Framing Session
A structured session that turns a vague AI or product idea into a clear, buildable brief using Context → Goals → Constraints.
What's involved
Systems & UX judgement · AI build speed
Fast is easy. Right is the hard part.
You know AI should be helping your team, and can't quite make it stick. I deliver a fully working solution to a real problem you want solved, then teach your team to keep solving problems with AI on their own.
Twenty years of UX inside teams like
G-Star RAW
De Bijenkorf
Bestseller
Whirlpool
Hogeschool Utrecht
Why this is hard
Most AI pilots stall in one of two ways
You have budget and intent. What you don't have is a low-risk way to find out whether AI is worth building on, without hiring a consultancy that disappears or a trainer who never ships anything.
Failure mode 1
Shelfware
An agency builds something, hands over a repo, and leaves. Nobody on your team can change it, so it quietly dies. You paid six figures for a demo.
Failure mode 2
Theory only
A training vendor runs workshops on prompts and principles. Everyone is a little more literate. Still nothing real got built, and the momentum fades.
How I work
Fall in love with the problem, not the tool.
The teams that get real value from AI are the ones who understand their problem well enough to describe its edges. I build the first working solution that way, then teach your team the same shift in how they think.
Safety before speed
Anyone with a laptop can now generate something that looks like a product. I check what's safe to build, where data can't leak, and what people will actually use. That judgement is twenty years of ICT and UX, not a prompt.
Problems are constraints in disguise
This is the mindset shift I teach. Stop treating limits as blockers. Your security rules, your legacy systems, the things you don't know yet: those are the constraints that tell AI exactly what to build. Get the problem, its goals and its constraints clear, and the build follows fast.
Ownership transfer, on purpose
I teach the same framing method against your real product, so your team keeps solving problems with AI after I leave, instead of the tool sitting in a box.
Ways to work together
Start small. Build real. Transfer ownership.
Three tiers, from a low-commitment session to a full build-and-train engagement.
Build Sprint
I learn the real need and build a working product your team can actually use on the problem you defined. Not a prototype, not a mock-up.
What's involvedBuild & Embed
Everything in the Build Sprint, plus hands-on training so your team can keep extending the tool using the same method.
What's involved
Built with the framework
A few things I've shipped
Real products, built solo with AI. The point isn't a stopwatch. It's that a clear frame plus fast iteration removes the long discovery phase.
Learning
Techtok Missies
Kids learn to use technology, but not to question or shape it.
Read the case
Track record
Twenty years of knowing what to build
~20 years
Leading UX across industries like e-commerce and consumer products.
Hogeschool Utrecht
UX/UI coach, teaching the next generation of product designers.
Build, not just advise
Every project on this site was designed and built hands-on, not outsourced.
She blew the audience away with her expertise, passion and enthusiasm about her strategy to deliver an impeccable customer experience.
I could always count on her to bring fresh ideas and enthusiasm into anything from packaging design to new user interface concepts for connected appliances.
Got something you wish already existed?
Start with a Framing Session. Ninety minutes to turn it into a brief you could build from.