Case Study

Production

FDC WhatsApp Cheque Bot

A WhatsApp bot that reads photographed cheques with GPT-4o Vision and posts receipt vouchers straight into Busy accounting — the part I built independently on a team project.

FDC WhatsApp Cheque Bot

Stack

Technologies

FastAPIOpenAIEvolution APIPostgreSQLpgvectorLangGraphRedisGCPNginx

Context

Tags

aiautomationocr

Overview

For a distribution business that wanted every cheque in their Busy accounting software without staff learning it. An employee photographs a cheque on WhatsApp; the bot extracts amount/date/party/bank with GPT-4o Vision, matches the party (including learned alternative names), fetches unpaid bills, and writes the receipt voucher into Busy.

It was a team project; the WhatsApp cheque bot was mine to build and own.

Features

Cheque processing

  • GPT-4o Vision OCR with orientation handling; documented 10+ OCR approaches to get handwriting right.
  • Queue-based bulk processing — send a stack of cheques at once.
  • Accuracy is measured from operator approve/disapprove feedback in the dashboard.

Quotation & catalog

  • RAG catalog search with LangChain/LangGraph + pgvector + OpenAI embeddings.
  • Busy integration over an RDP/SQL-Server bridge for parties, bills, and vouchers.

Architecture

Stack

  • FastAPI + PostgreSQL/pgvector + Redis
  • WhatsApp via Evolution API
  • Deployed on GCP Compute Engine behind Nginx

Impact

Metric Before After Impact
Accuracy (self-reported, dashboard-measured) — — ~95%+ overall; ~99% catalog/party
Manual cheque entry 5–30 min photo on WhatsApp no accounting-software training