🚀 Resolvix AI

AI-Powered Customer Support Platform built with Microservices, RAG & LLMs

An intelligent customer support platform that automatically analyzes support tickets, retrieves similar historical cases using Retrieval-Augmented Generation (RAG), and generates contextual AI-powered response suggestions.

---

📖 Overview

Resolvix AI is a production-style AI-powered customer support platform designed using a microservices architecture.

The system automatically analyzes incoming support tickets using Retrieval-Augmented Generation (RAG) and Large Language Models (LLMs) to:

• 🎫 Categorize tickets
• 🚨 Predict ticket priority
• 😊 Detect customer sentiment
• 🔍 Retrieve similar historical tickets
• 💬 Generate contextual AI reply suggestions

The platform follows an event-driven architecture where services communicate asynchronously through Redis Pub/Sub, making the system scalable and loosely coupled.

---

🏗️ System Architecture

Replace the image below with your architecture diagram.

---

✨ Features

Authentication

• JWT Authentication
• User Registration & Login
• Role-Based Authorization

Ticket Management

• Create Support Tickets
• Dashboard
• Ticket Details
• Ticket Tracking

AI Ticket Triage

• Automatic Ticket Classification
• Priority Detection
• Sentiment Analysis
• Context Retrieval using RAG
• AI Generated Reply Suggestions

Architecture

• Microservices
• API Gateway
• Redis Pub/Sub
• Event-Driven Processing
• Dockerized Services
• PostgreSQL
• Chroma Vector Database

---

⚡ System Workflow

Customer
    │
    ▼
Next.js Frontend
    │
    ▼
API Gateway
    │
    ▼
Ticket Service
    │
    ▼
PostgreSQL
    │
    ▼
Redis Pub/Sub
    │
    ▼
AI Triage Service
    │
    ├── HuggingFace Embeddings
    ├── Chroma Vector Database
    └── Groq LLM
    │
    ▼
Redis
    │
    ▼
Ticket Service
    │
    ▼
Frontend Updates

---

🏛️ Microservices

Service	Responsibility
API Gateway	Entry point for all client requests
Auth Service	Authentication & JWT
Ticket Service	Ticket CRUD operations
AI Triage Service	AI Analysis using RAG & LLM
Notification Service	Event notifications

---

🧠 AI Pipeline

Ticket Created
      │
      ▼
Generate Embeddings
      │
      ▼
Retrieve Similar Tickets
      │
      ▼
Construct Prompt
      │
      ▼
Groq LLM
      │
      ▼
Category
Priority
Sentiment
Suggested Reply

---

🛠️ Tech Stack

Frontend

• Next.js
• React
• TypeScript
• Tailwind CSS

Backend

• FastAPI
• Python
• SQLAlchemy
• JWT Authentication

AI

• Groq LLM
• Hugging Face Embeddings
• Chroma Vector Database
• Retrieval-Augmented Generation (RAG)

Infrastructure

• PostgreSQL
• Redis
• Docker
• Docker Compose

---

📂 Project Structure

Resolvix-AI
│
├── frontend
├── auth-service
├── ticket-service
├── ai-triage-service
├── notification-service
├── api-gateway
│
├── docker-compose.yml
├── ARCHITECTURE.md
└── README.md

---

🚀 Getting Started

Clone the repository

git clone https://github.com/cl0ud08/Resolvix-AI.git

cd Resolvix-AI

Create environment variables for each service.

Example

DATABASE_URL=
REDIS_URL=
JWT_SECRET=
GROQ_API_KEY=

Run

docker-compose up --build

---

🎯 Future Enhancements

• Agent Assignment
• Email Notifications
• Slack Integration
• Kubernetes Deployment
• CI/CD Pipeline
• Monitoring with Prometheus & Grafana
• Multi-tenant Support
• OAuth Login

---

📚 Key Concepts Demonstrated

• Microservices Architecture
• Event-Driven Systems
• API Gateway Pattern
• JWT Authentication
• Retrieval-Augmented Generation (RAG)
• Vector Databases
• Semantic Search
• Redis Pub/Sub
• Asynchronous Processing
• Dockerized Deployment
• AI-powered Ticket Triage

---

👨‍💻 Author

Harshit Gupta

AI/ML Engineer • Backend Developer • Full Stack Developer

• GitHub: https://github.com/cl0ud08
• X: https://x.com/0xhrshit

---

⭐ If you found this project useful, consider giving it a star!