AI Practices
The ability to show, not just tell, is powerful.
AI has opened up new ways for designers, and really, anyone, to bring ideas to life. As someone who loves to build, it’s been a dream to take an idea from a half-formed thought to something you can actually see, test, and share.
Intention.
Any tool is only as useful as the intention and thought process behind it. I believe in staying in the driver’s seat: using AI to explore, shape, and interrogate ideas, but always directing the process.
Brainstorming.
AI is a good brainstorming partner, especially when working solo. It helps stress-test, break apart, and rethink ideas. I still believe the best ideas come from people, but AI speeds up the process.
Curiosity.
I’m naturally curious, so I love using AI, especially NotebookLM, to dig into research papers, YouTube videos, and other sources when I want to go deeper into a topic.
AI in my practice
Tools I use, things I’ve built
How I use AI
Process feedback and interview transcripts to surface themes.
Review designs for usability and accessibility.
Research with NotebookLM (targeted, source-backed).
Research with Perplexity (with validation).
Turn technical docs into actionable design briefs.
Prototype designs or a product vision.
Experiment with image generation in Midjourney.
Use Claude Cowork for brand and content pillars.
Things I’ve built
Custom Claude Skills for brand pillars, content strategy, and LinkedIn workflows.
A Time tracking app for my client work.
An automation that saves recipes from Instagram to Notion.
A daily curated feed for UI Goodies.
A widget customiser I built for Zendesk’s Voice of Customer team to brand the Messaging Web widget for prospect demos.
Made with Midjourney
