AI Tools & Tactics in My Design Practice.
Over the past few years, I’ve gradually woven AI tools into my daily work as a designer. I can now be more creative and do more, sustainably. These tools help me generate ideas, communicate clearly, untangle complex information, and bring concepts to life in their most tangible form. I use AI mainly for content, imagery, and prototyping.
Tools like GPT have been especially helpful, not just for cleaning up writing or drafting documents, but also for organizing messy information, spotting feedback patterns, and quickly finding a clear starting point. I even built a Figma plugin using GPT and Claude to extract content strings.
Below, I’ve shared examples of how I use AI in different ways, depending on the task and goal. I’m always learning how to use these tools more effectively, since the technology keeps evolving. And while AI is powerful, it still requires human judgment and oversight.
Making sense of feedback and doing competitive analysis
ChatGPT, Claude, Notebook LLM
I often work with raw feedback like interview transcripts, help forum comments, and public reviews from sites like G2 and Reddit. Tools like GPT and Claude make this much easier. I use GPT to find themes, pull key quotes, and organize feedback into a clear format. Claude organizes content as “artifacts,” and GPT lets me create custom workspaces to keep context and explore the data further.
Simple design feedback
ChatGPT
Working as an IC, even at a big company, can feel isolating, designers need feedback, but it’s not always easy to get without regular critiques or team time. I use GPT to help fill that gap, especially on technical projects where I want to double-check my thinking or catch oversights.
While it’s not a replacement for real design feedback, I’ve been surprised by what it finds, like accessibility issues or clarity problems. It’s a helpful first pass, even if it can’t match a designer’s eye.
Turn information into something you can listen to and learn from
Notebook LLM (Google)
Notebook LLM is one of my go-to tools for working with dense or complex material, especially research transcripts. It’s great for analyzing batches of research papers or competitive data—helping me spot patterns, surface insights, and make the content more approachable. It’s also more reliable than tools like GPT for accuracy, as it cites sources and avoids hallucinations. I often use the “Audio overview” feature to create a podcast-style summary of the material, which makes it easy to absorb and understand. You can even ask questions during playback and get real-time answers from the AI hosts.
Turn technical docs into design briefs
ChatGPT
PM specs and engineering docs often come in too technical or too vague to use directly. I used to spend a lot of time reworking them, but now I use GPT to quickly rephrase and restructure them into design briefs. It still takes judgment and editing, but it saves time and gives me a clearer, more usable starting point.
I started using Midjourney in 2023, initially just to experiment. Realizing there was a learning curve, I took Nick St. Pierre’s Midjourney for Creatives course and have been exploring ever since, following other creators and being inspired by what’s possible.
Lately, I’ve been using AI tools like Vercel, Claude, and Figma Make to prototype and build working software. I used to rely on Framer, but these new tools have made it much easier to go from concept to execution. The gap between idea and reality is smaller than ever.
Messenger widget customization app
Vercel
At Zendesk, I built an internal tool for the Voice of the Customer team to easily customize the Messaging Widget. Personalizing the widget is important for showing prospective customers how the experience can look in context. Previously, this process was manual and slow, but the new tool made it much faster and simpler to create tailored demos.
Admin center dashboard
Figma Make
I used Figma Make to turn static design mocks into a working prototype. Once the first version was ready, I easily added new components and widgets. By prompting and iterating, I refined the design into a realistic, high-fidelity concept.
DJ Mixes app
Vercel
I’ve been using these tools to explore and build small apps for fun. One example is a DJ mixes app, inspired by YouTube videos of people DJing from unique spaces. I created it as a simple way to bring my favorite mixes, my personal bookmarks, into my living room.
3D image prompt library
Vercel
ChatGPT is now surprisingly good at creating images, and many creators are developing unique styles and prompts. I’ve started bookmarking my favorites and built a small gallery to collect them. Each entry has sample images, the original prompt, and often JSON instructions, so you can browse styles and easily copy prompts to try yourself.
Let’s talk
Feel free to get in touch anytime, I’m happy to help.