UXSnack

› tags human-ai interaction

#Human-AI Interaction

[07] pieces archived

How humans and AI systems communicate, collaborate, and influence each other. Mental models, conversational patterns, archetypes of relationship.

[01]
Design for AI 5 min

Observability in agentic UX

When the agent works in the background, trust dies. How Claude Code solved it, and what you can copy.

[02]
Design for AI 5 min

Capability discovery in agentic UX

People don't know what AI can do. It's one of the biggest UX challenges for AI products today, and there are concrete patterns that work.

[03]
Design for AI 6 min

Prompt engineering as design work

Two years ago this was an engineering job. Today, designers write prompts side by side with data science. Five patterns and a reverse-engineering trick that changes everything.

[04]
Design for AI 6 min

From deterministic to probabilistic design

For 60 years we designed for predictable screens. AI broke that. What changes in the method when the output is no longer fixed.

[05]
Design for AI 5 min

Mental models for AI design

Co-pilot, Assistant, Learning Partner: three ways people understand AI, and how each one changes the interface you'll design.

[06]
Design for Health 5 min

Grice's Maxims as conversational design heuristics

Four 1975 principles about what makes a conversation work. Applied to health chatbots, they're one of the best evaluation rubrics I know.

[07]
UX 4 min

5 Human-AI interaction types 🤖 The State of UX and Latest news on tech

Thinking about all the Human-AI interaction possibilities allows us to innovate and avoid falling back on the easiest solutions that we think are ‘shiny’ for users.