🚀 InfraWatch AI – Multi-LLM DevOps Copilot

AI-Powered Infrastructure Monitoring, Incident Analysis & Multi-LLM Platform

Monitor infrastructure, analyze logs with Gemini AI, and compare responses across multiple LLMs from a single platform.

Built with Next.js, Clerk, Supabase (PostgreSQL), Docker, Redis, Nginx, Python, Gemini API, and GitHub Actions.

---

📌 Overview

InfraWatch AI started as a Multi-LLM Chat System and evolved into a complete DevOps + AIOps platform.

The system enables users to:

• Compare responses from multiple LLMs simultaneously
• Persist chat history securely
• Monitor infrastructure health in real time
• Analyze server logs autonomously using AI
• Detect incidents automatically and log them
• Track API usage and token metrics
• Manage deployments using DevOps best practices

---

🏗 Architecture

User
 │
 ▼
Clerk Authentication
 │
 ▼
Nginx Reverse Proxy
 │
 ▼
Next.js Application
 │
 ├────────► Gemini API (Native)
 │
 ├────────► OpenRouter API (GPT-4o, Claude 3.5, Mistral, etc.)
 │
 ├────────► Supabase PostgreSQL (via Prisma ORM)
 │
 └────────► Redis Cache
                 │
                 ▼
          Python Monitoring Service
                 │
                 ▼
         AI Log Analyzer (Gemini)
                 │
                 ▼
        InfraWatch Dashboard

---

✨ Core Features

🤖 Multi-LLM Chat System

• GPT-4o Support
• Gemini 1.5 Flash Support
• Claude 3.5 Sonnet Support
• Mistral, Perplexity, & Cohere Support
• Side-by-side response comparison
• Auto-resizing input & full Markdown rendering
• Customizable temperature settings

---

🔐 Authentication

Powered by Clerk.

Features:

• Secure Session Management
• Row Level Security (RLS) in Database
• Protected Dashboard Routes

---

💬 Persistent Chat History

Stored in Supabase PostgreSQL via Prisma ORM.

Features:

• Secure conversation storage (RLS enforced)
• Cost calculation per message
• Side-navigation to revisit past chats

---

📊 API Usage Analytics

Track:

• Total LLM Requests
• Token Consumption per model
• Estimated API Cost (USD)
• Model Usage comparisons

---

⚡ Redis Caching

Features:

• Prompt caching to reduce identical API calls
• Faster repeated queries
• Reduced API costs
• Local Docker deployment

---

🐳 Dockerized Infrastructure

Application runs in isolated containers (managed by docker-compose.yml).

Services:

• Next.js App
• Redis Cache
• Nginx Reverse Proxy

---

🔄 CI/CD Pipeline

GitHub Actions automatically:

• Install dependencies
• Run ESLint checks
• Run TypeScript type checking
• Build Next.js application
• Build Docker image (infrawatch-ai)

---

📈 Infrastructure Monitoring

Python background workers track:

• CPU Usage
• RAM (Memory) Usage
• Disk Usage
• Container Health (App & Redis)

Metrics are pushed to Supabase every 30 seconds and displayed in real time on the frontend.

---

🧠 AI Log Analysis

Powered by Google Gemini API.

The autonomous Python worker (log_analyzer.py) automatically analyzes:

• System Errors and Application Logs

Generates:

• Root Cause Analysis
• Severity Classification
• Suggested Fixes
• Confidence Score

Example Output:

{
  "root_cause": "Redis Container Unavailable",
  "severity": "High",
  "suggested_fix": "Restart Redis Container",
  "confidence": 91
}

---

📊 DevOps Dashboard

Route:

/monitoring

Features:

• Live CPU/RAM/Disk Gauges
• Container Uptime Status
• AI Incident Reports
• 30-Day API Token/Cost Metrics

---

🛠 Tech Stack

Frontend & Backend

• Next.js 16 (App Router)
• React 19
• TypeScript
• Tailwind CSS v4
• Framer Motion & GSAP

Authentication

• Clerk

Database & ORM

• Supabase PostgreSQL
• Prisma ORM

DevOps

• Docker & Docker Compose
• Nginx
• GitHub Actions
• WSL Ubuntu

Background Services

• Python (psutil)

AI APIs

• Gemini API (Google)
• OpenRouter API (OpenAI, Anthropic, Mistral)

---

📂 Project Structure

infra-watch-ai/
├── app/
│   ├── api/          # Next.js API Routes (Chat, Health, Monitoring)
│   ├── monitoring/   # InfraWatch Dashboard UI
│   └── page.tsx      # Main Multi-LLM Chat UI
│
├── components/       # React Components (Modals, Chat Inputs)
├── lib/              # Prisma Client & Redis Configuration
│
├── services/
│   └── monitoring/
│       ├── monitor.py           # Infrastructure metric collector
│       ├── log_analyzer.py      # Gemini-powered incident analyzer
│       ├── cpu.py, ram.py, disk.py, docker_status.py
│       └── requirements.txt
│
├── nginx/
│   └── infrawatch.conf          # Nginx Reverse Proxy Configuration
│
├── .github/
│   └── workflows/
│       └── ci.yml               # GitHub Actions CI/CD
│
├── supabase/
│   └── schema.sql               # PostgreSQL Schema & RLS Policies
│
├── prisma/
│   └── schema.prisma            # Prisma Models
│
├── docker-compose.yml           # App & Redis Containers
├── Dockerfile                   # Next.js Production Build
└── README.md

---

⚙️ Installation

Prerequisites

• Node.js 20+
• Docker Desktop
• Python 3.10+
• Supabase Project
• Clerk Account
• Gemini API Key

---

Clone Repository

git clone https://github.com/Darshan3690/DevOps_Copilot.git
cd DevOps_Copilot

---

Environment Variables

Create .env.local

# Clerk Auth
NEXT_PUBLIC_CLERK_PUBLISHABLE_KEY=
CLERK_SECRET_KEY=

# Prisma / Supabase
DATABASE_URL=
DIRECT_DATABASE_URL=

# AI APIs
GEMINI_API_KEY=
# Add your OpenRouter/OpenAI/Claude keys for the multi-llm chat

# Redis
REDIS_URL=redis://localhost:6379

---

Local Development (Frontend)

npm install
npx prisma generate
npm run dev

Open http://localhost:3000

---

Run Python Monitoring Services

cd services/monitoring
pip install -r requirements.txt
python monitor.py

---

🐳 Run With Docker

Build & Start the containers:

docker compose up --build -d

The app will be available at http://localhost:3000 via Nginx/Docker.

---

🔄 CI/CD Workflow

Push Code
     │
     ▼
GitHub Actions
     │
     ▼
Lint (ESLint)
     │
     ▼
Type Check (tsc)
     │
     ▼
Build (Next.js)
     │
     ▼
Docker Build
     │
     ▼
Pipeline Complete ✅

---

🚀 Future Enhancements

• Smart Email/Slack Alerts: Real-time notifications for critical system failures.
• Kubernetes Deployment Manifests
• Grafana Dashboard Integration
• AWS Terraform Deployment Scripts
• AI Auto-Remediation

---

🤝 Contributing

Contributions are welcome.

1. Fork Repository
2. Create Feature Branch (git checkout -b feature/amazing-feature)
3. Commit Changes (git commit -m "Add amazing feature")
4. Push (git push origin feature/amazing-feature)
5. Create Pull Request

---

📄 License

Distributed under the MIT License.

---

Made with ❤️ by Darshan Rajput

Building the future of AI-powered DevOps 🚀