People keep asking me some version of the same question: “I want to do what you do: design and build real products with AI. Where do I start?” Usually it’s a woman early in her design or dev journey, and honestly, it makes my heart happy every time. So here’s everything I’d tell her over coffee. ✦
The role that’s emerging is what I call an AI-Native Design Engineer: someone who designs and builds, with AI as a copilot instead of a crutch. It’s the most valuable, most fun profile in tech right now, and the barrier to entry has never been lower. Here’s the map.
🎨 UX/UI fundamentals (what AI can’t replace)
AI can generate a hundred layouts in a minute, but taste and judgment are yours. This is the part that makes you irreplaceable, so invest here first:
- Visual hierarchy, typography & spacing: know why something feels right, not just how to copy it.
- Color theory & systems, covering tokens, scales, and contrast/accessibility.
- Real UX: user flows, information architecture, jobs-to-be-done, usability.
- Design systems thinking, meaning reusable components, not one-off screens.
- Figma: solid auto-layout, variables, and components. It’s still the bridge between design and code.
💻 Front-end (just enough to build & collaborate)
You don’t need to become a systems engineer; you need to speak the language well enough to direct the work and know when something’s off:
- Semantic HTML + modern CSS (flexbox, grid, responsive), the foundation of everything.
- JavaScript, then React and a framework like Next.js.
- Tailwind CSS: it’s fast and pairs beautifully with AI.
- Git & GitHub, covering versioning, branches, and deploys.
- Basics of deployment (Vercel/Netlify) and technical SEO/AEO.
🤖 The real superpower: working with AI
This is the new skill that changes everything:
- Learn to direct AI tools like Claude Code, not just ask, but guide.
- Prompt with intention: give context, reference images, and iterate in small steps.
- Develop the judgment to review what AI generates: spot what’s wrong and ask for the right fix.
- Stay curious about AEO/SEO for AI answer engines; it already matters.
✨ My most honest advice
- Build in public. One real portfolio teaches more than ten courses. (This whole site is my proof.)
- Ship > perfect. AI lowers the cost of trying so much that the advantage goes to whoever actually does.
- Your voice matters. The little details, the personality, the magic: that’s exactly what AI can’t invent, and what makes you memorable.
- Design + code. The hybrid design-engineer profile is the most valuable one right now.
The magic formula is simple: a designer’s taste + technical fluency + knowing how to direct AI. None of it happens overnight, but all of it is learnable, and the best time to start is today.
If you’re that girl reading this: you can absolutely do this. Start small, ship something, and let a little magic in. I’m cheering for you. 🔮