diff --git a/.gitignore b/.gitignore
new file mode 100644
index 0000000..07eca80
--- /dev/null
+++ b/.gitignore
@@ -0,0 +1,43 @@
+# Dependencies
+node_modules/
+__pycache__/
+*.py[cod]
+*$py.class
+.venv/
+venv/
+env/
+.env
+
+# Build outputs
+.next/
+out/
+build/
+dist/
+*.egg-info/
+
+# IDE
+.idea/
+.vscode/
+*.swp
+*.swo
+
+# Environment files
+.env
+.env.local
+.env.*.local
+
+# OS files
+.DS_Store
+Thumbs.db
+
+# Logs
+*.log
+npm-debug.log*
+
+# Testing
+coverage/
+.coverage
+htmlcov/
+
+# Misc
+*.pem
diff --git a/README.md b/README.md
index f919b49..a4d90b9 100644
--- a/README.md
+++ b/README.md
@@ -1,2 +1,268 @@
-# hack-nation template
+# 🚀 Hack Nation 2026 - Full-Stack Template
+
+A comprehensive hackathon template with **Next.js**, **FastAPI**, **Supabase**, and **AI integration**.
+
+## 📁 Project Structure
+
+```
+hack-nation/
+├── frontend/                 # Next.js 14 + React
+│   ├── src/
+│   │   ├── app/             # App Router pages
+│   │   │   ├── layout.tsx   # Root layout
+│   │   │   ├── page.tsx     # Home page
+│   │   │   ├── login/       # Auth page
+│   │   │   ├── dashboard/   # Protected page
+│   │   │   └── demo/        # Feature demos
+│   │   ├── components/      # React components
+│   │   │   ├── ui/          # Shadcn/ui components
+│   │   │   └── ...          # Custom components
+│   │   └── lib/             # Utilities & services
+│   │       ├── supabase/    # Supabase client & provider
+│   │       └── api.ts       # Backend API client
+│   └── package.json
+│
+├── backend/                  # FastAPI (Python)
+│   ├── main.py              # FastAPI app entry
+│   ├── config.py            # Environment configuration
+│   ├── routers/             # API route handlers
+│   │   ├── ai.py            # AI generation endpoints
+│   │   ├── projects.py      # CRUD example
+│   │   └── health.py        # Health checks
+│   └── services/            # Business logic
+│       ├── ai_service.py    # AI provider integration
+│       └── database.py      # Supabase helpers
+│
+└── README.md
+```
+
+## 🛠 Tech Stack
+
+| Layer | Technology | Why? |
+|-------|------------|------|
+| **Frontend** | Next.js 14 | React framework with App Router, SSR, and great DX |
+| **UI Components** | Shadcn/ui | Beautiful, accessible, customizable components |
+| **Styling** | Tailwind CSS | Utility-first CSS, rapid development |
+| **Backend** | FastAPI | Fast Python API, auto docs, async support |
+| **Database** | Supabase | Postgres + Auth + Realtime, generous free tier |
+| **AI** | OpenAI/Anthropic | GPT-4 or Claude for AI features |
+
+## 🚀 Quick Start
+
+### 1. Clone & Install
+
+```bash
+# Clone the repo (or use this template)
+cd hack-nation
+
+# Install frontend dependencies
+cd frontend
+npm install
+
+# Install backend dependencies
+cd ../backend
+pip install -r requirements.txt
+```
+
+### 2. Configure Environment
+
+```bash
+# Frontend
+cd frontend
+cp .env.local.example .env.local
+# Edit .env.local with your Supabase keys
+
+# Backend
+cd ../backend
+cp .env.example .env
+# Edit .env with your API keys
+```
+
+### 3. Run Development Servers
+
+```bash
+# Terminal 1: Frontend (localhost:3000)
+cd frontend
+npm run dev
+
+# Terminal 2: Backend (localhost:8000)
+cd backend
+python run.py
+```
+
+### 4. View Your App
+
+- **Frontend**: http://localhost:3000
+- **Backend API**: http://localhost:8000
