Aimee van de Schoor

Something I brought with me from design into product/engineering: Don’t start with markers. Start with pencils. At the art & design academy in the Hague, we were forced to sketch: explore multiple branches of the solution tree, throw ideas away fast (“kill your darlings”), and easily return to the last good version. I think AI can accelerate that same principle in product development: Agent-assisted sketching = learning velocity. Not spending months building “until it’s done,” but running fast iterations with a clear decision gate: Small experiments • Fixed evaluation moments (human-in-the-loop) • Short handover notes between iterations • Hard criteria for “better / worse” • Allowing actual customers to have an opinion The result: faster decisions, less sunk cost and only then investing in the enterprise-grade version. How do you organize “pencil mode” before something becomes “marker mode”?

11

2 commentaren

Excited to share that we have four impressive speakers sharing their knowledge and experiences with future foresight, curated by Phil Balagtas. Every week, during the month of May, there's something to learn about the practice of futures. A practice that is becoming ever more important for any organisation wanting to be prepared for the uncertainties to come. This series builds to a transformational Innovated by Design community for all invested in shaping the future.

7

Understanding clients is always one of the trickiest parts of design collaboration. In my 15 years as an industrial designer, working in design collaborations, creative consulting, and talent development, one common situation I keep encountering is clients or companies not being fully transparent when sharing information. Sometimes it’s about production capacity, available resources, objectives, timelines, or even budgets. And honestly, I always wonder why this happens. Is it fear of losing opportunities? Uncertainty? Protecting internal limitations? Or maybe just part of the negotiation culture? For me, honest information is not about limiting creativity, it actually helps designers develop better, more realistic, and more meaningful solutions. Curious to hear other perspectives on this. Have you experienced something similar?

10

Last night we had 40 people in a room talking about Framer, AI workflows, and how to actually make money as a freelancer, beginner or owner of a studio. Huge thanks to Andreas Kruszakin-Liboska and David Pukha for co-hosting 🤝🏻 My talk was about something I don't see discussed enough. How to combine Framer with AI tools like Claude Code and Codex, for example through MCP to actually improve your workflow. Not just generate components, but automate the repetitive parts that slow you down. A fresh angle most people in the room hadn't seen before. We also talked about how to make real money with Framer. Templates, freelancing, and leveraging your Framer (Pro) Expert status. Practical stuff students and freelancers can act on immediately. The no-code community in the Netherlands is growing fast. Designers, developers, students, founders, all in the same space, hungry for more. This won't be the last one. 🚀👀

29

3 commentaren

Today’s inspiration: when designing for disability, be aware of your biases, cultural assumptions, prejudices, and especially ableist narratives. Design is never neutral, even if we like to think it is. 📚 From: The Politics of Design, Ruben Pater (BIS Publishers, 2016) 👉 One of the most interesting books and research projects on how design directly impacts people, communities, and the environment. #accessibility #designforall #citizendesign #design #disability #universaldesign

53

6 commentaren

Design system management and maintenance is finally over. Like, truly over. A thing of the past. For years it was one of the most painful parts of designing products. Always important, always neglected. Always pulling focus away from what actually matters. Now we build once → and the design system documents itself. Automatically. Here’s how we do it now: In any project you’re working on (it has to be in an AI design tool obviously), whether you start from scratch or from a UI kit, just ask it to: “Create a full design system page with all components, styles, and states used so far.” (In a new page or URL) That’s it. You now have a fully generated, ever-updating design system that grows with the product. You can swap icon sets, update themes, add components with prompts. And just like that, the most boring thing in the design world is gone. Design systems finally adapt at the speed of the product, which means we can go back to what the canvas was meant for - exploring and imagining. #designsystems #productdesign #aiworkflow #uxui #superfine

73

25 commentaren

This upcoming Algorithm Audit's webinar offers a great deep dive into AI validation, addressing some of the most key challenges in the field of #AIBias and #AIValidation. Key topics include: > Methods to detect bias in AI systems, even when protected attributes are missing > The growing role and practical benefits of synthetic data generation > Introduction to two new open-source tools: an unsupervised bias detection tool and a synthetic data generator As #AI systems increasingly impact real-world decisions, understanding these validation techniques is essential to ensure fairness, accountability, and reliability. If you’re serious about advancing trustworthy AI, this is a great #webinar!

