I used Claude to help analyze my Lyssna usability study from organizing responses to spotting patterns. It also caught a recruiting setting I’d missed. A useful reminder to check both the study setup and the AI’s analysis.

I tried Lyssna for an app I’m designing and found the setup straightforward. Its range of research methods gave me plenty to consider: what to ask, what to observe, and in what order.

A small experiment turning two mobile screens into an interactive prototype. The interaction worked, but unexpected visual changes showed me what I’d need to check before bringing it into a user test.

What I learned building a design token system from scratch, from three token layers to dark mode.

I’m building my first design system from scratch with Claude as my guide and learning how much thought goes into every little detail.

Nearly 3 in 4 organizations report measurable AI value, but only 13% have scaled it as planned. Curious where teams really are today.

A first look at OpenAI Dot: what changes when AI starts anticipating what you need instead of waiting for a prompt?

I tested whether an AI Skill could handle research judgment. It didn't make Claude smarter, just more careful.

The more I use AI, the more I'm drawn to analog design. I'm not the only one.

Figma Weave lets AI change only the layers you choose. Here's my first try.

Figma isn't just a design tool anymore. It wants to own the whole workflow. My take on Figma's plan to own the whole design workflow, which drew 44K+ views on LinkedIn.