The Map
What do you want to learn about AI UX?
Pick the question that sounds like you. Each one opens onto the courses, patterns and checklists that answer it.
9 courses · 95 lessons · 38 patterns · all free, no account
Daily AI UX news and pattern breakdowns, straight to your inbox.
I'm designing my first AI feature. Where do I start?
The usual UX playbook assumes the system does what you told it to. Start with what changes when it does not, and the handful of patterns you will reach for in the first week.
Using AI UX Skills with Claude Code
A from-zero course on Claude Code and AI UX skills, 6 lessons and about 20 minutes: what Claude Code is if you have never opened a terminal for it, what a skill file is, how it differs from a pattern, how to install a skill pack, how triggering works without a slash command, and how to manage skills day to day.
Course · 6 lessons · 20 min
Conversational UI
Design intuitive, engaging, human-like interactions via chat and voice interfaces.
Pattern
Progressive Disclosure
Gradually reveal information, options, or AI features to reduce cognitive load and simplify complex tasks.
Pattern
Human-in-the-Loop
Balance automation with human oversight for critical decisions, ensuring AI augments human judgment.
Pattern
Error Recovery & Graceful Degradation
Fail gracefully with clear recovery paths when things go wrong.
Pattern
The AI UX Handbook
The patterns written up as chapters, if you would rather read straight through.
Kit
How do I make people trust what the AI tells them?
Most AI features fail on confidence, not capability. These cover showing uncertainty, explaining a decision, and leaving people a way back when it gets something wrong.
Confidence Visualization
Display AI certainty levels through visual indicators, helping users understand prediction reliability and decide when to trust or verify outputs.
Pattern
Explainable AI (XAI)
Make AI decisions understandable via visualizations, explanations, and transparent reasoning.
Pattern
Trust Calibration
Design a system that progressively builds appropriate trust through demonstrated competence - showing track records per domain, celebrating milestones, and adjusting oversight based on actual agent performance.
Pattern
Intent Preview
Before any significant action, the agent presents a clear, scannable summary of what it intends to do - showing planned steps, reversibility status, and edit controls for user approval.
Pattern
Action Audit Trail
Provide a timestamped, structured log of every action the agent took - grouped by task, with reversibility status, selective undo, and diff views - so users can review and correct agent behavior after the fact.
Pattern
How do I design the conversation itself?
Turn-taking, memory, and knowing when to hand back to a person. The course covers the shape of a conversation end to end; the patterns are the decisions inside it.
Build a Conversational UI
A practical guide to building conversational UIs: 11 lessons covering chat interfaces, voice assistants, streaming responses, conversation context, error recovery, agentic AI patterns, and accessibility. Includes code examples in React and design patterns used by ChatGPT, Claude, Slack, and Siri.
Course · 11 lessons · 39 min
Mixed-Initiative Control
Design interaction models where control flows seamlessly between human and agent - supporting parallel work zones, interruptible agent activity, and natural handoffs without formal 'take over' actions.
Pattern
Graceful Handoff
Seamless transitions between AI automation and human control.
Pattern
Selective Memory
Control what AI remembers, forgets, or ignores with transparent settings.
Pattern
Session Degradation Prevention
Strengthen safety checks during extended conversations with session limits.
Pattern
How do I get AI tools to build what I designed?
Five hands-on paths for the tools that turn a design into working code. Pick the one you already have open — they do not need to be taken in order.
Claude Code Course for Designers
Your complete Claude Code education: 24 sequential lessons covering setup, bidirectional Figma workflows, prototyping, version control, professional best practices, and customizing Claude Code with mods. Go from zero to confident in one comprehensive course.
Course · 24 lessons · 41 min
Claude Design Course
A practical 12-lesson path through Claude Design, from your first prompt to design system integration, team rollout, and production handoff. Sourced from Anthropic's official docs and hands-on accounts from designers who've spent real time in the tool.
Course · 12 lessons · 30 min
Cursor Course for Designers
Twelve lessons on Cursor 3 for designers: the Agents Window, the agent and its modes, Design Mode, building from Figma, and rules for your design system.
Course · 12 lessons · 20 min
GitHub Copilot Course for Designers
Your complete GitHub Copilot guide for designers: 10 lessons across 4 modules covering setup, code suggestions, prototyping workflows, and developer collaboration.
Course · 10 lessons · 18 min
GitHub Course for Designers
Your complete GitHub guide: 10 lessons covering account setup, repositories, version control basics, branching, pull requests, and design team workflows.
Course · 10 lessons · 18 min
How do I check my work before it ships?
Run the design past something other than your own judgement. Two checklists, an audit kit, and the patterns that catch what checklists miss.
Agent Readability Audit Kit
Check whether an AI agent can actually read and act on your interface.
Kit
Agentic UX Checklist
The pass to make before shipping anything that acts on a user’s behalf.
Checklist
Accessibility Checklist for AI Designs
Where AI interfaces commonly fail assistive technology, and what to do instead.
Checklist
AI Interaction Toolkit
Interaction patterns, component specs and implementation checklists as a PDF.
Kit
Vulnerable User Protection
Detect vulnerable users and apply graduated age, crisis, and dependency protections.
Pattern
Universal Access Patterns
Ensure equitable access for all abilities, languages, and assistive technologies.
Pattern
Why guides
The line between designer and developer is dissolving. Not because designers are learning to code in the traditional sense, but because AI has changed what “building software” means. You no longer need to memorize syntax, master APIs, or spend years on computer science fundamentals to participate in how your product actually gets built.
These guides teach you the tools that make that possible, whether you want cleaner handoffs with engineers, faster prototypes, a clearer mental model of the technical decisions shaping your product, or just the confidence to make changes yourself. Each course is a sequential learning path you can work through end-to-end, and every lesson has its own page so you can bookmark, share, and return to specific steps whenever you need them.
Frequently Asked Questions
Which AI tool should designers learn first?
Start with Claude Code. It has the most forgiving learning curve, works with Figma via MCP for design-to-code workflows, and lets you describe what you want in plain English instead of memorizing syntax. Cursor is great if you already know a bit of code. GitHub Copilot is best for people already using VS Code.
Do I need to know how to code to use these guides?
No. Every guide is written for designers with zero coding background. We cover everything from terminal basics to version control to shipping real prototypes. You describe what you want to build; the AI handles the implementation details.
Are these guides really free?
Yes, all guides are free. No email required to read them. The tools themselves (Claude Code, Cursor, GitHub Copilot) have free tiers, so we recommend starting with those and upgrading only when you hit limits.