16

🎯 The fastest way to kill creativity? Only looking where everyone else is looking. I was working on a bicycle design project… and no matter how much inspiration I collected, everything felt too safe. Why? Because I was only looking at other bicycle designs. Sure, it worked,but the outcome looked generic. No spark. No “wow.” Then I made a small shift 👇 Instead of staying inside the niche, I pulled ideas from fashion, furniture, tech even a random coffee shop layout I loved. And that changed everything. ✨ The layouts felt fresh ✨ The design language got stronger ✨ The whole project had a new vibe The lesson I learned: 👉 For content ideas, niche inspiration works. 👉 For design innovation, look beyond your niche. That one change completely reshaped my design process. Curious where do you look for inspiration outside your niche?

15

1 commentaar

Design decisions now land in places design teams do not control. ➡️ The EU has put interaction patterns under regulatory scrutiny. Scroll, autoplay, notifications. Not content or data, but the patterns themselves. In most organisations I have worked in, nobody truly owns whether those patterns cause harm. ➡️ At the same time, AI-native products have collapsed the marketing page into the product. First-run experience now sits in a gap between teams, with no clear owner. At Google, the Gemini visual design team is carrying trust the product cannot yet guarantee. Load-bearing work the org chart still treats as a finishing layer. ➡️ Three different problems. One shared gap underneath. ➡️ Design choices are shaping compliance, acquisition, and user trust, but accountability has not followed the work. ➡️ Where is the biggest gap between what your org chart says design owns and what it actually carries? Read the full edition of Three Things in the link in the comments. ⬇️

9

1 commentaar

Last week I attended UX Rotterdam Conference.  It's been a while since I attended a conference, happy to participate in the first edition of this event. And UX Rotterdam nailed it. Two days, a lot of conversations, and ideas that actually stuck. The sharpest one: AI is accelerating delivery but it's made product discovery the bottleneck (from Julia Bastian). LLMs are good to summarise but not that good to synthesise. The product teams that move the needle from here won't be the fastest to ship. They'll be the ones who stay closest to the problem. The second day reframed this even further. You can't synthesise from an empty well. Eelco Van Collenburg put it simply: flood yourself with the raw. Break your patterns. Build a rich interior. The quality of your output is downstream of the quality of your input and most of that input has nothing to do with work. The other thing that landed, Alvin Chan gave a great idea: instead of performing, try to play it. Pressure is real, but performance optimises for evaluation. Play optimises for discovery. When the work is connected to something you genuinely care about, the pressure becomes fuel. Three things I'm taking back to the desk: - Run an AI audit: search your brand the way a user would, and see what the model actually says - Lead stakeholder conversations with a story before the framework - Ask honestly: where am I performing, and where am I playing? Be curious, seek depth, don’t accept the first answer. Thank you UX Rotterdam Conference for organizing it, and Freeletics and Confidence Udegbue for making it possible.

37

10 commentaren

If you can’t lead yourself through stress and change, you can’t lead a project or a team to success. In UX, we obsess over the user's journey. But the most important journey you'll ever manage is your own. For me, leadership isn't about a job, it’s about mental resilience and taking charge of your own growth and well-being. Ready to take the lead? I’ll be diving deeper into this during my active keynote at the upcoming UX Event Rotterdam. This won't be a passive talk, we’re going to get to work on your own mental resilience leadership path. Stop following the status quo. Start leading your own journey. Watch/ listen to the podcast episode with yours truly and Nauko Leong #mentalhealth #leadership https://lnkd.in/e_JEb7rJ

19

3 commentaren

Every new tool and technique seems to trigger the very same question: what still counts as “real” craft? In the past, painters like Jan Van Eyck, Peter Paul Rubens or Rembrandt made their own paint (or had their pupils do it). And some artists still do today. But when did it become acceptable to buy paint and still be considered an authentic artist? (According to ChatGPT: somewhere in the mid-19th century, when paint in a tube became available.) When photography was developed (pun intended), was that real art or just technique? Does a real writer write only with a pen? Or is a BIC allowed? An Olivetti 82 typewriter (Flemish people +55 will understand)? WordPerfect 5.1, perhaps? Or, brace yourself, Microsoft Word? Now programming. Is someone only a “real” developer if they write everything from scratch? Or are frameworks allowed? Libraries? Stack Overflow 🪦? And today: vibe coding with AI, such as Lovable? I see generative AI and Large Language Models as something similar. It’s a tool. Nothing more. Nothing less. With a lot of impact, yes, comparable to (or perhaps greater than) Gutenberg's printing press. Not all paintings made with store-bought paint are good art. Not all photographs are masterpieces. Not all software is well written or easy to use. Not all films with CGI are Oscar-winners. And honestly: most of us are simply building further on the shoulders of giants who paved part of the road before us. What’s changing today isn’t the tool, but what we expect from the maker. That’s where the idea of the Renaissance Developer resonates for me. This concept from Werner Vogels (CTO at Amazon) sees the developer (and by extension a Digital Service Design Engineer 🤓) not as a code typist, but as a modern polymath. For me, that's somebody who: • uses AI to accelerate repetitive work • understands systems rather than isolated components • takes ownership of the whole process • connects business, technology and users as context • and, above all, remains deeply curious As in the Renaissance: art, science, and engineering converging again. The difference is rarely in the tool. Almost always in the vision, judgment, and craftsmanship of the person using it. (Triggered by a LinkedIn post about the “Renaissance Developer” from Dr. Werner Vogels. Got some help from ChatGPT. The illustration was generated with Nano Banana, inspired by Leonardo da Vinci’s Vitruvian Man. The text in the illustration comes from the final page of John Fante’s Dreams from Bunker Hill, itself inspired by “The Walrus and the Carpenter” by Lewis Carroll.)

