Wapptel

The official WhatsApp Business API is expensive and slow to get approved for a small business that just wants to send a bulk promotion to its customer list. Wapptel solves that by automating WhatsApp Web itself — connect an account by scanning a QR code, then manage contacts, run bulk-messaging campaigns and track delivery from a dashboard, without touching Meta's paid API.

How it works

• A three-tier design: a browser SPA talks to a Django backend, which talks to a separate Node.js microservice that holds the actual WhatsApp Web session.
• Campaigns accept an Excel/CSV recipient list, validate Bangladeshi phone numbers, and send with rate limiting and progress tracking to keep accounts from being flagged as spam.
• Multi-tenant from the start — organizations, users and subscriptions are first-class, not bolted on later.

Technology

Django REST Framework with JWT and API-key auth, PostgreSQL (+PostGIS), Celery and django-redis for background work, drf-spectacular for API docs, and Stripe for subscription billing. The WhatsApp connection itself is a Node.js/Express microservice built on @whiskeysockets/baileys (WhatsApp Web protocol), with Socket.IO for live QR/connection status and PM2 for process management. LangChain/LangGraph on top of OpenAI power smarter reply and campaign-content features.

My Contributions

• Designed the three-tier architecture and built the Django backend end-to-end.
• Built the Node.js/Baileys microservice for WhatsApp Web connection and messaging.
• Implemented the bulk-campaign feature — recipient upload, phone validation, rate-limited sending and progress tracking.
• Set up Celery-based background processing and Stripe subscription billing.