Dev Blog

Live Design: When Building Becomes the New Designing

Written by Funnel Dev | Oct 28, 2025 8:28:54 AM

By: Elena Geroska, Product Designer in Funnel's AI team

A few weeks ago, I opened Cursor for the first time. Three hours later, I had designed and shipped a live feature without touching Figma. Such an aha moment. That’s when I realised the gap between design and implementation is disappearing.

Cursor opened a whole new world where ideas move directly from intent to product. It’s time for the era of Live Design, a way of shaping product experiences directly in code and in their real environment. Designing through implementation, not around it.

The End of Prototype Illusions

For years, design tools have kept creativity inside polished simulations. We design, hand off, and hope the final product resembles what we imagined.

That boundary between “the idea” and “the implementation” made sense when coding was slow and expensive. Today, with AI-assisted environments, that’s no longer true. Studies of AI coding assistants report productivity increases of 20-40% in tasks like documentation, completion, debugging, and repetitive code work.

If building the real thing is faster than mocking it, building high fidelity, polished, clickable Figma prototypes becomes a constraint. Instead of shipping representations, we can shape products directly in code, their native medium.

Designing in the Real Medium

After shipping my first interface through Cursor, the traditional workflow suddenly felt outdated. Cursor compresses design, feedback, and implementation into a single loop: think, build, test, and iterate, all in the product.

That loop is what shortens the distance between insight and impact. The point isn’t speed alone. It’s the ability to test real interactions earlier and ground design choices in live user feedback.

The result isn’t “designers coding.” It’s designers designing within the codebase, where every change immediately meets the realities of logic, layout, and performance.

The Real Skill: Communication, Not Code

What quickly becomes clear to anyone using Cursor is that the challenge isn’t syntax, it’s clarity. The better I articulated intent, the better the tool executed.

This reframes the designer’s core skill: precision of thought and language. The designer who can describe intent clearly to an AI, a developer, or a future self becomes exponentially more effective. Cursor simply exposes how well we communicate design intent.

Vague ideas rarely produce good outcomes. When I built the search-chats feature, screenshots weren’t enough. I had to spell out the components to use, when to trigger search, how to display results and empty states, the search depth, and the sorting rules. It forced me to think harder and articulate every detail of the experience.

Where Cursor Fits in a Designer’s Day

Here’s how to start live designing with Cursor.

  1. Micro-UX Improvements

    Cursor is ideal for small copy or layout adjustments the kinds of changes that normally wait weeks for engineering time. Even minor fixes can increase conversion by measurable margins

  2. Design System Contributions

    One of the biggest gaps between design and code lies in systems work. Cursor allows designers to update tokens, variables, and documentation directly, reducing friction between product and brand.

  3. Iterating on Live Flows

    Instead of guessing how a tweak will behave and if it will reduce friction, designers can implement iterations directly in production. This reduces the time needed for experimentation and learning.

  4. Designing Directly in the Code

    Designing with Cursor means moving ideas to product faster and with fewer resources. It understands your design system, components, tokens, grids and builds what you describe in plain language. No handoffs or specs, just working results in less time.

  5. Bridging Figma and Code

    Tools like Figma’s Model Context Protocol (MCP) now let designers move seamlessly between design and live environments. Cursor complements this, grounding creative decisions in code without demanding full developer proficiency.

  6. Talk like a developer

    Cursor’s Ask mode helps designers understand how code works, the logic and terminology. I use it daily to build fluency in technical language, prepare more effectively for meetings, and engage in deeper, more productive conversations with developers. The result is smoother collaboration and stronger ideas born from shared understanding.

 

So why aren’t we all moving to Cursor already?

Because it changes more than just our tools. It changes habits, boundaries, and identities. Designers need curiosity to step into code, and organisations need openness to let them. Without a strong design system or collaborative dev culture, the shift can feel intimidating.

At Funnel, we’ve always been forward-looking, constantly improving our tools, processes, and ways of working, moving towards becoming an AI-enhanced company. We have strong design foundations and a dev culture that embraces change.

Without strong design systems and transparent collaboration practices, this shift will expose weak foundations. But that’s the point. Cursor doesn’t just make design faster. It makes the invisible visible, the inconsistencies, dependencies, and communication gaps we’ve ignored.

Don’t get me wrong. It’s tricky and by no means replaces anyone. Every line of code I generate with Cursor is reviewed and refined by a developer. Sometimes the output is verbose or needs refactoring, but that’s part of the learning. Even then, it saves time by making design intent clearer and collaboration smoother.

AI as a creative amplifier

What excites me most about Cursor isn’t just what it can do, but what it represents, a shift in how we think about design. AI isn’t a threat to creativity. it’s an amplifier. Cursor handles the repetitive and mechanical, leaving more room for thinking, exploring, and creating. It redefines the design process and the designer role. Design isn’t a phase anymore. It’s alive. It’s not about producing hand-offs, it’s about producing outcomes.

The boundaries between disciplines are not being erased, but are getting flexible. Tools like Cursor encourage empathy between roles, curiosity across domains, and shared ownership of the product.

When we up-skill cross-functionally, the whole team levels up. Collaboration becomes smoother, faster, and more meaningful. We spend less time translating or waiting, and more time on what truly drives outcomes: being strategic, innovative and human-centric.

 

Written with a little help from my AI friends. They make it easier to say what I mean :)