Building a design portfolio in the AI era
What stands out now that every portfolio looks polished.

Zander Whitehurst
9 min read
When anyone can generate a polished screen in minutes, a portfolio full of polished screens stops being impressive. Hiring managers we speak to say the same thing: the visuals all look good now, so they look for something else.
I have reviewed thousands of portfolios over the years, and I will let you in on a secret. I usually decide within the first thirty seconds whether I want to keep reading. Not because I am harsh, but because the good ones make it so obvious. They tell me a story. The forgettable ones show me a gallery.
That something else is evidence of how you think, how you work with others and how you get real things shipped. Here is how to show it.
What hiring managers are actually looking for
Most hiring managers spend a few minutes on a portfolio before deciding whether to talk to someone. In that time they are trying to answer three questions. Can this person solve real problems? Can they work with a team? Will their work be good enough to ship?
Beautiful mockups answer none of these on their own. They show that you can make something look nice, which AI can now do too. The parts of your portfolio that answer those three questions are what get you the interview.
Show decisions, not deliverables
The strongest case studies explain why, not just what. What problem were you solving? What options did you consider and reject? What changed after you tested it? That reasoning is the part AI cannot fake for you.
A good structure is to tell the story in four parts:
- The problem. What was wrong, for whom and how you knew.
- The constraints. Time, technology, business goals and anything else that shaped the work.
- The decisions. The key choices you made and the alternatives you rejected.
- The outcome. What happened after it shipped, with numbers if you have them.
Notice that the final designs only appear in part three. They support the story rather than being the story.
Include something that works
A live prototype or a shipped feature carries more weight than ever. If you can link to something people can actually use, do it. It signals that you understand how design survives contact with real constraints.
If your best work is under a confidentiality agreement, build something on your own. Recreate a flow you designed with fictional data, or design and build a small tool you wish existed. A working side project often says more about your skills than a heavily redacted client case study.
Be honest about your AI workflow
Do not hide that you use AI. Show how you use it well. Explain where it saved you time, where it got things wrong and how your judgement shaped the final result. Teams want designers who can get leverage from these tools without losing quality.
A short section in a case study that walks through your process is enough. For example: you used AI to synthesise interview notes, prototyped three directions in code in a day, then refined the strongest one by hand. That tells a hiring manager you are fast, thoughtful and in control of your tools.
Show collaboration
Product design is a team sport. Hiring managers want to see that you can work with engineers, PMs and stakeholders, not just produce work on your own.
Mention who you worked with and how. Describe a moment where feedback changed your direction, or where you had to negotiate a trade off with engineering. These details show maturity and make your work feel real.
Keep it short
Three strong case studies beat ten average ones. Lead with the one most relevant to the role you want, and make the first thirty seconds of each one count.
Put a short summary at the top of every case study: the problem, your role and the outcome in two or three sentences. Many people will only read that part, so make sure it stands on its own.
Cut anything that does not support the story. Process photos of sticky notes, every iteration of a screen and long lists of tools rarely add much. Show the moments that mattered.
Tailor it to the role
A portfolio for a design engineering role should look different from one for a strategy focused product design role. Lead with the work that matches what the team needs.
You do not need a completely different portfolio for every application. But reordering your case studies and adjusting your introduction for each role makes a noticeable difference.
The best portfolio I saw this year had one case study with a slightly scrappy prototype and an honest write up about what went wrong. I hired that person. Polish is everywhere now. Honesty and clear thinking are far rarer, and far more interesting.
Keep it current
Portfolios go stale quickly, especially now. Add new work as you do it rather than saving everything for a big update. Even a short write up of a recent project shows that you are actively learning and working with current tools. If you want fresh projects built that way, the AI Design Training Path takes you through the whole AI design workflow.
Your portfolio is not a gallery any more. It is evidence of how you think.
Memorisely