LexiOps - Intelligent Legal AI Platform

Advanced AI-powered legal information retrieval and DevOps automation platform with RAG (Retrieval-Augmented Generation) and MCP (Model Context Protocol)

📊 Live Project Tracker - Architecture, Kanban & Development Progress →

🏗️ System Architecture

📊 Complete System Overview

🤖 DevOps Copilot Architecture

🌟 Application Architecture

🎯 Overview

LexiOps is a cutting-edge dual-platform system that revolutionizes both legal information access and DevOps automation through advanced AI technologies. The platform combines RAG (Retrieval-Augmented Generation) with MCP (Model Context Protocol) to deliver precise legal insights and intelligent Kubernetes cluster management.

🌟 Dual Platform Architecture

🏛️ Legal AI Platform

Intelligent RAG System: Advanced retrieval-augmented generation for legal document analysis
Real-time Legal Chat: WebSocket-powered instant legal assistance with streaming responses
Multi-Source Integration: Legal databases, web search, and knowledge APIs
Vietnamese Legal Focus: Specialized for Vietnamese legal corpus with 180k+ documents
Vector Search: Semantic search with Qdrant database and OpenAI embeddings

🤖 DevOps AI Copilot

Agentic Workflow System: GPT-4 powered automation for Kubernetes operations
MCP Integration: Model Context Protocol with 51+ dynamic Kubernetes tools
3-Tier Architecture: FastAPI + MCP Client + Tools Server
Intelligent Decision Making: LangGraph-based workflow orchestration
Cluster Management: AI-powered infrastructure automation

🚀 Key Innovations

Dual AI Engines: Legal RAG + DevOps Automation in unified platform
Real-time Processing: Sub-second response times with intelligent caching
Microservices Design: Scalable, containerized architecture with specialized workers
Production Ready: Enterprise-grade security, monitoring, and deployment
Open Source: MIT licensed with comprehensive documentation

🏷️ Technology Stack

AI & Machine Learning:

Backend & Infrastructure:

🚀 Quick Start

Prerequisites

Docker 20.10+ and Docker Compose 2.0+
OpenAI API Key for AI functionality

Installation

# 1. Clone repository
git clone https://github.com/tinhnguyen0110/LexiOps.git
cd LexiOps

# 2. Setup environment
cp .env.example .env
# Edit .env with your OPENAI_API_KEY

# 3. Start system
make quick-start
# OR: docker-compose up -d

# 4. Verify health
make health-check

Access Points

Service	URL	Description
Frontend	http://localhost:3000	Main chat interface
API Docs	http://localhost:8000/docs	Backend documentation
Monitor	http://localhost:5555	Celery task monitor

📁 Project Structure

LexiOps/
├── 📁 src/                           # Main legal retrieval system
│   ├── 📁 app/                       # 🚀 FastAPI Backend
│   │   ├── brain.py                  # 🧠 AI reasoning engine
│   │   ├── main.py                   # 🌐 API server entry
│   │   ├── celery_config.py          # 🔄 Task queue config
│   │   ├── 📁 api/                   # API endpoints
│   │   ├── 📁 config/                # Configuration
│   │   └── 📁 utils/                 # Utilities
│   │
│   ├── 📁 legal-chatbot-fe/          # ⚛️ React Frontend
│   │   ├── src/App.jsx               # Main component
│   │   ├── 📁 components/            # UI components
│   │   └── 📁 hooks/                 # Custom hooks
│   │
│   └── 📁 streamlit-fe/              # 🎯 Alternative frontend
│
├── 📁 lexiops-copilot/               # 🤖 DevOps AI Copilot System
│   ├── 📁 agent/                     # Core AI agent
│   │   ├── main.py                   # FastAPI copilot server
│   │   ├── graph.py                  # LangGraph workflow
│   │   ├── mcp_server.py             # MCP server configuration
│   │   └── 📁 nodes/                 # AI processing nodes
│   ├── 📁 k8s/                       # Kubernetes deployments
│   │   └── mcp-k8s-deployment.yaml   # MCP tools server
│   ├── 📁 scripts/                   # Deployment scripts
│   │   ├── run.sh                    # Start copilot system
│   │   └── deploy-3tier.sh           # 3-tier deployment
│   ├── 📁 test/                      # Integration tests
│   ├── 📁 docs/                      # Architecture docs
│   └── 📁 notebooks/                 # Development notebooks
│
├── 📁 scripts/                       # 🔧 Deployment scripts
│   ├── build-docker.sh               # Docker builder
│   ├── system-check.sh               # Health validator
│   └── deploy-pipeline.sh            # Deployment automation
│
├── 📁 tests/                         # 🧪 Test suite
│   ├── run_docker_tests.sh           # Test runner
│   ├── DOCKER_TESTING_GUIDE.md       # Testing docs
│   └── test_*.py                     # Test files
│
├── 📁 helm/                          # ☸️ Kubernetes deployment
│   ├── helmfile.yaml                 # Helm orchestration
│   ├── 📁 charts/                    # Custom charts
│   └── 📁 values/                    # Environment configs
│
├── 📁 data/                          # 📚 Legal corpus
│   ├── 📁 data_corpus/               # 180k+ legal documents
│   ├── 📁 mongo_data/                # Database storage
│   └── 📁 qdrant_storage/            # Vector database
│
├── 📁 deployment/                    # 🚀 Deployment configs
├── 📁 terraform/                     # 🌩️ Infrastructure as Code
├── 📁 reports/                       # 📈 System reports
├── docker-compose.yml               # 🐳 Container orchestration
├── Makefile                          # 🛠️ Commands
└── README.md                         # 📖 Documentation

