Vestito

AI-powered fashion that fits you — personalized styles, tailored visuals, no code required

Comment

Vestito – Real Fit Fashion for Real Bodies

Inspiration

As a woman, I often find myself frustrated with fashion. Sizes like "Large" or "Medium" are supposed to fit everyone in that category — but in reality, no two bodies are the same.
Some women have hip dips, others carry weight in the belly, some have broader shoulders, or a curved back. Add to that modesty preferences, personal style, and overlooked sizes — and suddenly most clothing just doesn't work.

This sparked an idea:

What if customers could visualize and customize fashion pieces to suit their actual body shape and preferences — without needing to be designers?

What it does

Vestito is a no-code fashion customization platform where users can:

  • Browse branded items like dresses, basics, and tops
  • Submit natural language prompts such as:
    “Make this dress longer with full sleeves and a higher neckline”
  • Generate instant AI-based visual previews of the customized design
  • Pay securely using Stripe
  • View and manage past orders and customizations
  • Experience a unique shopping journey tailored to their body and preferences

How I built it

This app was built entirely with Bolt.new using:

  • Supabase – for user authentication, PostgreSQL database, and Row-Level Security
  • Stripe – for payments and checkout
  • Together AI – to generate visual product customizations from text prompts
  • Netlify – for continuous deployment
  • GitHub – for version control and syncing with Bolt
  • Supabase schema – for modeling users, products, carts, and orders

All features were created visually or via prompt engineering — with zero traditional coding.

Challenges I ran into

  • Integrating accurate AI generation was tough — many models (like Runway and Replicate) failed or were too expensive
  • Together AI finally offered high-quality, affordable generation
  • Bolt and Supabase RLS policy integration required extra effort to isolate user data securely
  • Voiceover tools like Descript weren’t delivering quality narration, so I explored better alternatives
  • Matching generated visuals with imagined designs was an ongoing refinement challenge

Accomplishments that I'm proud of

  • Fully functional and deployed e-commerce platform built entirely without code
  • Integrated real-time AI generation into a product visualization flow
  • Built an app that’s both inclusive and scalable
  • Seamless CI/CD setup with GitHub and Netlify
  • Used Stripe for monetization and completed multiple hackathon challenges

What we learned

  • AI image generation APIs differ significantly in quality, cost, and control
  • Supabase is powerful for real-world authentication and data isolation
  • Bolt.new can handle professional use cases when extended via external APIs
  • Non-coders can build real businesses when given the right tools
  • UI automation, deployment, and API calling can all be handled with prompts

What's next for Vestito

  • Add a virtual fitting room based on different body shapes
  • Let users upload a photo and preview styles on themselves
  • Expand product categories (hijab fashion, plus size, formalwear)
  • Onboard fashion designers to sell custom designs directly
  • Introduce user-style profiles to automate suggestions

Challenge Submissions

This project qualifies for:

  • Supabase Startup Challenge
  • Netlify Deploy Challenge
  • Stripe Monetization Challenge
  • AI Image Customization Challenge (via Together AI)

Built With

Share this project:

Updates