Chat360- Agentic AI

Advanced Django-based agentic AI backend system with knowledge base management, multi-service integrations, and real-time processing capabilities

---

🚀 Project Overview

This project was developed during my internship at Chat360 as a comprehensive backend solution for agentic AI applications. The system integrates advanced AI capabilities with robust backend infrastructure, featuring intelligent knowledge base management, multi-platform integrations, and asynchronous task processing.

Key Highlights

• 🧠 Agentic AI System: Advanced AI agent management with dynamic tool integration
• 📚 Knowledge Base: Vector database integration with Pinecone for intelligent document retrieval
• 🔗 Multi-Service Integration: Seamless connectivity with Shopify, Zoho, WhatsApp, and more
• ⚡ Real-time Processing: Celery-based async task queue for scalable operations
• 📄 Document Intelligence: Advanced processing for PDF, DOCX, Excel, and web content
• 🔐 Enterprise Security: Custom Bearer token authentication with role-based access

---

🛠️ Technology Stack

Category	Technologies
Backend	Django 5.2, Django REST Framework
Database	PostgreSQL, Redis (Cache)
Vector DB	Pinecone
Task Queue	Celery + RabbitMQ
AI/ML	OpenAI GPT, LangChain, Scikit-learn
Cloud	AWS S3, Docker
Document Processing	PyPDF2, python-docx, pandas
Integration APIs	Shopify, Zoho, Jina AI

---

⚙️ Quick Setup

Prerequisites

• Python 3.10+
• Node.js 18+
• Docker + Docker Compose (used to run Postgres, Redis, and RabbitMQ locally)

docker-compose.yml provides the infrastructure (Postgres, Redis, RabbitMQ) as containers; the Django backend, Celery worker, and frontend run natively on your machine for a faster dev loop (hot reload, easy debugging).

Installation

1. Clone the repository

git clone https://github.com/AaryanPuri/Chat360_Internship_Agentic-AI.git
cd Chat360_Internship_Agentic-AI

2. Set up environment variables (from the repo root, before cd backend)

cp .env.example .env

Edit .env with your actual credentials. At minimum:

• DJANGO_SECRET_KEY - generate one with python -c "import secrets; print(secrets.token_urlsafe(50))"
• OPENAI_API_KEY - required even to start the server, since several modules construct an OpenAI client at import time. A placeholder string works if you just want the server to boot; a real key is needed for AI features to actually respond.
• The POSTGRES_* / REDIS_* / RABBITMQ_* values already default to match docker-compose.yml, so you don't need to change those for local dev.

3. Start infrastructure (Postgres, Redis, RabbitMQ)

docker-compose up -d

4. Backend setup

cd backend
python -m venv venv
source venv/bin/activate   # On Windows: venv\Scripts\activate
pip install -r requirements.txt

5. Database setup

python manage.py migrate
python manage.py createsuperuser

6. Run the backend

python manage.py runserver

7. Start the Celery worker (separate terminal, from backend/, with the venv activated)

celery -A backend worker --loglevel=info

8. Run the frontend (separate terminal)

cd frontend
npm install
npm run dev

The frontend dev server proxies /api requests to http://localhost:8000, so make sure the backend (step 6) is running first. It's served at http://localhost:5173 by default.

---

🚀 Core Features

🤖 Agentic AI System

• Dynamic Tool Integration: 18+ built-in tools for various operations
• Custom Tool Creation: User-defined tools with OpenAPI schema support
• Multi-Agent Workflows: Parallel and sequential agent execution
• Context-Aware Processing: Intelligent conversation management

📚 Knowledge Base Management

• Vector Search: Pinecone-powered semantic search
• Multi-Format Support: PDF, DOCX, Excel, web scraping
• Real-time Indexing: Automatic document processing and embedding
• Hybrid Retrieval: Dense + sparse vector search

🔗 Platform Integrations

• Shopify: Order tracking, product recommendations, returns
• Zoho CRM: Customer data synchronization
• WhatsApp Business: Chat analytics and automation
• Custom APIs: Flexible webhook-based integrations

---

Screenshots

Dashboard	Ask AI
Knowledge Base	Abilities & Integrations
AI Assistants	Assets

---

📚 API Documentation

Comprehensive API documentation is available in API_DOCUMENTATION.md.

Key Endpoints

Endpoint	Description
POST /api/analytics/chat/	AI chat interface
GET /api/analytics/assistants/	Agent management
POST /api/analytics/knowledgebase/create/	Create knowledge base
POST /webhook/	Webhook handler
GET /oauth/zoho/	Zoho OAuth integration

---

🧪 Testing

Run all tests

python manage.py test

Run specific test modules

python manage.py test analytics.tests

With coverage

pip install coverage
coverage run --source='.' manage.py test
coverage report

---

🏗️ Development Workflow

This project follows professional development practices:

• Pre-commit Hooks: Automated code formatting and linting
• Docker Support: Containerized development environment
• Comprehensive Logging: Structured logging with rotation
• API Documentation: Auto-generated OpenAPI specifications
• Error Handling: Robust exception management and recovery

---

📈 Performance & Scale

• Async Processing: Celery for background tasks
• Caching Strategy: Redis-based multi-level caching
• Database Optimization: Efficient query patterns and indexing
• Vector Storage: Scalable Pinecone integration
• Load Balancing Ready: WSGI/ASGI compatible

---

📄 License

This project was developed by me during my internship at Chat360, and is shared here for showcase purposes only.

---

👤 Aaryan Puri