top of page
SARAH A. SCHMIDT
UX Writer | AI Content Designer | AI Interaction Designer
I design how people interact with AI through language, workflows, and content systems.
AI Content Writer & UX Writer (Outlier AI) • Columbia University
GYRO • JWT • DDB
AI Systems
Designing how AI interactions, outputs, and user experiences behave across generation, evaluation, and decision-making.
Focus areas: AI UX Writing, Content Design, AI Interaction Design, Prompt Systems, User Flows, Content Systems, Output Evaluation
AI Interaction Systems
VOICE DESIGN AI PROTOTYPE
⸻
System Logic
AI voice creation requires more than generation—it requires trust, consent, customization, and user control.
⸻
HOME/LOGIN
JOIN VOICE DESIGN
VOICE READY
HOME/LOGIN
1/7
⸻
Method
* Design onboarding flow * Establish consent requirements
* Guide voice upload and analysis
* Structure customization options
* Support editing and export workflows
⸻
System Capabilities
* Guides AI interactions * Supports user trust
* Enables voice customization * Improves workflow clarity
* Creates consistent user experiences
⸻
What This Reveals
AI products succeed when users understand what the system is doing and maintain control throughout the process.
Also Demonstrates
UX Writing · Content Design · AI Interaction Design · User Flows · Information Architecture · Figma
⸻
MULTI-TURN INTERACTION SYSTEM
⸻
System Logic
AI interactions aren’t single responses—they’re sequences.
Clarity emerges as intent is refined across turns.
⸻
Simplified Flow
Input → Intent → Response → Next
Full System Flow
Input → Intent → Clarification → Response → Next Step → Fallback
Before / After
Before (Unstructured Response)
* Generic output * No guidance * Dead end
After (Structured Interaction)
* Clarifies intent * Guides next step * Adaptive flow
From reactive responses → guided, adaptive interactions
⸻
Method
* Map interaction flows * Define intent + edge cases * Structure for clarification + recovery
* Evaluate across turns
⸻
System Capabilities
* Maintains state * Disambiguates intent * Adapts to context * Guides toward resolution
⸻
What This Reveals
AI performs best when interactions are structured as sequences.
Also Demonstrates
AI tutoring · UX writing · structured communication
⸻
CONVERSATIONAL TONE SYSTEM
⸻
System Logic
Tone can be defined, tested, and controlled.
⸻
⸻
Method
* Define tone variables * Generate variations * Compare clarity + perception
* Refine for context
⸻
System Capabilities
* Controls voice * Adapts tone * Maintains consistency
⸻
What This Reveals
Tone is a system—not a stylistic choice.
Also Demonstrates
Copywriting · Brand voice · adaptive language
⸻
Generation Systems
AI CREATIVE WORKFLOW SYSTEM
⸻
System Logic
AI improves through iteration—not one-off prompts.
⸻
⸻
Method
* Define intent * Generate outputs * Evaluate * Refine * Select
⸻
System Capabilities
* Structures generation * Reduces iteration cost * Improves consistency
⸻
What This Reveals
Direction—not generation—determines output quality.
Also Demonstrates
Art direction · prompt design · iterative thinking
⸻
VARIATION & CONSTRAINT SYSTEM
⸻
System Logic
Constraints define outputs.
⸻
⸻
Method
* Fix variables * Introduce variation * Generate outputs
* Compare + measure * Refine constraints
⸻
System Capabilities
* Controls variability * Enables comparison * Improves consistency
⸻
What This Reveals
Consistency comes from control, not randomness.
Also Demonstrates
Copywriting · art direction · testing systems
⸻
GENERATIVE CONSISTENCY SYSTEM
⸻
System Logic
Outputs depend on input structure.
⸻
Consistent Outputs
Prompt Variation
Edge Conditions
Failure Systems
MODEL FAILURE SYSTEM
⸻
System Logic
Failures are driven by ambiguity.
⸻
⸻
Method
* Introduce ambiguous inputs * Generate outputs * Identify breakdowns * Analyze patterns
⸻
System Capabilities
* Identifies failure modes * Surfaces ambiguity
* Improves robustness
⸻
What This Reveals
Most failures come from unclear input—not weak models.
Also Demonstrates
AI tutoring · UX thinking · clarity design
⸻
SEGMENTATION FAILURE SYSTEM
⸻
System Logic
Perception breaks under complexity.
⸻
⸻
Method
* Introduce visual noise * Generate outputs * Compare failures
* Analyze conditions
⸻
System Capabilities
* Identifies breakdowns * Evaluates robustness
* Surfaces edge cases
⸻
What This Reveals
AI perception fails when context becomes complex.
Also Demonstrates
Visual analysis · AI evaluation · edge case thinking
⸻
⸻
Object Segmentation System - NEW paste HERE
A system exploring how tone functions as a controllable variable within AI-generated responses
⸻
System Logic
Tone is not subjective—it can be structured, tested, and applied systematically.
By holding informational content constant and varying tone, this system examines how shifts in language alter perception, emotional response, and user experience.
⸻
Process
• Consistent input across tone variations
• Structured shifts in tone (neutral, supportive, creative)
• Evaluation of clarity, perception, and interaction
⸻
Outcome
A scalable system for generating consistent, human-centered responses across different tones and conversational contexts
⸻
System Capabilities
• Adapts tone based on user intent
• Maintains clarity across variations
• Produces consistent, usable responses
• Scales across conversational use cases
Perception Systems
OBJECT SEGMENTATION SYSTEM
⸻
System Logic
Perception is context-dependent.
⸻
⸻
Method
* Provide inputs * Generate outputs * Compare accuracy
* Evaluate context
⸻
System Capabilities
* Detects boundaries * Evaluates context * Improves interpretation
⸻
What this reveals
Objects are understood through context—not isolation.
Also Demonstrates
Art direction · perception · evaluation
⸻
Exploration
COLOR IMPACT SYSTEM
⸻
System Logic
Small visual shifts change perception.
⸻
System Output
⸻
Method
* Keep composition * Vary color * Compare outputs
* Evaluate perception
⸻
System Capabilities
* Controls tone * Shapes perception * Enhances clarity
⸻
What This Reveals
Perception is highly sensitive to visual input.
Also Demonstrates
Art direction · brand · visual storytelling
⸻
IDENTITY FRAGMENTATION SYSTEM
⸻
System Logic
Consistency requires constraints.
⸻
.jpeg)
⸻
Method
* Generate variations * Compare identity * Identify drift
* Analyze breakdown
⸻
System Capabilities
* Tracks consistency * Identifies drift * Improves control
⸻
What This Reveals
Without constraints, identity breaks down.
Also Demonstrates
Art direction · brand systems · narrative
⸻
bottom of page