WhatsApp AI Expense Tracker.
Lets individuals and small teams log every expense by sending a photo, voice note, or quick text message to a WhatsApp number. AI extracts the amount, category, vendor, and date — eliminating the friction that kills 95% of expense-tracking apps.
Key features.
- WhatsApp Business API integration for two-way conversational logging
- Vision model parses receipts, invoices, and handwritten notes
- Automatic categorisation with custom user-trained tag rules
- Real-time dashboard with monthly burn, category breakdown, anomaly detection
- Export to CSV / Excel / accounting tools (Xero, QuickBooks)
Architecture.
Event-driven webhook pipeline. WhatsApp messages hit a webhook, queue into Redis, get processed by an AI worker, write to PostgreSQL, and trigger a real-time dashboard update via WebSockets. Failures retry with exponential backoff.
Tech stack.
- Frontend
- Next.js 14 · Tailwind · Recharts
- Backend
- Node.js · WhatsApp Cloud API webhook
- Database
- PostgreSQL · Redis (queue)
- AI
- OpenAI Vision · Whisper for voice transcription · structured-output extraction
- Infrastructure
- Vercel (edge) · Cloudflare R2 for receipt storage
Target users.
Freelancers, indie hackers, small business owners, expat households
Unique selling points.
- Zero learning curve — uses an app every user already has open
- AI handles unstructured input (handwritten notes, multi-language receipts, voice memos)
- Privacy-first: receipts never leave your tenant; AI calls run on isolated keys
Monetization.
Freemium SaaS — 50 logs/month free, paid tiers for unlimited and family/team accounts