13

2 commentaren

End of October we hosted our first Product Leadership Roundtable at the Netlight Amsterdam office! We had engaging and intimate sessions with great energy and thoughtful discussions on the topic of Reimagining Product Organisations for AI-Powered UX. Let's share some key takeaways: - Faster path to MVP, kept human: AI can speed up exploration and reduce friction on the way to a meaningful MVP. The reminder was consistent: keep it human, ground decisions in user insights and do not use AI outputs alone. - Evolving roles and overlapping skills: as AI expands what different roles can do, we’re seeing more overlap across product, UX and engineering. Design engineers are a strong example: blending design intuition with implementation creates new collaboration patterns and shifts in ownership. Clear expectations and shared principles become even more important. - Product managers and AI for vision: PMs can use AI to quickly visualise concepts and product narratives, making vision tangible earlier. This helps teams and stakeholders align faster around “what good looks like”. - More time for interaction and alignment: with AI accelerating certain tasks, teams can invest more in the human work, facilitated conversations, decision-making and building shared understanding across disciplines and stakeholders. Thanks again to everyone who contributed and made it a genuine session. Thank you Amelie Duchardt & Eeva de Cazenove (Tervahartiala) for your insights and sharing your experiences with us, and Johanna Hermansson for making this event possible! If you’re interested in continuing the conversation or joining a future roundtable, feel free to reach out!

45

2 commentaren

Lately, I find myself talking about AI all the time. With colleagues, architects, developers. And that's probably true for most of us right now. Conversations about productivity, quality, responsibility or simply: "how are you using AI?" At some point I noticed I kept explaining the same things. So I wrote them down. Not to hype AI, but to share where I currently stand. What started as a few notes turned into a full reflection on how AI shows up in my daily work as an architect and engineer. It's a longer read, but worth it, I promise: https://lnkd.in/eYKSHxdN

20

2 commentaren

I recently read a book that included a quote by Jeroen de Flander: "You cannot be everything to everyone. If you decide to go north, you cannot go south at the same time." This is something that happens more often than not when I talk with founders. They have a great product that's working with a specific audience, and instead of going all in with something that works, they start adding a bunch of features to attract new audiences. This approach might work sometimes, but in most cases it's complete chaos. Your current loyal customers start to get confused about your product. You were a database, now you've turned into a full CRM, with a new SaaS platform, and a GPS tracker for pets. The key to growth isn't adding fancy things. It's focusing on what got you where you are, and scaling it. If a bakery is famous for its cheesecakes, adding new cakes won't make them grow. You're already doing great with a single thing, why add new products? There's a lot more you could do: ◾ Clarify your messaging so the right customers find you faster ◾ Build a website that actually converts your traffic into leads ◾ Create case studies that show the real impact you're making ◾ Position yourself as the obvious choice in your specific niche ◾ Optimise your sales process to close bigger deals Going deeper into what you already do well beats spreading thin across things you don't. #branding #designagency #webflow

17

So we’ve been talking about UX Laws for some time… now let’s talk about what Tobias van Schneider said. “If you want to be a better designer, don’t just study design books. Study sculpture. Study paintings. Study cars, watches, philosophers, movies, fiction, music, people. Study the world.” – Tobias van Schneider Design inspiration isn’t locked inside the design world. When we explore art, culture, history, science, and everyday life, we bring fresh perspectives to our work. A great UI might be influenced by the rhythm in a song, or the storytelling in a movie, or how people solve problems. The more experiences we draw from, the richer our designs become. Then most importantly experiment with everything you’ve learnt from these experiences. User experience is not just about having the knowledge, It’s about applying them in ways that make people’s lives easier and more enjoyable. ✌ ❤️ 😊 #UXDesign #DesignThinking #UserExperience #UIDesign #CreativeDesign

