EasyBela

Problem

Egypt's e-commerce stack is fragmented: most small merchants either sell through Facebook DMs or rent a heavy shared platform that doesn't speak Arabic well and doesn't support local payment rails like Paymob and cash-on-delivery. EasyBela is the consolidated marketplace those merchants need, with the operational discipline (moderation, payouts, dispute flow) they don't want to build themselves.

Architecture

Three surfaces over one backend:

• API: Django + DRF, Postgres, Celery for async work (notifications, payouts, image processing).
• Storefront: Next.js App Router + TypeScript + Tailwind + Ant Design.
• Mobile: Flutter — single codebase, iOS and Android.

The accounts app is fully built (auth, OTP, profile, merchant onboarding); catalog, orders, payments are scaffolded with end-to-end happy paths and hardening in flight.

Key decisions

• Multi-merchant from day one. A single-merchant build would have been faster, but the moderation and payout flows are the actual moat — they had to land in the schema before any seller went live.
• Bilingual at the model layer. Translatable string fields are on the models, not bolted on as a JSON afterthought. Arabic is the primary locale; English is the fallback. Layout direction switches per locale.
• Paymob first, COD second. Paymob covers card and wallet rails Egypt expects. COD is a hard requirement for the trust gap with new buyers.

Outcome

Foundations are stable; the marketplace is in pre-launch with the first cohort of merchants onboarding. The build is intentionally narrative-only on this site — the live URL is a customer-controlled property and not shareable publicly.