Workshop design

Designing in the AI era

A UX workshop for non-designers

Half day · hands-on · no design background required

By Dusty Baars

The promise in one sentence

Building has become cheap. Judgement is expensive. In half a day, you will learn the two things that matter now: looking like a designer, and building with AI. By the end, you will have built a working prototype of a real problem from your own work with your own hands.

For a generation, building was the hard part. That is over.

You used to need a team, weeks of time, and technical skill to turn an idea into something tangible. Today, anyone without a design background can build a working prototype in an afternoon, simply by describing what they want. Building has become cheap.

This shifts the scarcity. The question is no longer whether you can build it, but whether you know what is worth building, and for whom. That is exactly what good design has always been: understanding the human, seeing the friction, building the right thing. Not the prettiest button, but the shortest distance between what someone wants and what they get.

AI does not replace that thinking. It speeds up the hands, not the compass. Whoever has only the accelerator builds the wrong thing faster. Whoever has only the compass keeps talking without building anything. This workshop gives you both, and lets you use them together in the same afternoon.

The core idea. The compass points (the human, the friction, the value). The accelerator builds (AI). A modern designer masters both, and knows when to lean on which.

Who it is for

For people who are not designers but do design, or will do so soon. Entrepreneurs, marketers, product owners, decision-makers. Anyone who has ideas and can now build something tangible themselves using AI, but lacks the tools to judge whether it is actually good.

What you can do by the end

See

Turn a vague problem into a sharp, human problem statement.

Choose

Think broadly and then choose with focus, without letting the loudest voice win.

Build

Go from words to a working, clickable prototype with AI, even without code.

Judge

Look critically at what AI builds and remove the friction before a user encounters it.

No theory for the sake of theory. Every participant works all afternoon on a real problem from their own work, and goes home with something that runs.

Half day, about four hours including breaks. Four blocks, each following the same rhythm: explain briefly, do it yourself, apply AI.

0:00

Opening — the shift

Welcome and context. Icebreaker: the last digital thing that annoyed you. UX becomes tangible immediately. The core shift: building became cheap, judgement became expensive. (20 min)

0:20

Block 1 — Seeing what the user sees

Empathie and the real problem. Mini-lesson, own problem statement, quick empathy map, AI as a sparring partner. (45 min)

1:05

Break

(10 min)

1:15

Block 2 — From problem to idea

First broad, then narrow. Crazy 8s, AI as a brainstorm multiplier, choosing with a value-vs-effort matrix. (45 min)

2:00

Break

(10 min)

2:10

Block 3 — From idea to tangible · the heart

Vibe coding: from description to a working prototype with AI. Everyone builds. (60 min)

3:10

Block 4 — Testing and critical review

Output is free, judgement is scarce. Heuristics, and a short user test of your own prototype. (30 min)

3:40

Closing

What you take home. The compass and the accelerator as a lasting lens. One thing you will do differently on Monday. (20 min)

For a tight 3.5-hour version, shorten Block 3 to 45 minutes and keep breaks at five minutes. The four blocks remain, because the strength lies in the complete journey from problem to tested prototype.

Opening — the shift

20 min

You do not open with a definition of UX, but with an experience. Everyone names the last digital moment that annoyed them. Within two minutes, the entire group is in the middle of the topic, because frustration is friction, and friction is exactly what design is about. Then the context in a few sentences: building became cheap, judgement became expensive. Today we practice both.

Block 1 — Seeing what the user sees

45 min
Mini-lesson

UX is the friction between what someone wants and what they get. People hire a product to get a job done, they do not care about your product, they care about their job. (10 min)

Do it yourself

Choose a real problem from your own work. Write a sharp problem statement. Create a quick empathy map: what the user says, thinks, does, and feels. (25 min)

AI leverage

Let AI play a synthetic user to test your assumptions, and summarize raw interview notes in minutes. With a warning: synthetic is for early hypotheses, not for final validation. The real human remains the source of truth. (10 min)

Block 2 — From problem to idea

45 min
Mini-lesson

First diverge, then converge. The alone-together principle: first think quietly by yourself, then share, so the loudest voice does not hijack the room. (10 min)

Do it yourself

Crazy 8s: eight ideas in eight minutes. Then AI as a multiplier to broaden your directions. Converge with a simple value-vs-effort matrix, and choose one. (25 min)

AI leverage

Let AI formulate and challenge your chosen direction: where does this idea break, for whom does it not work. (10 min)

Block 3 — From idea to tangible · the heart

60 min
Mini-lesson

Vibe coding. From words to a working prototype, without programming. What the tools can do now, and when to grab which: Figma Make for fast design-in-context, Lovable for a complete working app, v0 for separate components. (10 min)

Do it yourself

Build your chosen idea. Everyone, even those who have never written a line of code, goes from description to a clickable prototype. This is the moment it clicks. (45 min)

Reflection

What was easy, and where was your judgement still required. The prototype is not the end, it is material to test. (5 min)

Block 4 — Testing and critical review

30 min
Mini-lesson

Output has become free, judgement has not. A handful of heuristics to look critically: is it clear what is happening, do you recognize rather than recall, can you make errors without damage. Plus the human scale: AI invents with confidence, you verify. (10 min)

Do it yourself

Let someone else use your prototype without explanation. Watch where they stumble. One round of feedback, sharp and concrete. That is worth more than an hour of talking about it. (20 min)

Closing

20 min

No top-down summary, but one from the group itself. What everyone takes home. The compass and the accelerator remain as the two handles for everything that follows. Each names one thing they will do differently on Monday. Small, concrete, applicable today.

The space

  • Tech. A laptop per person and stable Wi-Fi, because we will actually build.
  • Space. A large screen for demonstrations, and sticky notes plus markers, or a digital FigJam or Miro board.
  • Time. A visible timer. Timeboxing is the silent hero of a good workshop.

Setup beforehand

  • Accounts. On the AI building tools, the free tier is sufficient (Lovable and Figma Make).
  • Fuel. Everyone brings one real problem from their own work. No fictional cases, that is half the success.
  • Rehearsal. A quick test of the tools the night before, so you do not get stuck on a login during Block 3.

Content is half the work. The other half is how you lead the room, especially with people who are not designers and quickly feel uncertain about creative work.

The title sets the tone before you begin. Three directions, choose what fits the audience.

Subtle

Designing in the AI era

Calm, professional, fits any schedule.

Middle

The compass and the accelerator

Intrigued, already carries the core idea.

Bold

Anyone can build now. What will you make?

Direct, slightly provocative, immediately wakes up a room.

Research and building are democratized. Non-researchers conduct research, non-designers build prototypes. This unlocks massive speed, and at the same time a risk: more people build things faster without the judgement to see if they are good.

This workshop addresses exactly that. It gives a broad team the thinking to use new tools responsibly: building faster, but building the right thing. This is not design training for designers, it is a way to make an entire organization build with sharper focus, in a time when building stops no one.

What it also shows

This design unites two worlds that rarely sit in one head: the human side of design and the build side of modern AI tools. Whoever can run this workshop understands both, and knows how they work together.

The design draws from the state of the art in 2026: the rise of AI prototyping and vibe coding, the shift in UX research toward AI synthesis and synthetic users, and proven facilitation principles for design thinking.