Core Components

Component	Purpose	Key Features
brain.py	AI reasoning engine	Query analysis, tool selection, response generation
legal_chat.py	WebSocket API	Real-time chat, conversation management
App.jsx	Frontend app	Chat interface, message handling
useChat.js	Chat logic	WebSocket management, state handling
Copilot System	DevOps automation	K8s management, AI-powered operations

🛠️ Development

Backend Development

# Docker development (recommended)
make dev-up

# Local development
cd src/app
pip install -r requirements.txt
uvicorn main:app --reload --port 8000

Frontend Development

cd src/legal-chatbot-fe
npm install
npm run dev          # Development server
npm run build        # Production build

DevOps Copilot

# Start copilot system
cd lexiops-copilot
./scripts/run.sh

# Test copilot functionality
python test/test_fastapi_mcp.py

Environment Setup

# Required in .env file
OPENAI_API_KEY=your_openai_api_key_here
MONGODB_URL=mongodb://admin:password123@localhost:27017/legaldb?authSource=admin
REDIS_URL=redis://localhost:6379/0
QDRANT_URL=http://localhost:6333

🐳 Deployment

Docker Deployment

# Local deployment
docker-compose up --build -d

# Production deployment
docker-compose -f docker-compose.yml --env-file .env.production up -d

# Scale workers
docker-compose up --scale celery-worker-rag=3 -d

Kubernetes Deployment

# Deploy with Helm
helmfile -e development apply    # Development
helmfile -e production apply     # Production

# Deploy copilot MCP server
kubectl apply -f lexiops-copilot/k8s/mcp-k8s-deployment.yaml

# Monitor deployment
kubectl get pods -n legal-retrieval-prod

Available Commands

make quick-start     # Quick setup and start
make health-check    # System health verification
make test            # Run test suite
make clean-all       # Complete cleanup
make help            # Show all commands

🧪 Testing

Test Categories

Smoke Tests: Quick validation (2-3 minutes)
API Tests: Endpoint validation
Integration Tests: Service interaction
Performance Tests: Load and stress testing

Running Tests

# Quick smoke tests
make test-smoke

# Full test suite
make test

# Docker-isolated testing
./tests/run_docker_tests.sh --command test --test-type smoke

# Copilot system tests
cd lexiops-copilot
python test/test_fastapi_mcp.py

For detailed testing: tests/DOCKER_TESTING_GUIDE.md

🔧 Configuration

Core Settings

# API Configuration
FASTAPI_HOST=0.0.0.0
FASTAPI_PORT=8000
DEBUG=true

# AI Configuration
OPENAI_MODEL=gpt-4o-mini
EMBEDDING_MODEL=text-embedding-3-small
VECTOR_SIZE=1536

# Worker Configuration
CELERY_CONCURRENCY=2
CELERY_BROKER_URL=redis://localhost:6379/3

Service Scaling

The system supports horizontal scaling:

RAG Workers: CPU-intensive AI processing
Embed Workers: Vector embedding generation
Retrieval Workers: Web search and content retrieval

System Architecture

Legal Retrieval System

Frontend: React chat interface (port 3000)
Backend: FastAPI with WebSocket (port 8000)
Workers: Celery for RAG processing
Databases: MongoDB + Redis + Qdrant

DevOps Copilot (3-Tier Architecture)

Tier 1: FastAPI main server (lexiops-copilot/agent/main.py)
Tier 2: MCP client adapter (langchain_mcp_adapters)
Tier 3: MCP tools server (Kubernetes deployment)

📊 Monitoring

Dashboards

Celery Monitor: http://localhost:5555 - Task monitoring
API Health: http://localhost:8000/health - System status
Qdrant Dashboard: http://localhost:6333/dashboard - Vector DB

Performance Metrics

Response Time: < 2s simple queries, < 5s complex queries
Throughput: 100+ concurrent users, 50+ RPS
Resource Usage: 2-4GB RAM, 2-4 CPU cores

Troubleshooting

# Check service status
docker-compose ps

# View logs
docker-compose logs -f backend-api

# System health check
./scripts/system-check.sh

🏆 Production Ready

LexiOps includes enterprise-grade features:

✅ Legal AI System - RAG-powered legal information retrieval
✅ DevOps Copilot - AI-powered Kubernetes cluster management
✅ Scalable Architecture - Microservices with auto-scaling
✅ Real-time Chat - WebSocket-powered interface
✅ Comprehensive Testing - Full test coverage
✅ DevOps Ready - Docker, Kubernetes, CI/CD
✅ Performance Optimized - Sub-second response times

Additional Resources

📈 Reports: reports/OPTIMIZATION_REPORT.md
🚀 Deployment: deployment/ configs
📚 Documentation: docs/ guides
🤖 Copilot Docs: lexiops-copilot/docs

LexiOps - Transforming Legal Information Access with AI 🏛️🤖

Built with ❤️ using FastAPI, React, OpenAI, Docker, and modern DevOps practices

Name		Name	Last commit message	Last commit date
.github		.github
.motia		.motia
argocd-manifests		argocd-manifests
docs		docs
helm		helm
lexiops-copilot		lexiops-copilot
scripts		scripts
src		src
terraform		terraform
tests		tests
.env.config		.env.config
.env.example		.env.example
.gitignore		.gitignore
CODE_OF_CONDUCT.md		CODE_OF_CONDUCT.md
CONTRIBUTING.md		CONTRIBUTING.md
LICENSE		LICENSE
Makefile		Makefile
README.md		README.md
docker-compose.dev.yml		docker-compose.dev.yml
docker-compose.yml		docker-compose.yml
requirements.txt		requirements.txt

Latest commit

History

Folders and files

Repository files navigation