VectorWorkspaces

Multi-tenant, API-first knowledge platform with complete tenant isolation, versioned storage, and async processing. Features a RAG-powered query interface and modern React frontend.

Why This Exists

Organizations need a way to:

• Query internal documentation and knowledge bases
• Keep data private (no external API calls)
• Support multiple teams/tenants securely with workspace collaboration
• Scale to handle high-volume queries
• Track document versions and freshness

This platform provides all of that, ready to deploy.

Features

• Multi-tenant Architecture - Complete tenant isolation at database, vector storage, and API layers
• Workspace Management - Team collaboration with invite codes, member roles (Admin/Member), and RBAC
• API-first Design - RESTful API with service-layer architecture
• Async Processing - Celery + Redis for high-volume task orchestration
• Versioned Storage - Document versioning with freshness-aware retrieval
• Database Engineering - PostgreSQL with Alembic migrations and proper schema design
• RAG Query Interface - FAISS vector search + local Hugging Face LLM
• Modern Frontend - React + Tailwind CSS with routing and state management
• Production Observability - Prometheus metrics + structured logging
• Cloud-Ready Deployment - Dockerized with health checks and production configs

Quick Start

Docker (Recommended)

# Clone and start
git clone <repository-url>
cd Enterprise-AI-Knowledge-Agent
docker-compose up -d

# Run migrations
docker-compose exec api alembic upgrade head

# Access services
# API Docs: http://localhost:8000/docs
# Frontend: http://localhost:8080 (if running frontend)

Local Development

Backend:

cd backend
python -m venv venv
source venv/bin/activate  # Windows: venv\Scripts\activate
pip install -r requirements.txt

# Start PostgreSQL & Redis
docker-compose up -d postgres redis

# Run migrations
alembic upgrade head

# Start API
uvicorn app.main:app --reload

# Start Celery worker (separate terminal)
celery -A celery_worker worker --loglevel=info

Frontend:

cd frontend
npm install
npm run dev
# Frontend runs on http://localhost:8080

See docs/GETTING_STARTED.md for detailed setup.

Usage

1. Register & Login

POST /api/v1/auth/register  # Creates user + workspace (tenant)
POST /api/v1/auth/login     # Get JWT token

2. Workspace Management

GET  /api/v1/workspace/info              # Get workspace info
POST /api/v1/workspace/invite-code       # Generate invite code (Admin)
POST /api/v1/workspace/join              # Join workspace (creates account if needed)
GET  /api/v1/workspace/members           # List workspace members
PATCH /api/v1/workspace/members/{id}/role # Update member role (Admin)
DELETE /api/v1/workspace/members/{id}   # Remove member (Admin)

3. Ingest Knowledge

POST /api/v1/ingestion/upload    # Upload PDF/TXT/MD
POST /api/v1/ingestion/ingest    # API-based ingestion

4. Query Knowledge

POST /api/v1/chat/query  # RAG-powered answers

API Docs: http://localhost:8000/docs (when running)

Tech Stack

Backend:

• FastAPI, PostgreSQL, FAISS, Redis, Celery
• Hugging Face (transformers, sentence-transformers)
• Alembic, SQLAlchemy, Pydantic

Frontend:

• React 18, React Router, Axios
• Tailwind CSS, Vite

Infrastructure:

• Docker, Docker Compose
• Prometheus, Grafana

Project Structure

.
├── backend/
│   ├── app/
│   │   ├── api/v1/       # API endpoints (auth, workspace, ingestion, chat, etc.)
│   │   ├── core/         # Security, config, logging
│   │   ├── db/           # Database models & session
│   │   ├── services/     # Business logic
│   │   ├── tasks/        # Celery async tasks
│   │   └── schemas/      # Pydantic models
│   ├── alembic/          # Database migrations
│   └── requirements.txt
├── frontend/
│   ├── src/
│   │   ├── pages/        # React pages (Login, Dashboard, Documents, Query, Workspace)
│   │   ├── components/   # Reusable components
│   │   └── contexts/     # React context (Auth)
│   └── package.json
├── monitoring/            # Prometheus & Grafana configs
├── storage/              # FAISS indices & uploaded documents
└── docker-compose.yml

Configuration

Create backend/.env file and configure:

• DATABASE_URL - PostgreSQL connection string (default: postgresql://postgres:postgres@localhost:5432/knowledge_agent)
• JWT_SECRET_KEY - Change this in production! (default: dev-secret-key-change-in-production)
• EMBEDDING_MODEL - Hugging Face embedding model (default: sentence-transformers/all-MiniLM-L6-v2)
• LLM_MODEL - Hugging Face LLM model (default: facebook/opt-1.3b)
• REDIS_URL - Redis connection string (default: redis://localhost:6379/0)
• CELERY_BROKER_URL - Celery broker URL (default: redis://localhost:6379/0)
• CELERY_RESULT_BACKEND - Celery result backend (default: redis://localhost:6379/0)

See docs/GETTING_STARTED.md for complete configuration details.

Monitoring

docker-compose -f docker-compose.monitoring.yml up -d

• Prometheus: http://localhost:9090
• Grafana: http://localhost:3000 (admin/admin)

Deployment

Local Development: docker-compose up -d

Production/Cloud: This platform is cloud-ready and can be deployed to:

• AWS (ECS, EC2, EKS)
• Google Cloud (Cloud Run, GKE)
• Azure (Container Instances, AKS)
• Any Kubernetes cluster
• Or any server with Docker

Documentation

• Getting Started - Setup guide
• Architecture - System design
• Developer Guide - How to work on this codebase
• AI Agent Guide - Guide for AI agents

Contributing

Contributions welcome! See CONTRIBUTING.md for guidelines.

License

MIT License - see LICENSE file