• Figma Make
  • AI
  • Prototyping
  • Supabase
  • Research

From Static Designs to Functional Prototypes with AI

How I helped shift prototyping at Stepstone from static simulations to functional, data-driven products, using Figma Make, AI models, and real data integrations to test real behaviour instead of imagined scenarios.

The Problem

Traditional design prototypes simulate interactions, but they don't behave like real products. During research, this creates a fundamental disconnect, users are asked to pretend, inputs don't behave realistically, data is static or fabricated, and insights are limited by the prototype itself.

As our ambition for research and product quality increased at Stepstone, traditional tools created hard constraints: no real data, no logic, no persistence, no adaptability.

We weren't testing real behaviour. We were testing how well users could imagine.

The Opportunity

Advancements in AI and tools like Figma Make introduced a new possibility, prototypes that could process real user input, connect to live datasets, adapt dynamically to user behaviour, and function similarly to production systems.

Prototypes that behave like products.

I focused on combining three elements to make this possible:

Real Data Integration

Using structured datasets, real job listings, real content, prototypes could reflect actual product scenarios rather than fabricated ones.

AI-Driven Logic

Natural language input could be interpreted and transformed into structured queries, enabling intelligent, adaptive behaviour without predefined filters.

Persistent State

User interactions stored and reused across sessions via Supabase, creating continuity that mirrors real product behaviour.

Adaptive Responses

Logic that responds to what the data returns, broadening or refining results dynamically based on volume and relevance.

What I Built

1. Conversational Job Discovery Prototype

Concept

Instead of traditional filters, users describe what they're looking for in natural language, and the system interprets intent rather than waiting for structured input.

How it works

A user inputs a query like "UX Designer in London for at least £50k". AI interprets job title, location, and salary expectations, maps them to structured fields, and returns results dynamically from real data.

Conversational UI with parsed query + results
  • No predefined filters, dynamic interpretation of intent
  • Handles synonyms and variations automatically
  • More natural interaction model for job discovery
  • Opened new directions for product thinking

2. Adaptive Result Logic

Problem

Search results can overwhelm or underdeliver, too many results with no guidance, or too few with no clear path forward.

Solution

Adaptive logic: if results < 50, broaden the search. If results > 50, refine it. A balanced experience that mimics intelligent system behaviour without manual filtering.

Flow diagram showing adaptive branching logic
  • Maintains relevance of results without user effort
  • Mimics real intelligent system behaviour
  • Reduces friction in the discovery journey

3. Research-Ready Prototypes

Problem

Research participants are often asked to engage with artificial scenarios, pretending data is real, imagining how a feature might work.

Solution

Prototypes designed to accept real user inputs, use real-world data, and respond dynamically, allowing participants to engage with scenarios relevant to their actual lives.

Prototype used in research session with real inputs
  • More authentic user feedback
  • Reduced bias from hypothetical scenarios
  • Improved quality and confidence in research insights

4. Persistent User State

Problem

Traditional prototypes reset after each interaction, no memory, no continuity, no way to test multi-step or returning-user journeys.

Solution

Using Supabase to store user sessions, inputs persisted across flows, prototypes could "remember" users and simulate real product behaviour across multiple interactions.

Diagram showing session storage + retrieval
  • Enabled testing of multi-step journeys
  • Simulated real returning-user behaviour
  • Improved testing of long-term and re-engagement flows

The Bigger Shift

This work redefined what a prototype could be, not just in terms of fidelity, but in terms of capability and purpose.

Static Dynamic
Scripted Intelligent
Simulated Functional
Disposable Reusable

Prototypes became environments for testing real behaviour, not imagined scenarios.

My Role

  • Led exploration of AI-powered prototyping at Stepstone
  • Designed system logic for conversational interfaces
  • Integrated real datasets into Figma Make prototypes
  • Defined interaction models for adaptive behaviour
  • Connected prototypes to persistent storage via Supabase
  • Supported research teams with new testing capabilities

Impact

  • Reduced time from idea to testable prototype from weeks to hours
  • Increased confidence in research insights
  • Enabled testing of complex, data-driven concepts
  • Expanded the role of design into product experimentation

What I'd Do Next

  • Integrate analytics and heatmaps directly into prototypes
  • Introduce automated research instrumentation
  • Expand AI reasoning capabilities for more complex flows
  • Enable reusable prototype frameworks for teams