# AI builds your product in hours. I make sure it fits together. I'm a senior product design lead for startups and scaleups building with AI tools. I set the design system, review what AI generates, and decide what ships, so your design system lives in your code, not in a Figma file nobody opens. [Book a 30-minute product review](https://superflow.design/contact) ## Generating is easy. Orchestrating and judging is the bottleneck. - AI tools now produce screens, components and front-end code in minutes. - Nobody on the team knows which of it is actually good. - The result: five button styles, three spacing systems and a product that feels assembled, not designed. ## What I do ### Product Sparring Session One hour with a senior UX designer on your product. Bring a flow, a screen or a workflow question. You leave with concrete advice on what to change and why. **Who it’s for:** SaaS teams where UX decisions end up with AI tools or developers by default, and teams setting up an AI design-to-dev workflow who want a second opinion. **What you get:** - Advice on flows, patterns and interaction decisions in your product - A view on your design-to-dev workflow and where AI helps or hurts - Three written takeaways by email after the session - A clear next step, if there is one **Typical duration:** 60 minutes · free · one session per company · remote or in Ghent ### Design-to-Dev Setup I (help) set up the foundation your team needs to ship consistent UX without me: design system, UX patterns and an AI design-to-dev workflow. Then I hand it over. **Who it’s for:** Teams with developers and AI tools in place but no senior designer, who want to ship good UX on their own. **What you get:** - Two-tier token architecture (brand → semantic) and a component library with state logic - UX patterns and flow rules for recurring problems: forms, tables, navigation, empty and error states, among others - A Figma → Figma MCP → Claude Code pipeline (or any other) connected to your codebase - Automated product management and analytics, set up and maintained, so every release stays consistent and converts. - Handover: documentation, a team walkthrough and one check-in 4 weeks later **Typical duration:** Estimated budget on an agreed scope · typically 6 weeks to 12 months · ends with handover, unless you don't want it to of course. ### Embedded UX & Product Design Lead I join your team and own the UX of your product or module. I make the flow and pattern decisions, direct the AI output and keep the system evolving as the product grows. **Who it’s for:** Teams building a product that keeps changing, where UX decisions would otherwise end up with developers or AI tools. **What you get:** - Ownership of flows, patterns and interaction decisions - Senior review of AI-generated design and code, before it ships - A design system that grows with the product instead of drifting - Alignment between product, design and engineering - Guest experts from my network when needed, such as branding, sales strategies or marketing. **Typical duration:** monthly budget · on-site in Belgium or remote · Depends on availability ## Selected work - [NxtPeople: A design system for a multi-product HR SaaS](https://superflow.design/work/nxtpeople.md): One design system now supports multiple products and a separate brand. Product teams build new screens from shared components instead of designing from scratch. - [GoTemp: One design system, a second brand](https://superflow.design/work/gotemp.md): GoTemp launched as a distinct brand with its own light, recognisable look, separate from NxtPeople. Every screen is made from the same set of reusable shadcn components, and each one has written specs for its states and behaviour. Colours, type and spacing are set once in Figma as tokens and carried into the live product automatically, so one change updates every screen. - [Abilitee: Making complex VR training easier to create](https://superflow.design/work/abilitee.md): The concept and interface were refined through internal and client testing. The application was ultimately discontinued in production because creating and maintaining customised 3D environments required too much manual work. ## How I work - **Audit:** Review what AI and the team have already produced. - **Decide:** Keep, fix or throw away, with the reason written down. - **Systemise:** Turn the decisions into tokens, components and rules. - 20+ years - SaaS product experience - Tech start- & scaleups - Ghent, Belgium. ## Latest insights - [Design the delegation, not the screen.](https://superflow.design/insights/design-the-delegation-not-the-screen.md) (Imported 2026-10-20): Agentic SaaS turns product design into the design of trust. - [Is AI making designers obsolete?](https://superflow.design/insights/is-ai-making-designers-obsolete.md) (Imported 2026-10-20): Why a human designer still has a place at the table ## Your AI ships fast. Let's make sure it ships right. [Book a call](https://superflow.design/contact)