AYNA — Premium Bridal Atelier

Inspiration

Buying bridal wear is very different from buying a normal piece of clothing.

For a bride or groom, the outfit is expensive, emotional, and connected to one of the most important days of their life. Yet online bridal shopping still mostly works like a traditional catalog: customers see an outfit on a model and have to imagine how it might look on them.

In physical stores, trying on multiple heavy sarees, lehengas, sherwanis, or panjabis can also be time-consuming and exhausting.

That led us to one simple question:

What if customers could see themselves wearing a bridal outfit before physically trying it on?

That idea became AYNA, meaning “mirror” in Bangla.

AYNA is an AI-powered premium bridal commerce platform for Bangladesh where brides and grooms can discover outfits, virtually preview selected clothing on themselves, save their favorite looks, and continue toward a physical fitting or purchase.

What it does

AYNA combines a premium bridal e-commerce experience with AI-assisted fashion technology.

Customers can explore bridal collections for both women and men, including sarees, lehengas, gowns, salwar-kameez, sherwanis, panjabis, and other premium wedding outfits.

The core experience is the AI Virtual Try-On.

A customer can upload or capture a photo, select a supported outfit, and generate a visual preview of how that outfit may look on them.

The same concept can also be used inside a physical bridal store through the AYNA AI Mirror. A customer can stand in front of a screen and camera, take a photo, select different outfits from the store, compare AI-generated looks, save favorites, and then proceed to a real fitting or purchase.

AYNA does not replace physical fitting. Instead, it helps customers discover and shortlist the right outfits before spending time trying them on.

AI-Assisted Product Management

AYNA also uses AI to reduce work for store owners.

When an administrator uploads a new product image, the system can analyze the image and generate editable product information such as:

• Product title
• Category
• Target audience
• Color
• Visible embroidery or design work
• Style
• Occasion tags
• Product description
• SEO description
• Image alt text
• Searchable tags

Every AI-generated field can be reviewed and modified before saving.

AYNA deliberately does not allow AI to guess critical business information such as price, stock, SKU, exact sizes, or exact fabric composition.

This keeps AI useful while preventing inaccurate information from becoming trusted product data.

How we built it

AYNA is built as a modern full-stack web application using Next.js, React, and TypeScript.

For the interface, we use Tailwind CSS, shadcn/ui, Motion, Three.js, and React Three Fiber to create a premium, cinematic bridal shopping experience while keeping the website responsive and usable.

We use Supabase for PostgreSQL, authentication, user data, commerce data, and private customer media.

Cloudinary is used for product and catalog image management, optimization, and delivery.

The Gemini API is used for multimodal product-image understanding and AI-assisted product information generation.

The Virtual Try-On system is designed behind a separate AI provider layer so that dedicated try-on models can be integrated without tightly coupling the entire application to one AI provider.

Challenges we faced

One of the biggest challenges is virtual try-on accuracy.

Generating an attractive image is relatively easy, but preserving the customer's identity, body position, garment structure, colors, embroidery, and overall realism at the same time is much harder.

South Asian garments create additional complexity because sarees, dupattas, lehengas, and sherwanis often contain detailed draping, layering, and ornamentation.

Another challenge was deciding what AI should and should not infer.

For example, AI may recognize that an outfit appears maroon with golden embroidery, but it cannot reliably know its exact fabric composition, price, stock, or available sizes.

We therefore designed AI-generated catalog information as editable suggestions rather than verified facts.

We also had to balance premium visuals with performance. Luxury fashion websites benefit from rich imagery, animation, and 3D experiences, but excessive effects can make the website slow, especially on mobile devices.

For this reason, AYNA treats 3D and advanced animation as progressive enhancements rather than requirements for core shopping functionality.

What we learned

Building AYNA taught us that useful AI products are not created by simply adding a chatbot to an existing application.

AI becomes valuable when it removes friction from a real workflow.

For customers, AYNA helps answer:

“How might this outfit look on me?”

For store owners, it helps answer:

“How can I add and describe products faster?”

We also learned the importance of human review around generative AI, especially when AI-generated information becomes part of an e-commerce system.

Finally, we learned that AI features need reliable fallbacks. If AI analysis or generation fails, customers should still be able to browse products, and administrators should still be able to add products manually.

What makes AYNA different

AYNA is not simply an e-commerce website with AI added on top.

Our vision is:

Premium Bridal Commerce + Digital Atelier + AI Mirror + Physical Fitting

The customer journey becomes:

Discover → See Yourself → Compare → Save → Book Fitting → Customize → Buy

Our goal is to make premium bridal shopping in Bangladesh more personal, immersive, intelligent, and convenient.

What's next

We plan to expand AYNA with:

• Improved Virtual Try-On for South Asian garments
• Couple outfit recommendations
• Personalized bridal styling
• Made-to-measure workflows
• Private fitting appointments
• AI-powered product search
• Bangla and English support
• QR transfer from an in-store AI Mirror to a customer's phone
• Personalized recommendations based on saved looks
• Bangladesh payment gateway integration
• Store analytics and customer insights

Our long-term vision is to make AYNA a digital operating system for premium bridal shopping, connecting AI discovery, physical showrooms, personalized styling, and commerce into one experience.

Built With

• ai
• antigravity
• api
• cloudinary
• css
• fiber
• gemini
• google
• motion
• next.js
• node.js
• postgresql
• react
• shadcn/ui
• supabase
• tailwind
• three
• three.js
• typescript
• vercel
• vertex
• zod