How we prototype with AI tools: from a brief to an interactive prototype in days

What arrives in the first days

On every product we design, the first deliverable is the same: a high-fidelity interactive prototype. Real screens, real flows, real copy, running on a phone or in a browser, that you can click through, show to customers and put in front of investors. It arrives in days. That is not because we work faster than everyone else. It is because the tools changed, and we changed how we work with them.

The brief that goes in

A prototype is only as good as what it is asked to be. Before any screen exists, the discovery sessions produce a short written brief: the goal, who the users are, what already exists, and the flows the first version needs. The last part is the designer's decision, and it is the most important one on the project. A product has twelve possible flows and a first version needs three; choosing which three is judgment, not generation.

Research goes into that brief too. We synthesize the market, the competitors and the audience with Claude, and the designer reads the synthesis and decides what matters. That takes hours where it used to take days, and it means the brief is grounded before the first screen is drawn.

The tools, and what each is for

Claude Design takes the brief and the flows and produces the first version of the screens: layout, hierarchy, copy, states, as an interactive prototype. It is where the prototype starts.

Figma Make works inside the design file, generating variations of a screen or a component against the design system that already exists, so the options the designer chooses between are consistent with the product rather than generic.

Replit comes in when a prototype needs real behavior: data that persists, logic that branches, an integration that has to be seen working. Some validation questions cannot be answered by a click-through, and a working build answers them.

Figma remains where the interface is finished: every screen, the design system, brand identity when the product needs its own.

The tools are named because they are what we use today. They will change; the division of labor will not.

What the designer decides

Everything that matters. Which flows go in. The structure of each screen and the order of the steps. What the first version does when something goes wrong, which the tools reliably forget: empty states, errors, the second visit, the user who has no data yet. Consistency with the design system, which generators drift from within a few screens. And taste, which is the part nobody has automated.

Anyone who has pointed these tools at a regulated domain has seen the failure mode: they produce plausible screens quickly, and plausible is not the same as right. Every screen in a WeaveLines prototype has been reviewed and shaped by a designer before you see it. The tools produce; the designer decides.

What we validate with you

The prototype goes to you first, and the first walkthrough usually changes it. Then it goes where it earns its keep: in front of your customers and your investors, who react to a real interface rather than to a description of one. What they say shapes the next version, and it shapes the scope, because this is the point where changing your mind is cheap. A flow that does not survive the walkthrough is removed in an afternoon; a flow that does not survive after it has been built is a week.

From prototype to product

Once the prototype holds up, two things happen at the same time. Engineers start building from it, in the stack the product will grow in. And the designer moves to Figma to finish the interface: the design system of components and tokens that keeps the product consistent as it grows, brand identity if the product needs its own, accessibility checks along the way. Design runs alongside engineering rather than before it, which is how the whole thing, prototype included, fits inside an MVP that goes live in 2 to 6 weeks.

Why the first days matter

Decisions are cheapest before anything is built. A prototype in the first days means the scope is settled on something you can click, the investors' questions are answered with a product rather than a plan, and the engineers start from an interface that real people have already reacted to. That is the case for prototyping first, and it is why product design at WeaveLines starts there.

What to bring

You do not need wireframes or a brief. Bring what the product has to do for its first users, what already exists, and what you are unsure about. The first days turn that into something you can hold.