Imad Najam I build products end to end.

Interface, API, database, deploy I own the whole path. Payments, multilingual storefronts and AI document pipelines, running in production today.

Casablanca, Morocco Remote-friendly · CET/UTC+1

Imad Najam, full-stack developer, in a suit and glasses.
Scroll

Three products, all running in production.

Shipped end to end interface, API, data layer, deploy. Open a row for the engineering detail.

  1. Lara

    Multi-tenant commerce & payments platform Live 2024

    A commerce platform that lets local merchants take payments, track inventory and sell tickets from one dashboard.

    Problem
    Small merchants were running inventory, sales and payouts across disconnected tools, with no single view of what was actually selling.
    My role
    Full-stack: architecture, payment integration, i18n, deployment

    How it was built

    • Integrated Stripe for card payments, handling webhook-driven state so an interrupted checkout never leaves an order in limbo.
    • Built the interface in Next.js and TypeScript with shadcn/ui, typed end to end against the API contract.
    • Shipped full internationalisation, so the storefront runs in multiple languages from a single codebase.
    • Backed it with an Express and MongoDB service layer, deployed on Vercel.

    Stack

    • Next.js
    • TypeScript
    • Stripe
    • Node.js
    • Express
    • MongoDB
    • i18n
    • shadcn/ui
  2. FRIDA

    AI document-to-Markdown pipeline Live 2025

    Converts Word, Excel, PowerPoint and PDF files into clean Markdown while preserving document structure.

    Problem
    Naive document extraction flattens everything. Headings, tables and lists all collapse into undifferentiated text, so the output needs rewriting by hand before it is usable.
    My role
    Full-stack: model integration, conversion pipeline, interface

    How it was built

    • Used GPT-NeoX to recover semantic structure rather than raw text, so headings, tables and lists survive the conversion.
    • Split the system across a Next.js front end and a FastAPI service, keeping the Python ML dependencies isolated from the web tier.
    • Added real-time preview so users can verify the output before committing to a download.
  3. Undermovies

    Streaming platform Live 2024

    A film and series streaming application with accounts, catalogue browsing and transactional email.

    Problem
    Building a media catalogue that stays responsive as the library grows means getting pagination, caching and access control right from the start.
    My role
    Full-stack: application, auth, external API integration, deployment

    How it was built

    • Built on Laravel with authenticated accounts and role-gated access to the catalogue.
    • Integrated a third-party catalogue API for metadata, with a transactional email flow for account actions.
    • Deployed and maintained on Railway.

More on GitHub (opens in a new tab)

What I can own, from interface to database.

Grouped by responsibility rather than listed alphabetically. The bold items are what I reach for first.

Frontend

Interfaces that stay fast as they grow

  • TypeScript
  • React
  • Next.js
  • JavaScript
  • Sass
  • shadcn/ui
  • i18n

Backend

APIs, auth, payments, background work

  • Laravel
  • FastAPI
  • Node.js
  • PHP
  • Python
  • Express
  • C# / .NET

Data

Schema design, queries, migrations

  • MySQL
  • MongoDB
  • SQL Server

Platform

Getting it shipped and keeping it up

  • Git
  • Docker
  • GitHub Actions
  • Vercel
  • Railway
  • Stripe

Have something you need built?

Tell me what you're working on and I'll come back to you within a day.

Reach me directly

Email imad.najam.j@gmail.com LinkedIn Message me on LinkedIn (opens in a new tab) GitHub Imadnajam (opens in a new tab)

Based in Casablanca, Morocco · Remote-friendly · CET/UTC+1

Goes straight to my inbox. No newsletter, no list.