AI custom merch marketplace
MerchMind
A multi-vendor merch marketplace where shoppers describe a design in words, generate artwork with AI and preview it on products.
- Role
- Co-developer (final-year project with a classmate)
- Platforms
- Web, Backend & API
- Status
- Final-year project (frontend demo live)
- Timeline
- 2024 – 2025
- Web application
- E-commerce
Overview
MerchMind is a MERN-stack marketplace for custom printing and merchandise. A shopper types a short prompt, the platform generates an artwork with an AI image API, and the design is placed onto eight product mockups that can be downloaded or added to the cart.
Around that design flow sits a full multi-vendor store with three roles: customers who browse, buy and track orders, sellers who run their own shops, and an admin who oversees users, sellers, orders and payouts. Checkout supports Stripe and PayPal, and buyers can message sellers in real time.
I co-developed MerchMind with a classmate as our final-year project. The marketplace layer builds on an existing multi-vendor MERN e-shop codebase that we adapted and rebranded. Our main original work was the AI design studio, the mockup pipeline and the MerchMind branding, and I made the deployment commits for the public frontend and backend repositories.
Built for: Shoppers who want custom-printed merchandise without hiring a designer, independent print and merch sellers, and the marketplace operator.
The problem
Ordering custom merch usually starts with a finished design. People without design skills have to find a designer, wait for drafts and then guess how the artwork will look on a hoodie or a mug before they order.
Print sellers face the opposite problem: they need a storefront, payments, order handling and a way to talk to customers, which is a lot to build for a small shop.
The solution
We built a design studio inside the store. The shopper describes an idea in plain words, the prompt is wrapped in a quality template and sent to an image generation API, and the result is rendered onto eight product templates at once so it can be compared side by side and added to the cart with a chosen quantity.
The commerce side gives each seller a shop dashboard for products, timed events, coupons, orders, refunds and withdrawals, while customers get a familiar storefront with wishlist, cart, multi-step checkout and order tracking. An admin dashboard ties the marketplace together.
Main features
AI design studio
Shoppers enter a prompt on the Generate Designs page and get an AI-generated artwork back, with loading and error states while the image is generated.
Automatic product mockups
Each generated design is rendered onto eight product mockup templates in parallel through the Dynamic Mockups API, then shown in a grid with a quantity selector and Add to Cart.
Design download
The generated artwork can be saved as a JPG directly from the browser, so the design is usable outside the store as well.
Seller shops
Sellers register with email activation and manage products, timed events with countdowns, coupon codes, orders, refunds, shop settings and withdrawal requests from their own dashboard.
Customer storefront
Home, best selling, product listings, product details with ratings and reviews, events, FAQ, wishlist, cart, checkout, profile and order tracking.
Stripe and PayPal checkout
Card payments use Stripe Elements with payment intents created on the server, alongside PayPal buttons and a cash on delivery option.
Buyer and seller messaging
Customers and sellers chat in real time over Socket.IO, with conversations and messages stored in MongoDB.
Admin dashboard
An admin role can review all users, sellers, orders, products, events and seller withdrawal requests.
Inside the product
Screens from the software itself. Design prototypes and illustrative mockups are labelled as such.
Development and architecture
MerchMind is split into two repositories: a React single-page app and an Express REST API backed by MongoDB. The frontend uses React Router for navigation, Redux Toolkit with thunks for state, Tailwind CSS for styling and Material UI data grids for the seller and admin tables.
The backend is a Node.js and Express service organised into controllers per domain (users, shops, products, events, coupons, orders, payments, conversations, messages and withdrawals), with Mongoose models and a central async error handler. The backend is no longer hosted, so the live site now shows the interface without store data.
- Parallel mockup rendering
- The eight mockup requests are fired together with Promise.all and failed renders are filtered out, so one broken template does not break the rest of the grid.
- Prompt template
- User prompts are wrapped in a fixed quality prefix before generation, aiming for more detailed artwork from short inputs.
- Separate user and seller sessions
- Authentication uses JWTs in cookies with separate tokens for customers and sellers, bcrypt password hashing and email activation links sent through Nodemailer.
- Role-based route guards
- The React app has distinct protected routes for customers, sellers and admins, so each dashboard is only reachable by the right role.
- Runtime Stripe configuration
- The Stripe publishable key is fetched from the API at runtime, and the Elements provider mounts only after it loads, so the Stripe configuration lives on the server.
- Cloud deployment
- The frontend was deployed to Netlify and the API moved from Vercel to Railway, with CORS restricted to the frontend origin. Uploaded images were stored on Cloudinary.
Technology
- React
- Node.js
- Express.js
- MongoDB
- JavaScript
- Tailwind CSS
- Redux Toolkit
- Stripe
- PayPal
- Socket.IO
- Cloudinary
- Netlify
Related projects
AI video captioning SaaS
Caption AI
A captioning web app for South Asian video creators, with Gemini transcription, timing measured from the audio and a full caption editor.
- Web application
- SaaS product
Multi-vendor optical marketplace
AsaanOptics Marketplace
A multi-vendor marketplace for optical shops in Pakistan, with prescription ordering, nearby search and escrow-style checkout, currently in development.
- Web application
- E-commerce
- SaaS product
Fashion e-commerce with made-to-measure tailoring
Shehmina's Wardrobe
A custom Laravel e-commerce prototype built around a womenswear label's catalog, with made-to-measure tailoring options and a full admin back office.
- Web application
- E-commerce
- Business system
Have a software idea or a business process that needs a proper system?
Tell me what you are trying to fix. I will reply with honest questions, a practical plan and a realistic estimate.