Skip to content

P.06 — AI · Product · Design

AI Design Platform

An AI-driven UI/UX design platform used daily by CodeGen's internal design team — from requirements analysis to multi-page designs, prototypes and feedback.

Role
Associate Software Engineer · Primary author
Year
2025 — 26
Company
CodeGen International · Lumos
Special Recognition · National AI Awards 2026 · Best AI Solution
Designer working on interface layouts
Internal platform · screens under NDACodeGen International

01 — Overview

Overview

The design platform is where my engineering and design work meet. It's used daily by CodeGen's internal design team of about ten for requirements analysis, multi-page design generation, prototyping and collaborative feedback.

It has had positive feedback from the team, and potential clients have shown interest in it.

Started from zero as part of the Lumos ecosystem; I wrote most of the modules I own.

02 — The problem

Designers spend a lot of time turning requirements into first-pass screens before the real design thinking starts.

The platform had to speed that up without taking the designer out of control — generation is a starting point, not the final word.

03 — The system

From brief to prototype

AI does the heavy lifting between a brief and a reviewable prototype; designers stay in the loop at every step.

  1. 01

    Requirements

    Briefs are analysed into structured requirements.

  2. 02

    Generate

    Multi-page designs are generated from the requirements.

  3. 03

    Prototype

    Pages are linked into clickable prototypes.

  4. 04

    Feedback

    Collaborative review and feedback with the team.

04 — My contribution

Team, and what was mine.

Team

  • Lumos R&D team
  • CodeGen's internal design team of about 10 as daily users

My contribution

  • Started the platform from zero as primary author
  • Requirements analysis and multi-page generation flows
  • Prototyping and collaborative feedback features

05 — Technical depth

What it's built with, layer by layer.

AI
LLM-driven generation · Prompt & context engineering · Structured outputs
Product
UI/UX design · Prototyping · Design systems
Infrastructure
Microsoft Azure

07 — Outcome

Where it landed.

  • ~10

    Designers

    using it daily

  • 4

    Stages

    requirements to feedback

In daily use by CodeGen's design team, with positive feedback and interest from potential clients.