2

Working with Ron Kohavi and Lukas Vermeer 🃏on the Trustworthy A/B Patterns Project, I noticed a few interesting insights emerge on how power calculations are done on real projects by leading stat experts (as seen from a UI/UX designer's point of view). And so another side project was born - www.changeandmeasure.com - a little playground where I'm colliding: experimentation tools + heuristics + simulations. The first tool is a power calculator - that I can now work on, tweak, evolve and share freely. It has the following affordances and design properties based on what I've seen on our multiple projects over the last year (or two?): 1️⃣ Affordance to Explore Multiple Power Calculations We would often explore not one, but multiple power calculations in order to: achieve 80% power, maintain a realistic MDE (informed by historical priors), and aim to bring down the testing duration. That's no easy task when getting into 1 - 2% MDE territory, even with millions of visitors. Often it would involve falling back on and comparing proxy metrics (with higher base rates and often more power), as well as using other methods like capping, CUPED, and triggering. Interestingly, the MDE was the second most respected parameter, and rarely inflated due to lack of power. And so I built out the ability to compare multiple power calcs. 2️⃣ Heuristics as Smart Defaults Striving to keep power above 80%, as mentioned earlier, was one of the most important heuristics. But there were many other ones as well. This overlaps with the general UX principle of building in smart defaults into tools. As I come across heuristics from other experts in the field, I would like to use them as overlays that hopefully guide people as far as what values to enter. And so I entered a few heuristics from the people I work closely with on the project, but also from other respectable experts such as Georgi Georgiev and Ishan Goel and hopefully others in the future. Here I want to admit that this still needs more work. As: * some heuristics are in draft mode (not fully approved yet) * other heuristics will need conditionals * while others will open up to interesting differences (unresolved conditionals?) (Georgi for example is more open to one-sided tests, whereas Ronny leans towards two-tailed. Or Ronny often caps the MDE at 5%, whereas in some cases I may encourage teams to raise the MDE to 10% for high probability exploit / compound tests.) 3️⃣ Custom Traffic Volume / Custom Days I've seen Andrey Andreev once provide us with an X volume count for Y days (uneven) for a power calc. And so I added a traffic volume input per custom days - a tiny friction removal. 4️⃣ See the Power Loss from Fixing To Weekly Elsewhere, I occasionally notice teams set the test duration to X weeks (4-6), because they want a quicker test or have it as a cap. And so I built a feature that shows the loss of power as a result of doing so. More to come. Draft mode. Enjoy. #experimentation #design #stats

42

7 commentaren

Sunday Reflection: Goodbye car ownership? 🚗🚌 2.5 years ago, I gave up my company car to test what everyone calls the "sustainable future of mobility." Moving to the Netherlands, known for excellent public transport and shared mobility, I committed to 2 years without car ownership to experience the shift away from traditional ownership firsthand. 📊 The Reality Check: -> Lynk & Co subscription year: Started at €550/month, rose to €600, €649 now €679 (15,000km limit). For the (above) average person already juggling housing, insurance, and other subscriptions, this was simply unsustainable. I couldn't save money anymore. -> Greenwheels + Public Transport year: Lower initial costs, but the convenience factor and planning required made daily life more complex than anticipated. Especially mentally it limited spontaneous freedom. 🎯 Where I landed: I'm back to a second-hand car that provides both the financial benefit subscription models (with new cars) can't offer AND the freedom that sharing services couldn't provide. 🌱 After genuinely testing what's promoted as our mobility future, I've learned that subscription models often oversell their benefits. The Accenture report I read years ago positioned leasing, subscriptions, and vehicle-on-demand as key enablers for circular automotive economy - and while the sustainability goals are admirable, the current financial reality doesn't match the marketing promises. 💡 What I'd love to see: Subscription models for 5-10 year old cars entering the market as a much more accessible alternative. 🤔 Questions for the community: -> Are you regularly using shared mobility services? What's your experience? -> What new service offerings would help bridge the gap - either making it more financially accessible OR providing the timing flexibility that sharing services currently can't offer? #SustainableMobility #SharedMobility #CircularEconomy #Netherlands #SubscriptionEconomy #BusinessModel

22

14 commentaren