AURA | Devpost

Dashboard : Outfit of the day
My Wardrobe
AI Theme based Stylist
Rate my outfit
Personal Stylist (Chatbot)

Inspiration

Choosing outfits daily is surprisingly time-consuming, especially when people already own good clothes but don’t know how to combine them well. Most fashion apps either promote shopping or give generic suggestions without considering a user’s actual wardrobe, occasion, or theme. AURA was inspired by the idea of turning a personal wardrobe into a smart, AI-powered stylist that helps users dress better with what they already have, while also suggesting matching products when needed.

What it does

AURA is an AI-powered personal styling platform that:

Generates outfits from a user’s wardrobe based on themes Provides an Outfit of the Day for daily styling based on the weather Includes a “Rate My Outfit” feature with AI feedback which provides pros and cons of out outfit Offers a conversational AI fashion chatbot through which we can query on our styling options Suggests matching products when wardrobe items are missing from the e commercial websites

How we built it

Frontend: React.js for a responsive and interactive user interface. Backend: FastAPI for handling APIs and business logic. Database: SQLite for lightweight and fast local data storage (wardrobe items, users, outfits). AI Integration: Gemini API for intelligent outfit reasoning, matching colors, styles, and themes.

Architecture:

Frontend sends user inputs (theme + wardrobe) Backend processes requests AI model generates outfit logic Results are stored and displayed back to the user

Challenges we ran into

Designing meaningful outfit logic instead of random clothing combinations. Structuring wardrobe data properly in SQLite for fast retrieval. Ensuring AI suggestions match real-world fashion sense. Connecting AI output cleanly with frontend components. Planning for scalability while initially using a lightweight database.

Accomplishments that we're proud of

Built an AI stylist capable of evaluating real outfits Implemented a daily personalized outfit system Designed a conversational chatbot for fashion assistance Combined vision AI and text-based AI in one platform

What we learned

How to integrate AI models into real-world applications. Backend–frontend communication using FastAPI and React. Practical database design using SQLite. Turning an everyday problem into a user-focused AI solution. Importance of clean architecture for future scalability.

What's next for AURA

Replace SQLite with Supabase / PostgreSQL for cloud scalability. Add virtual try-on using AI-generated models Implement user style profiles and fashion history. Deploy the platform publicly with authentication and personalization.