I've spent four years building AI into real products. Here's what I actually do.

Most introductions start with a job title. Mine won't, because a title doesn't tell you much about what I actually spend my time on.

My name is David Skelton. I'm a product designer based in Columbus, Ohio — but for the last four years, most of my work has lived at the intersection of design, AI systems, and getting things to actually ship.

Here's what that looks like in practice.

At Arloa, I was part of a four-person team building an AI platform that helps parents navigate special education. We didn't have the headcount to keep up with content, so I built an automated engine using Claude Code — it pulls in feeds, writes editorial summaries, and sends out emails. Publishing went from 4 pieces a month to 12+, no new hires.

Before that, at Mantium, I was designing one of the first enterprise GenAI platforms. The challenge was getting non-technical teams from "just signed up" to "actually using AI" without a 2-hour onboarding call. I redesigned the flows so that first deployment happened in under 30 minutes.

At Huntington Bank now, I've been building a Figma-to-HTML workflow that uses AI to extract design context and spit out production-ready code. In a place where most teams are still printing PDFs and routing them through email chains, we're shipping in hours instead of days.

I'm not a researcher. I'm not a prompt library. I'm the person sitting between the AI tools and the business outcome, figuring out what actually works.

That's what this blog is about.

Every post here comes from real work — use cases I've shipped, experiments that failed usefully, techniques I've refined across startups and enterprise. Some of it will be tactical. Some of it will be strategic. All of it will be tested.

If you're a designer, a product leader, or a founder trying to figure out how to actually integrate AI into your operations — this is written for you. And if you want to skip straight to the part where we build something together, you know where to find me.

Let's get into it.