Sparkhood ジェネレーティブ・コンパニオンUI by Jizheng Guan

Sparkhood ジェネレーティブ・コンパニオンUI by Jizheng Guan is a highly esteemed winner of the very prestigious 2026 A' Design Award 鉄賞 in デジタルプロダクトデザイン Competition Category.

鉄デジタルプロダクト· 2026

Sparkhood ジェネレーティブ・コンパニオンUI

by Jizheng Guan

Client: Sparkhood

A' Design Award 鉄賞 Winner in デジタルプロダクト Category

5,597

Verified Award Winner

Verify Award Status for #182301

View at A' Design Award Press Release Award Badge

Colors

The Design

About Sparkhood ジェネレーティブ・コンパニオンUI

Sparkhoodは、ユーザーが意義ある目標へ最初の一歩を踏み出せるよう支援するAI駆動のモチベーションアプリです。ビジュアルストーリーテリングと行動心理学にインスパイアされ、野心をインタラクティブな体験へと変えます。導き手となる「未来の自分」と、リアルタイムに応答する適応型UIを特徴としています。複雑な課題をシンプルな行動へと分解することで、Sparkhoodは勢いを生み出し、行動を通じたアイデンティティ形成を支え、ユーザーが自分自身を探求し理解できるようにします。

The Recognition

About the A' Design Award 鉄賞

Sparkhood ジェネレーティブ・コンパニオンUIは、よく設計された創造物に期待される専門的・産業的要件を満たす優れたデザインに授与される、権威ある評価である A' Design Award 鉄賞の受賞作品です。鉄賞の受賞作品は、制作者の高い技量、高度な専門性、専門家としての理解、創造的な能力を示しています。Sparkhood ジェネレーティブ・コンパニオンUIは、高く評価され、称賛され、愛される、実用的で革新的な創造物であることが期待されます。この A' Design Award 鉄賞を受賞したデザインは、業界のベストプラクティスを統合し、ノウハウと才能に結びついた有能な技術的特性を備えています。Sparkhood ジェネレーティブ・コンパニオンUIは、その徹底ぶりで尊敬され、世界をより良い場所にするための生活の質の向上をもたらすよう設計されています。

Behind the Design

Press Kit

Credits, team members and technical details provided by the designer.

Unique Properties: Sparkhood is an AI based system that helps users move from hesitation to action. It turns goals into a lived experience. A future self shows what is possible. A resistance character reflects doubt and fear.
Tasks are broken into clear, small steps. Users focus on one step at a time, instead of being overwhelmed. The system is not only about motivation. It builds identity through action, and helps users see their own patterns over time.
Dimensions & Packaging: Mobile app designed for most smartphones.
Tags: Mobile app, goal management, agentic UI, motivation, companion
Production Technology: The system uses prompt design, agent based logic, and modular UI blocks. User input such as goals, emotions, and concerns are mapped into structured states. The app then generates UI, tasks, and interactions in real time.

Visual elements like scenes, avatars, and motion are rendered from a prompt library. Each session is task focused and adaptive, helping users stay present and engaged.
Design Challenge: A key challenge was designing a system that supports users without creating pressure or judgment. Representing problems as characters had to feel helpful, not childish or distracting. Another challenge was guiding users toward action without forcing behavior. On the technical side, real time generation of UI, characters, and interactions had to remain consistent and clear.
Research: This project explores how reframing difficulty changes user behavior. From over 100 one to one sessions, a key insight emerged. People often turn problems into self judgment, which blocks action. By turning problems into external characters, users can observe them instead of internalizing them. This shift reduces self attack and makes challenges easier to approach. Combined with AI generated UI and adaptive sessions, the system creates a loop of observation, action, and reflection.
Inspiration: After graduation, I spent time helping over 100 alumni students in one to one sessions. Many of them felt stuck. They were unsure what to do, afraid to try, or simply followed the crowd into jobs that did not fit them. In the AI era, entry level chances to try are shrinking, while tools and resources are everywhere. This gap creates more hesitation, not less.
Sparkhood is built to close that gap. It helps people start before they feel ready. The app breaks down complex goals into small actions, and builds a sense of companionship so users do not feel alone. Instead of pushing users to be perfect, it helps them try, observe, and learn. Even if they quit, they understand themselves better. Only the user can know what truly fits.
Project Duration: Sep 2025 - Feb 2026, New York, United States
Operation Flow: Users begin by entering a goal and current challenge. Users then enter a focused session. Within the session, guidance characters help break down the task and support problem solving.
When friction appears, the system generates a character that represents the problem. This allows users to see it as something external and workable.
The interface adapts in real time to the user’s state, keeping attention on the current step. After the session, progress is recorded, reinforcing a sense of growth and understanding.
Team Members: Jizheng Guan