Home page
Dashboard
AI Stylist

Inspiration

Have you ever wondered what happens to the clothes that SME fashion boutiques fail to sell? Behind every local clothing brand is a founder struggling with overstocked inventory, fragmented data, and guesswork pricing. Without expensive analytical tools or dedicated styling teams, slow-moving items often just sit on racks, turning into waste. Inspired by these small business owners, we built CloThinks to tackle one of their biggest challenges: unsold inventory. CloThinks optimizes inventory in real time and transforms slow-moving products into sellable looks through intelligent outfit recommendations, helping retailers sell smarter, reduce waste, and maximize every piece.

Target Domain

AI-driven Inventory Intelligence for SME Fashion Retailers, focused on trend-aware inventory optimization, minimizing deadstock, and data-backed pricing strategies for the fashion and apparel sector.

What it does

CloThinks provides an interface in simple English to help SME Fashion Retailers improve inventory management and reduce waste by tracking stock in real time, identifying slow-moving items, and predicting demand. It also functions as an AI fashion stylist, generating outfit combinations by pairing slow-moving items with items customers are likely to buy, helping store owners turn unsold items into sellable looks. Hence, shopping experiences are improved by assisting customers with personalized style recommendations and accurate size matching.

Key Functionalities: Smart Inventory Intelligence | Conversational Interface | AI Outfit Matching Recommendation | Demand Prediction

How we built it

We built CloThinks by combining AI-powered demand prediction, inventory analytics, and an intelligent styling system into one platform. The system analyzes inventory data and product information to identify sales opportunities and generate outfit recommendations.

Frontend: Developed a sleek, user-friendly dashboard using HTML, CSS, and JavaScript for smooth interaction.
Backend: Built with Python, C++, and MySQL to handle real-time inventory data processing and API requests.
AI & LLM Integration: Leveraged OpenAI GPT-4o, Claude, and Gemini API for parsing natural language prompts and running conversational analytics.

Challenges we ran into

One of our biggest challenges was connecting multiple AI functions into one practical solution. We had to make sure inventory tracking, demand prediction, and AI styling worked together rather than functioning as separate features. We also had to balance AI accuracy, usability, and the needs of SME retailers.

Accomplishments that we're proud of

We are proud that we created more than just an inventory management tool. CloThinks can identify a problem and take action. Instead of simply telling retailers that an item is slow-moving, it suggests how that item can be styled and marketed to increase its chances of being sold.

What we learned

We learned that AI is most valuable when it solves a real business problem. A powerful model alone is not enough; the solution has to be practical, easy to use, and capable of creating measurable value for businesses.

What's next for CloThinks

Our next step is to make CloThinks smarter and more personalized by improving demand forecasting, learning from customer purchasing behavior, and integrating real-time sales and inventory data. Ultimately, we want CloThinks to become an AI-powered retail assistant that helps SMEs make better decisions from stockroom to checkout. We want to go beyond managing inventory and become the AI brain behind fashion retail. Our next step is to integrate real-time sales data, improve demand forecasting, personalize recommendations, use AI prompts, and eventually automate marketing to help retailers make smarter decisions from stockroom to checkout.

Built With

Submitted to

NTU InnovateX Hackathon 2026

Created by

I contributed to the ideation and development of CloThinks by researching the problem of unsold inventory. I helped to do slidedeck make me better understand the real-world problem.

Sharon Chan Wei Ting
I worked on ZAI with the team for the first time to build a simple user interface for CloThinks projects.

Angeline Teo
I contributed to the backend development for CloThinks. I implemented server‑side functionalities, managed database operations, and built APIs to connect frontend components with stored data, supporting the project’s core business logic.

Lye Jing Min
I worked on the front end of the project. My contribution involved designing and developing the user interface, ensuring that the website was visually appealing, user-friendly, and easy to navigate.

Qifang