A cloud-native RAG platform delivering trustworthy, citation-backed answers with dual-agent verification, academic research discovery, and automatic LLM failover.

VeriRAG is an intelligent document library system that ingests PDFs and answers user questions with verified, citation-backed responses. The system scores each AI-generated answer for faithfulness and automatically regenerates responses using a backup LLM if verification fails. Includes dedicated academic paper discovery for PhD research with integration to Semantic Scholar, arXiv, and CrossRef.

• ✅ Verified AI Responses — Every answer is scored for faithfulness against source documents
• ✅ Automatic Failover — Regenerates responses using backup LLM if verification fails
• ✅ Citation Grounding — Provides source references for every answer
• ✅ Academic Research Discovery — Native integration with Semantic Scholar, arXiv, CrossRef, and Google Scholar for PhD research
• ✅ OAuth Integration — Google & GitHub authentication with social sign-in support
• ✅ Cost Optimized — Built for a $97/month budget on Azure
• ✅ Production Ready — Cloud-native architecture for Azure Container Apps
• ✅ Observable — Real-time metrics with Prometheus, Grafana, and Azure Monitor
• ✅ Enhanced UX — Command palette, intelligent search, and streamlined interface

• Docker & Docker Compose
• Google API Key (Gemini + Embeddings) — Get here
• Groq API Key (optional fallback) — Get here
• Node.js 18+ (frontend)
• (Optional) Google & GitHub OAuth credentials for social sign-in

1. Clone the repository:

   git clone https://github.com/VaibhavKumar2005/cloud-native-ai-library-system.git
   cd cloud-native-ai-library-system

2. Create .env file:

   DJANGO_SECRET_KEY=your-secret-key
   GOOGLE_API_KEY=your-google-key
   GROQ_API_KEY=your-groq-key
   POSTGRES_USER=admin
   POSTGRES_PASSWORD=devpassword
   POSTGRES_DB=verirag_db
   POSTGRES_HOST=rag-db
   POSTGRES_PORT=5432
   REDIS_URL=redis://rag-redis:6379/0
   DEBUG=True
   ALLOWED_HOSTS=localhost,127.0.0.1,0.0.0.0,backend
   
   # OAuth (optional)
   GOOGLE_CLIENT_ID=your-google-client-id
   GOOGLE_CLIENT_SECRET=your-google-secret
   GITHUB_CLIENT_ID=your-github-client-id
   GITHUB_CLIENT_SECRET=your-github-secret

3. Start infrastructure:

   docker-compose up -d --build

4. Run migrations:

   docker exec -it rag-backend python manage.py migrate
   docker exec -it rag-backend python manage.py createsuperuser

5. Start frontend:

   cd frontend
   npm install
   npm run dev

See Google & GitHub OAuth Setup Guide for complete authentication configuration.

cloud-native-ai-library-system/
├── apps/
│   ├── backend/                    # Django REST API
│   │   ├── ai_engine/              # RAG engine + verification
│   │   │   ├── rag_logic.py        # Core dual-agent pipeline
│   │   │   ├── academic_views.py   # Academic paper discovery & RAG
│   │   │   ├── auth_views.py       # OAuth & authentication endpoints
│   │   │   ├── views.py            # Core API endpoints
│   │   │   ├── models.py           # Database models (Papers, Users, etc.)
│   │   │   ├── tasks.py            # Celery async tasks
│   │   │   └── rag_logic.py        # Academic RAG pipeline
│   │   ├── mcp_server.py           # FastAPI RAG backend (optional)
│   │   └── requirements.txt
│   └── frontend/                   # React + Vite app
│       ├── src/
│       │   └── components/         # Command palette, search, auth UI
│       └── package.json
├── ops/
│   ├── infrastructure/             # Terraform (Azure)
│   └── k8s/                        # Kubernetes manifests
├── scripts/
│   ├── demo/                       # Demo scripts
│   ├── setup/                      # Setup & initialization scripts
│   └── testing/                    # Test scripts
├── docs/
│   ├── guides/                     # Deployment guides
│   ├── OAUTH_SETUP_GUIDE.md        # Google & GitHub OAuth setup
│   ├── ARCHITECTURE.md             # System design details
│   └── README.md                   # Detailed documentation
├── docker-compose.yml              # Local dev orchestration
└── README.md                        # This file

Query your PDF library with verified, citation-backed answers:

curl -X POST http://localhost:8000/api/documents/ask/ \
  -H "Content-Type: application/json" \
  -d '{"question": "What are the key findings?"}'

Search and analyze academic papers across Semantic Scholar, arXiv, CrossRef:

curl -X POST http://localhost:8000/api/papers/search/ \
  -H "Content-Type: application/json" \
  -d '{"query": "machine learning", "source": "arxiv"}'

Features:

• 🔍 Paper Search — Query across multiple academic sources
• 📚 Personal Library — Organize and manage research papers
• 🔬 Gap Analysis — Identify research gaps and opportunities
• 💡 Topic Recommendations — Get smart topic suggestions based on your research
• ❓ RAG Q&A — Ask questions about papers with citation references

• Passwordless Email Login — Sign in with email
• OAuth 2.0 Social Sign-In — Google & GitHub authentication
• Secure Token Exchange — Industry-standard OAuth flow

cd apps/backend
pytest tests/ -v --tb=short
pytest tests/ --cov=ai_engine --cov-report=html

# Apply manifests
kubectl apply -k ops/k8s/

# Verify pods
kubectl get pods -n verirag

# View logs
kubectl logs -n verirag deployment/rag-backend --tail=50

Detailed documentation available in the docs/ directory:

• Deployment Guides — docs/guides/ for ACA, Kubernetes, and local setup
• Architecture Details — See docs/guides/ for system design documentation
• API Reference — See backend Swagger docs at /api/schema/swagger-ui/
• OAuth Setup — docs/OAUTH_SETUP_GUIDE.md

• ✅ Trivy vulnerability scanning & remediation in CI/CD
• ✅ Latest patched base images (Python 3.12, Alpine 3.21)
• ✅ OIDC authentication for Azure Container Registry
• ✅ Consolidated CI/CD workflows (10 → 4 lean pipelines)

• ✅ Academic RAG system with multiple source integrations
• ✅ FastAPI backend alternative for high-performance scenarios
• ✅ Enhanced frontend UX with command palette
• ✅ Optimized local development environment
• ✅ Comprehensive budget & deployment guides

• ✅ OAuth setup guide for Google & GitHub
• ✅ Terraform budget optimization guide
• ✅ Local dev environment guide
• ✅ Cleaned up and consolidated documentation

1. Fork the repository
2. Create a feature branch: git checkout -b feature/your-feature
3. Commit changes: git commit -am "Add feature"
4. Push to branch: git push origin feature/your-feature
5. Open a pull request

MIT License — see LICENSE for details

**Built for the Azure Cloud-Native Hackathon | Team 96