Workshop

Routesetting is a creative discipline with a logistics problem. A setter visualizes a sequence: the way a body should move through space. Then they spend significant time translating that creative intent into physical hold selection and placement. The gap between the vision and the execution is where a lot of the craft lives, but also where a lot of time gets lost.

SetCraft is an app I'm building to help with that translation. The core idea: AI-assisted generative design that takes a setter's creative intent and helps surface hold combinations and placements that match it. Not to replace the setter's eye, but to speed up the translation step so they can spend more time on the creative work.

It's early. But it's the most interesting thing I'm building.

Product designers have had online portfolios forever. It's just part of the job. Most of the rest of the professional world hasn't caught up, and that's especially true in AEC and product management. Your work lives in decks, in documents, in the memory of people who were in the room. That's not enough.

I built this site because I wanted a place that was mine. Not a LinkedIn profile, not a resume. Something that actually shows how I think and what I care about. Building it myself was part of the point.

As AI-assisted coding has matured, I decided to put it to use building something genuinely useful to me: a personal finance app I actually own. I'd been paying for a subscription service for years, but the more I used it, the more I recognized that there were features I wasn't using, and features from other applications that I wanted to use. Building it myself means I can shape it exactly to my needs, upgrade it when I want, and drop the monthly fee in the process.

I've been using Obsidian for years both personally and professionally. One of the things I was looking for was a way to make journaling easier and more accessible, so I could build the habit of writing more often. With the advancement of AI tools for creating applications, I thought it was a good time to build my own. It takes a timeline-based approach to journaling, making each entry a first class citizen. I'm still iterating, but what's great about AI-assisted coding is that I can build quickly without losing what's already working.

I've been spending serious time with Claude Code and Cowork: not as tools I use occasionally, but as a genuine area of study. I want to understand what it means to build with AI at the workflow level: how to set up memory systems, how to write skills that encode specific ways of thinking, how to use agents to automate tasks that used to require a lot of manual coordination.

Learning this well feels like the kind of leverage that compounds. The more fluent you get, the more you can build without needing to ask someone else to build it for you.

I came into product management from 18 years as an engineer. What I've found is that I love it: the intersection of vision, design, software, and creating something that actually delivers value to a real person. It's a different kind of problem solving than engineering, and I've fallen into it in a way I didn't expect.

Part of why I built this site was to have a place to show that work. Most people in this industry don't have a portfolio. I think that's a missed opportunity.