For intermediate UX and product designers who want to use Claude and Figma's agent in a consistent way. You will learn to build repeatable design workflows and choose AI tools sensibly for your own process.
9 lessons · 83 min, plus 1 optional (27 min more with the optional lessons)
Beginner (not covered)
Intermediate
Advanced (not covered)
What you’ll learn
✓Match AI design tools to the real bottlenecks in your own process
✓Write specific prompts that get usable results from Claude and Figma's agent
✓Run an AI-assisted workflow from requirements through UI, design system and prototype
✓Compare AI design tools on the same task before choosing one to adopt
✓Keep creative decisions in your hands while delegating repetitive design work
Start here: frames where AI fits in a design process and how to pick tools by bottleneck.
Explains how UX designers can pick AI design tools by matching each tool to a specific bottleneck in their process, trialling one tool on a real flow, measuring the gain and reviewing output by hand. It also covers a per-job tool map, selection criteria and data and privacy rules.
Shows the whole team arc, from research and decisions to prototyping and iteration.
A walkthrough of how the Eleken team uses AI across the product design workflow, from research and product decisions to interactive prototyping and iteration. It stresses that designers still need to guide the process and validate AI-generated solutions before fast prototypes reach users and engineers.
Adds first hands-on Claude prompts for ideation, microcopy and design tokens.
Shows five prompts for using Claude across UI/UX work, from expanding sketches and ideation through microcopy, prototyping and testing, to building a design system. The prompts are aimed at designers who want to speed up their process and generate more ideas.
Introduces prompting Figma's agent directly, covering specific prompts, auto layout and context.
Shows best practices for chatting with the Figma agent to generate and refine designs, including being specific, using auto layout, focusing context, providing UI examples and organising chats. The description notes that output quality depends on how the feature is used.
Builds on the basics with front-loading context, using skills and managing chats efficiently.
Best practices for prompting Figma's agent to get more out of it while using AI credits efficiently. Covers giving the right context, using skills, when to create generative plugins, and when to ask Fig support questions instead.
Chains your prompting into a full workflow from requirements to prototype, with you as project manager.
A walkthrough of an AI-assisted UI/UX workflow in which the designer acts as project manager and AI assists with requirements, UI design, a design system and a prototype. It uses Claude, Figma, UX Pilot and Netlify.
Moves to an advanced chained workflow using Claude and UX Pilot via MCP into Figma.
The video shows a workflow that moves from diagnosing a UX failure from conversion data with Claude, through building a wireframe with UX Pilot via MCP and a persona audit, to styling two directions with DESIGN.md and exporting to Figma. It is aimed at designers who want to avoid switching tools at each handoff.
Compares Claude Design and Figma's agent on the same page, to judge fit after trying both.
A comparison of Claude Design and Figma's AI agent, in which the same web page is designed in both tools to judge which suits product designers better. The description is brief and offers no detail on the steps or findings.
Widens the view to five UI tools, helping you choose a sensible stack for your work.
The video tests five UI design tools used through Claude MCP: Figma, Pen.dev, Paper.design, UX Pilot and Claude Design. It compares their UI quality, features and cost, and ends with a recommended UX/UI workflow.
Optional deep dive into newer agent features such as skills, connectors and canvas generation.
A Figma Config 2026 session on the Figma agent's newer capabilities, including web search, live connectors, teachable skills and generation on the canvas. It covers what designers can hand off to agents and what to keep under their own control.
For intermediate designers who want to take a UI from layout grid to working code. You will plan interfaces, build interactive components, document them for developers and use AI to generate screens from a design system.
For designers comfortable with a little JavaScript who want to automate repeat design work. You will be able to write small Figma plugin scripts, build tools with AI coding assistants, and judge when a modular framework fits.
For designers who already prototype with AI and want their output to follow real design systems, component libraries and code. You will be able to extend existing designs, keep generated UI on brand, and move work toward handoff.
For product designers who want AI features to be trustworthy, clear and well judged. You will learn to plan AI behaviour, make answers and intent legible, and evaluate output against real user needs.