LOOK AI – Decision Agent for Shopping

An AI agent that helps people decide what to wear and what to buy using their existing wardrobe.

Comment

Inspiration

Shopping today is no longer limited by a lack of choices—it is limited by decision fatigue.

Existing AI shopping tools are good at recommending products, but they rarely help users answer practical questions like: "Should I buy this?", "Do I already own something similar?", or "What piece is actually missing from my wardrobe?"

As both a fashion model and an independent product builder, I experienced this problem firsthand. That inspired me to build LOOK AI—not another styling app, but an AI shopping decision agent that helps users make better decisions with what they already own.

What it does

LOOK AI is an AI-powered shopping decision agent.

Users upload one clothing item, and the AI analyzes it, generates outfit suggestions, identifies missing wardrobe pieces, finds visually similar products, and helps users decide what to wear and what to buy.

Instead of generating endless recommendations, LOOK AI focuses on reducing decision fatigue and making shopping more intentional.

How we built it

LOOK AI is built as a WeChat Mini Program MVP.

The project combines multimodal image understanding, prompt engineering, AI workflow orchestration, cloud database services, and recommendation logic.

The current workflow is:

Upload → AI Analysis → Outfit Generation → Similar Item Search → Wardrobe Management

The architecture is modular, allowing different vision models or recommendation systems to be integrated in the future.

Challenges we ran into

The biggest challenges were not only technical but also product-related.

We faced issues including fashion image copyright, product data availability, similar-item retrieval, AI response latency, and designing an effective recommendation workflow.

The most important challenge was defining the product itself. During development, the project evolved from an AI styling app into a shopping decision agent focused on helping users make confident decisions.

Accomplishments that we're proud of

• Built the entire MVP independently. • Designed a complete AI shopping workflow from scratch. • Introduced the concept of a decision layer for AI shopping. • Successfully combined fashion expertise with AI product design.

What we learned

Building AI products is not only about choosing better models.

We learned that product design and user decision-making are equally important. Users don't necessarily need more recommendations—they need more confidence in the decisions they make.

What's next for LOOK AI – Decision Agent for Shopping

Our next goal is to develop LOOK AI into a complete AI shopping agent.

Future features include personalized wardrobe memory, long-term preference learning, smarter shopping planning, and integration with real e-commerce platforms to create a seamless shopping experience.

Built With

Share this project:

Updates