# Design at the Speed of Adjectives — Paul Bakaus, Renaissance Geek, Inc.

## Executive summary

The talk introduces 'Impeccable,' a design skill designed to enhance coding harnesses (like Copilot and Claude Code) by providing a vocabulary to steer AI-generated design. The core thesis is that good design cannot be 'oneshot' or fully automated because it is context-rich, iterative, and requires human decision-making. Instead of automating the process, Impeccable allows users to inject specific adjectives and verbs (e.g., 'bolder,' 'quieter,' 'distill,' 'harden') to guide the AI toward specific design goals, thereby improving the quality of the human-AI collaboration.

## Key takeaways

- Design is not a one-shot process: Good design must be context-rich and iterative. The speaker argues that fully automating design is currently impossible, as human judgment is required to define the emotional territory and audience.
- Impeccable facilitates steering, not abdication: The tool's purpose is to give users the necessary control to steer the agent using specific design vocabulary. The speaker explicitly rejects the idea of an automatic mode, stating that the point is guiding the process, not letting the AI do all the work.
- The traditional design handoff is collapsing: The traditional 'waterfall' process (PM $ ightarrow$ Design $ ightarrow$ Engineer) is rapidly breaking down, leading to a blurring of roles where engineers and designers must increasingly work in shared, fluid processes.

## Technical details

- Impeccable Functionality: Impeccable is a design skill that works across multiple coding harnesses (Claude Code, Copilot, Cursor, Codex, etc.). It allows users to apply specific design verbs and adjectives (e.g., 'bolder,' 'quieter,' 'distill,' 'polish,' 'harden') to guide the AI's output. For instance, 'bolder' resolves to hierarchy, scale, and decisive type, rather than generic gradients or glass effects.
- Design Workflow Mapping: The speaker mapped the messy design workflow into distinct, non-linear stages: initialization/shaping $ ightarrow$ crafting $ ightarrow$ iterating $ ightarrow$ hardening $ ightarrow$ design debt cleanup. Impeccable provides injection points along this entire route.
- AI Design Slop: The speaker notes that AI-generated design tends toward 'algorithmic unilo' (e.g., 'Claude beige' palettes, Instrument Serif fonts, and italics), which is a predictable, yet often bland, pattern.

## Practical implications

- Build engineers should view design as a continuous, multi-stage process (shaping, crafting, hardening) rather than a linear handoff.
- When integrating AI tools, focus on building 'steering' mechanisms (vocabulary/prompts) rather than attempting full automation.
- The concept of 'hardening' (ensuring design works across devices and maintains responsiveness) must be treated as a critical, explicit step in the development workflow, not an afterthought.
- The role of the engineer is shifting to one of design director, requiring a deep understanding of design language and emotional territory.

## Topics

AI-Assisted Design, Software Development Workflow, Design Systems, Prompt Engineering, Human-AI Collaboration, Paul Bakaus

Source: https://www.youtube.com/watch?v=v42opQpCy60