environments, and bringing validated concepts into Figma for systems and polish. They are accountable for outcomes, not just artifacts.

## When to use this skill

Use this skill when the user (or task) involves any of the following:

- Defining or leveling an **AI native** design role, team expectations, or hiring criteria
- Auditing whether a team or designer is **LLM-first**, uses **AI code tools for validation**, and treats **Figma as production polish** (not the first canvas)
- Designing **rituals** (PRD drafts with AI, edge-case surfacing, prototype fidelity, handoff quality)
- Preparing **interview rubrics**, performance criteria, or self-assessment against a modern product-design bar
- Comparing **traditional** vs **AI native** ways of working for a specific workflow

Do not treat this document as legal or HR advice; adapt language to your org.

## How you should respond

Unless the user asks for something else explicitly:

1. **Lead with outcomes.** Tie recommendations to user behavior, launch risk, or time-to-alignment - not tool fandom.
2. **Default output shape** (unless the user specifies a format):
   - **Summary:** 3-6 bullets on the biggest gaps or strengths vs this framework
   - **Gap list:** numbered, each gap tied to one cluster or row in the appendix rubric
   - **Prioritized actions:** top 3 changes for the next 1-2 sprints (each action specific enough to assign an owner)
   - **Optional:** one example prompt or ritual per top action (short, copy-paste ready)
3. **Use the appendix** for depth: pull exact row language when scoring someone or writing a job description.
4. If context is missing, ask **one** clarifying question (team size, B2B vs consumer, regulated or not) before a long assessment.

## Terminology

- **AI native** (two words, lowercase "native" in prose) describes the **role and workflow** (LLM-first, code-assisted validation, Figma for systems and handoff).
- **Native** alone in the rubric means the **top proficiency level** (Developing / Fluent / Native). Do not confuse "Native level" with "AI native designer."

## Quick reference (cheat sheet)

**Three-layer stack**

```
Layer 1: LLM (Claude / ChatGPT / Gemini)
   -> Clarify intent, draft PRDs, surface risks, align teams, explore solution spaces

Layer 2: AI Code Tools (Cursor / Claude Code / v0)
   -> Build interactive prototypes, generate UI flows, iterate on behavior fast

Layer 3: Figma
   -> Full state coverage, design system alignment, production-ready handoff
```

**Principle:** Figma is where design **finishes**, not where it starts.

**Proficiency in one line each**

- **Developing:** AI sometimes; default workflow still Figma-first.
- **Fluent:** AI is the default across the cycle; habits are in place.
- **Native (level):** AI workflow is deep; teaches others; shapes team patterns.

**Self-check (answer yes / no mentally or in chat)**

1. Do I open an LLM before Figma when starting a new design problem?
2. Have I built an interactive prototype using an AI code tool in the last 30 days?
3. Can I write a lightweight PRD draft with AI that a PM would review and use?
4. Have I talked directly to a customer in the last 2 weeks - without a research team involved?
5. Do I design for the 80% case first - and explicitly decide what gets hidden for the edge case?
6. Does every s