🤖 E-Commerce Customer Support AI Agent

This repository contains the source code and project structure for a modular, scalable E-Commerce Customer Support AI Agent. The agent is designed to provide intelligent, automated support for customers and can integrate with various APIs and tools for seamless functionality.

An AI-powered customer support agent for e-commerce, built with Rasa and FakeStore API. Handles order tracking, product inquiries, and returns, with seamless integration for mock data.

Live Demo (Not Ready Yet): Streamlit App | API Docs

Features ✨

• Order Tracking: "Where is my order #123?" → Real-time status from FakeStore API.
• Product Stock Checks: "Is the Fjallraven Backpack in stock?" → Inventory lookup.
• Returns/Refunds: Guided return process via dynamic forms.
• RestockNotification: Notify the customer when the product is back in stock
• Rasa Integration: NLP intents, entity extraction, and dialogue management.
• Integrated LLM: Integrated DeepSeek-R1 for Reasoning i.e. to handle ambiguous queries and generate dynamic responses.
• Mock Data: No need for a real e-commerce backend.

Tech Stack 🛠️

• NLP & Dialogue: Rasa (NLU + Core)
• Backend: Python, Flask, PostgreSQL
• Integrations: FakeStore API (mock data), Shopify/Stripe API connections, Gmail/SMTP email automation
• Deployment: Docker, AWS/Heroku
• Frontend: Streamlit with custom CSS design
• LLM for Reasoning: DeepSeek-R1

🚧 Roadmap

Completed

• Core Rasa/DeepSeek integration
• Basic order tracking & returns
• Streamlit UI framework
• FakeStore API integration
• Context-aware conversation handling
• LLM Integration for Reasoning build part(deployment WIP)
• Database Integration:
  • PostgreSQL for order history and product return/refund enquiry
  • Redis for real-time session storage
  • User preference persistence

Immediate Priorities/Work-in-progress Items:

• External Service Integration

  • Shopify/Stripe API connections
  • Gmail/SMTP email automation
  • Warehouse inventory system hooks

• Enhanced Autonomy

  • Automated refund processing
  • Proactive shipment updates
  • Smart cart recovery workflows

Advanced Features that are in plan

• Self-Improvement System

  • Rasa Interactive Learning integration
  • Hugging Face Transformers fine-tuning
  • User feedback analysis pipeline

• Production Readiness

  • Docker/Kubernetes deployment
  • Prometheus/Grafana monitoring
  • Load testing & scaling

Future Exploration

• Have thought of 4 amazing features or further integrations that can be added in the future but I'm not willing to disclose them now. If you're an Interviewer or a product owner, I can definitely share those future scopes in an offline conversation.