Writing · 5 min read
The hardest thing I’ve designed is what’s next
Twenty years into my design career, I’m turning the process on myself: figuring out where I belong in a profession that’s changing faster than ever.
For most of my career, I’ve helped organizations make sense of complicated problems.
Now I have one of my own.
Where does twenty years of design experience belong in an industry AI is reshaping? I’m approaching it the only way I know how: as a design problem.
After thirteen years at the same consulting firm, I’m in an unexpected career transition.
I have more experience, better tools, and more ability to build than ever, yet the path forward feels less obvious.
A conversation that made things clearer
Jenny Wen’s conversation with Lenny Rachitsky (opens in a new tab) about how AI is changing design matched much of what I’ve been living. Most of us learned design as a relay:
- Research, then mockups. Every detail defined before anyone builds.
- Handoff, then waiting. The design goes over the wall to engineering.
What’s replacing it is shaping the product as it’s built: designing alongside engineering, in the working product.
Earlier in my career I worked across both design and development, the kind of generalist she hires for. Over time, my work narrowed to enterprise UX, product strategy, interaction design, and design systems.
Tools like Claude Code have let me reconnect them: explore an interaction, build a working version, test it, and improve it.
In some ways, the profession is moving toward the kind of cross-disciplinary work that originally attracted me to it. But I don’t think learning to build faster is the complete answer.
When making things becomes easier, choosing what matters becomes harder
AI is getting better at taste and judgment too.
Review 5 customers before archiving
Can’t archive
Monarch Dental Group: it has an open invoice. It stays in the table.
Active Check these. Clear a box to keep the customer.
1 customer churned, past due, or on trial.
- Halvorsen Freight Active
- Brightwater Clinics Active
- Alder & Finch Trial
- Monarch Dental Group Past due
- Saltmarsh Studio Active
Producing a solution isn’t the same as deciding whether that solution is appropriate. Someone still has to take responsibility for the decision.
That distinction runs through my whole career. In enterprise systems, a workflow can function perfectly and still confuse its users. Getting it right means deciding what should happen, and when a person needs to step in.
It’s also what motivated my Agent Review prototype. An agent builds a way to archive customers in bulk, and one press can archive five at once, three of them still active. Fast cleanup or no customer lost by mistake? No automated check can say which matters more, so the agent asks a person.
Agent Review is still an experiment, but it’s the problem I want to work on: keeping AI’s work understandable and accountable.
Turning the design process on myself
So rather than jumping into a conventional job search, I’m asking where my combination of experience can create the most value now.
I’ve tried different labels. Senior product designer. Enterprise UX specialist. Visual designer. Designer who builds. Each describes part of my experience; none captures all of it.
The more I look at my work, the clearer the theme: I enjoy turning complexity into experiences that people can understand, trust, and use. I care about visual craft, and also about how a system behaves and how it’s built.
I don’t want to chase the newest definition of an AI designer, or keep an outdated one because it’s familiar. So I’m working from three hypotheses:
AI in enterprise productswhere oversight, governance, accessibility, and complex workflows matter
Product, visual, and codeone senior designer moving between them
A senior product design roleon one team starting something new
Research is useful. Eventually you have to ship.
It’s tempting to keep rewriting the portfolio until everything feels aligned. I know, because I’m doing it.
But a career strategy, like a product strategy, can’t be validated entirely in isolation. I need to find out, from hiring managers and other designers, where my experience solves a real team’s problem.
The next phase has four steps:
- Define. A stronger position, not everything to everyone.
- Demonstrate. Working examples, like Agent Review, that show how I decide, not only what I make.
- Test. Start conversations and learn which problems teams need solved.
- Refine. Treat the portfolio and résumé as drafts, and change the story when the evidence says to.
What I know about the road ahead
I don’t know what happens to product design next. I want to keep working as a hands-on senior designer who brings research, visual craft, systems thinking, implementation, and judgment to complex products.
Most of all, where AI is becoming part of how people work, I want to make those experiences more coherent and trustworthy.
I’m still testing this direction, and I expect parts of it to change.
I don’t need to design the rest of my career before taking the next step. I need a clear hypothesis, enough conviction to act on it, and the willingness to learn from what happens next.
For the first time in a while, that feels less like uncertainty and more like a direction.