AutomationSTATUS: Open Source 1 0

WhatsApp Automation Hub

NLP-powered customer interaction orchestrator

A Python backend designed to process concurrent WhatsApp Webhooks. By combining intent classification with Redis state management, it handles routine support queries and provides fallback mechanisms for human agents.

Inspect GitHub Source Code

01 // THE BUSINESS PROBLEM & SITUATION

Operational Context

Service businesses relying on WhatsApp face response bottlenecks during peak hours due to repetitive manual inquiries.

CLIENT / ENVIRONMENT: Created for service businesses using WhatsApp as a primary client communication channel.

02 // SYSTEM ARCHITECTURE & SOLUTION

What Was Built

Built a webhook orchestrator using Python and FastAPI. The system classifies incoming intent and dispatches defined logical flows, escalating complex cases to human staff.

WHATSAPP.ORCHESTRATOR // WEBHOOKSystem Architecture Diagram

WEBHOOK

WhatsApp API

CLASSIFIER

FastAPI NLP

STATE

Redis Cache

ESCALATE

Human Desk

SYSTEMS_ARCHITECTURE_EXPLODER // LAYERED_STACKFULL-STACK SYSTEM DECOUPLING

Frontend Layer

Next.js 16 App Router, React 19, Tailwind CSS, TypeScript

API & Middleware

FastAPI, Pydantic Schema Validation, Rate Limiters

Orchestration

LangGraph Multi-Agent Supervisor & State Machines

Model Providers

OpenAI GPT-4o, Anthropic Claude 3.5, Vision AI

Data Stores

PostgreSQL (Prisma), Qdrant Vector Store, Redis Caching

Observability

LangSmith traces, Pydantic audit logs, Error retries

03 // ENGINEERING CONTRIBUTION & FEATURES

Exact Contribution

Developed the FastAPI webhook receiver
Implemented the NLP intent classification module
Built a Redis-based session state store
Designed the human-agent escalation protocol

Key Capabilities Built

✓ Intent Classification

✓ Redis Session State Management

✓ Human-Agent Handoff Protocol

✓ Automated Flow Routing

DATA_PROVENANCE_TIMELINE // AUDIT_TRACETRACE_ID: #TR-992041

14:22:01.002Source Ingestion

Received webhook payloadOK

14:22:01.045Schema Validation

Pydantic structured field checkOK

14:22:01.210LangGraph Reasoning

Evaluated criteria rulesOK

14:22:01.350Human Audit Check

Confidence threshold evaluationPASSED

14:22:01.480Database Commit

Created audit record in PostgreSQLCOMMITTED

*Illustrative system trace — demonstration data, not a client result.

04 // EXECUTION PIPELINE

Step-by-Step Workflow

STEP 01Client sends a WhatsApp message.

STEP 02Webhook payload is parsed and validated.

STEP 03Intent classifier categorizes the message goal.

STEP 04Redis state machine maintains conversational context.

STEP 05Automated response or human escalation is dispatched.

05 // ENGINEERING LESSONS & OUTCOMES

Technical Challenges Overcome

• Maintaining conversational memory across asynchronous webhook callbacks.
• Handling low-confidence classifier predictions gracefully.

Measurable Outcomes

→ Automated tier-1 FAQ responses.

→ Standardized Redis state persistence across worker instances.

Lessons & Engineering Rules

"Stateless webhook handlers require explicit caching layers for conversational continuity."
"Fallback paths should be triggered when intent confidence drops below threshold."

NEXT RELATED CASE STUDY

Competitor Tracking Engine

View Study

PROJECT METADATA

ARSLAN'S ROLEBackend Automation Engineer

YEAR & STATUS2025 // Open Source

CATEGORYAutomation

TECHNOLOGY STACK

PythonWhatsApp Business APIFastAPIRedisNLP

GITHUB METRICS

Stars: 1

Forks: 0

Verified GitHub Update: 7/17/2026

Need a similar solution?

If your business faces similar data, automation, or software bottlenecks, let's discuss your requirements.

Start a Conversation