I Started Building an E-Commerce Platform to Help a Friend Launch a Fashion Brand. It Turned Into Something Much Bigger.

Marry — Wed, 24 Jun 2026 14:33:28 +0000

A couple of month ago, I wasn't planning to build an e-commerce platform.

A friend of mine is a fashion designer. He was preparing to launch a new collection and needed a website. At first, it sounded like a straightforward project: a storefront, product pages, checkout, and some content around the collection.

So I started building.

Like many side projects, the first version was supposed to be temporary. I thought we would launch the collection, make a few sales, and move on.

But very quickly I ran into a problem.

Most e-commerce solutions felt optimized for selling products, not for building brands.

Everything revolved around catalogs, filters, and templates. The brand story was usually an afterthought.

For a fashion brand, content is often just as important as products. Editorial photography, lookbooks, campaigns, stories, and product presentation all work together. Traditional storefronts separate these things instead of connecting them.

I wanted something different.

Instead of treating content and commerce as separate systems, I started building them as one product layer.

Lookbooks could lead directly into products.

Stories could become shopping journeys.

Editorial pages could connect naturally to the catalog.

At first, these were just custom features for a single project.

Then more requirements appeared.

Multiple languages.

Different markets.

Custom checkout flows.

Content management.

Inventory integration.

Admin tools.

AI-powered assistance.

What started as a small storefront slowly became a complete commerce platform.

At some point I realized I was no longer maintaining a project for a single brand.

I was building infrastructure.

Today that project is called SIB-ECOM.

It's an open-source commerce platform built around a simple idea:

Brands should own their infrastructure and have complete freedom to shape how their products are presented.

The stack currently includes:

SvelteKit storefront
Express + TypeScript API
PostgreSQL with Prisma
S3 / MinIO storage
Multi-market support
Content management tools
Checkout and customer accounts
AI and RAG-powered workflows
Self-hosted deployment

The most important thing is that it isn't a demo.

The platform is already running in production for a real fashion brand Shisterov.to

Building it also changed how I think about software.

The internet already has enough platforms encouraging people to produce more content, more products, and more noise.

I'm more interested in helping people build better products.

Many independent brands don't need another marketplace or another SaaS subscription.

They need tools they can control.

That's ultimately why SIB-ECOM became open source.

sibstones / sib-ecom

E-com AI platform for managing global e-commerce operations, content, and business workflows from a single dashboard. Open-source e-commerce platform built with TypeScript, PostgreSQL and Docker for launching online stores.

SIB CMS

Full-featured e-commerce platform for a store: storefront on SvelteKit, backend API on Express + TypeScript, PostgreSQL via Prisma, file storage in S3/MinIO, and a separate RAG setup for GPT Assistant.

Repository Layout

frontend/ — storefront and admin UI built with SvelteKit
backend/ — API, business logic, Prisma schema, seeds, and utility scripts
docker-compose.dev.yml — local development stack with PostgreSQL, pgvector, and MinIO
docker-compose.yml — production / self-hosted application stack
Makefile — commands for startup, updates, migrations, logs, and backups

Stack

Frontend: SvelteKit, Svelte 5, Vite, Tailwind CSS
Backend: Node.js, Express, TypeScript, Prisma
Database: PostgreSQL
File storage: MinIO or any S3-compatible storage
AI / RAG: OpenAI SDK, separate PostgreSQL + pgvector database for RAG
Infrastructure: Docker Compose, Makefile

Project Areas

Use this as a quick map of where each part of the…

View on GitHub

Not because open source is trendy, but because the people building thoughtful products should have ownership over the systems they depend on.

What started as a favor for a friend became a platform, a production system, and an ongoing experiment in building better tools for independent brands.

And honestly, that's a much more interesting project than the one I originally planned to build.