🤯 Agentic Design Systems are here - Learn the exact workflows
What if agents handled your Design System work?
Your Design System is drowning in manual work
Accessibility checks, design token updates, component reviews,
documentation that is constantly out of sync 😮💨
You spend 30% of your time just keeping the lights on 🥲
It is exhausting and it is getting worse as your design system grows.
What if agents handled that?
Agentic design systems are here!
AI agents that observe your design system, suggest fixes, update documentation, enforce standards.
These are production workflows from teams already using agents.
At our AI Conference on March 19-20, the designers behind them will show you their exact workflows.
Here’s what’s coming:
📌 Building Real Design Systems with Agents
Jan Six, Principal Product Designer, GitHub
Jan designs agent experiences for GitHub Copilot and created Tokens Studio, the tool that brought design tokens to Figma. He’s been working with agents before most people knew what they were.
In this session, he shares how to build design systems that agents can actually operate on.
What you will learn:
How to make your design system agent-ready
Designing instructions so agents collaborate, not hallucinate
How agents change the role of designers and design engineers
📌 Agentic Design Systems
Romina Kavcic, Design System Lead, The Design System Guide
Most design system teams spend 30-40% of their time just keeping things working. Accessibility issues, Design Tokens misuse. Documentation divergence.
Romina shows you a different model: AI-powered systems where agents automatically observe, detect, suggest, fix, and learn.
What you will learn:
The five agentic flows: Observe, Detect, Suggest, Fix, Learn
How to encode intent, not just values
Implementation strategies for any organization size
📌 Machine-Readable Design Systems for MCP and LLMs
Diana Wolosin, Sr. Design System Designer, Indeed
“Design systems today are built for humans, not machines.”
Diana has 10 years of experience building systems that scale.
She ran an 8-configuration MCP benchmark to find out what actually works when AI reads your design system.
What you will learn:
How to design metadata architecture for MCPs and LLMs
What makes “quality input” for AI reasoning
Using structured constraints instead of fragile prompts
📌 Encoding Governance on Agentic Design Systems
Cristian Morales Achiardi, Design Engineer, Enara Health
Documentation drifts. Design Tokens diverge from code. The “source of truth” depends on who you ask.
Cristian’s fix: make code the source of truth. AI agents that don’t just consume components, but actively enforce your design contracts. On-demand audits, generated dashboards, and documentation sites built from the system itself.
What you will learn:
The three structural layers: tokenization, indexing, and strategies
How to make components machine-readable for AI agents
Automated governance: audit reports and dashboards on demand
What’s included:
🟢 Live sessions, demos, case studies & live Q&A
📁 Slides, templates & files from speakers
📼 All recordings
🏆 Certificate of attendance
⚠️ Only 53 of 1,000 tickets left
Stop maintaining!
Start automating!
This shift is happening now
The question is whether you lead it or chase it
Your chance to learn from teams already using agents.
See you there!
Sil