+- **API Docs**: http://localhost:8000/docs
+
+## ⚙️ Configuration
+
+### Supabase Setup
+
+1. Create a project at [supabase.com](https://supabase.com)
+2. Go to Settings → API
+3. Copy the **URL** and **anon key** to frontend `.env.local`
+4. Copy the **URL** and **service role key** to backend `.env`
+
+### AI Setup
+
+1. Get an API key from [OpenAI](https://platform.openai.com) or [Anthropic](https://anthropic.com)
+2. Add to backend `.env`:
+   ```
+   OPENAI_API_KEY=sk-your-key-here
+   AI_PROVIDER=openai
+   AI_MODEL=gpt-3.5-turbo
+   ```
+
+## 📚 Key Concepts
+
+### Frontend Architecture
+
+```
+┌─────────────────────────────────────────┐
+│              layout.tsx                  │
+│  ┌────────────────────────────────────┐ │
+│  │           Providers                 │ │
+│  │  ┌──────────────────────────────┐  │ │
+│  │  │       Navigation             │  │ │
+│  │  ├──────────────────────────────┤  │ │
+│  │  │       page.tsx               │  │ │
+│  │  │   (Route-specific content)   │  │ │
+│  │  └──────────────────────────────┘  │ │
+│  └────────────────────────────────────┘ │
+└─────────────────────────────────────────┘
+```
+
+- **layout.tsx**: Wraps all pages, contains providers
+- **page.tsx**: Individual page content
+- **Providers**: Context for auth, theme, etc.
+
+### Backend Architecture
+
+```
+Request → Router → Service → Response
+           │          │
+           │          └── AI Service / Database
+           │
+           └── Validates input with Pydantic
+```
+
+- **Routers**: Handle HTTP requests, validate input
+- **Services**: Business logic, external integrations
+- **Config**: All settings from environment variables
+
+### Data Flow
+
+```
+┌──────────┐     ┌──────────┐     ┌──────────┐
+│ Frontend │────▶│ Backend  │────▶│   AI     │
+│ (Next.js)│◀────│ (FastAPI)│◀────│ Provider │
+└──────────┘     └──────────┘     └──────────┘
+      │                │
+      │                │
+      ▼                ▼
+┌─────────────────────────────────────────┐
+│              Supabase                    │
+│  (Database, Auth, Realtime, Storage)    │
+└─────────────────────────────────────────┘
+```
+
+## 🎨 Adding Shadcn/ui Components
+
+```bash
+# Initialize shadcn (first time only)
+cd frontend
+npx shadcn-ui@latest init
+
+# Add components as needed
+npx shadcn-ui@latest add button
+npx shadcn-ui@latest add card
+npx shadcn-ui@latest add dialog
+npx shadcn-ui@latest add dropdown-menu
+```
+
+## 🔌 API Endpoints
+
+### AI Endpoints
+
+| Method | Endpoint | Description |
+|--------|----------|-------------|
+| POST | `/api/ai/generate` | Generate text from prompt |
+| POST | `/api/ai/chat` | Chat with message history |
+| POST | `/api/ai/stream` | Stream AI response |
+| GET | `/api/ai/models` | List available models |
+
+### Project Endpoints (Example CRUD)
+
+| Method | Endpoint | Description |
+|--------|----------|-------------|
+| GET | `/api/projects` | List all projects |
+| POST | `/api/projects` | Create project |
+| GET | `/api/projects/:id` | Get project |
+| PATCH | `/api/projects/:id` | Update project |
+| DELETE | `/api/projects/:id` | Delete project |
+
+## 🗄 Database Schema (Supabase)
+
+Run these SQL commands in Supabase SQL Editor:
+
+```sql
+-- Projects table
+CREATE TABLE projects (
+  id UUID DEFAULT gen_random_uuid() PRIMARY KEY,
+  name TEXT NOT NULL,
+  description TEXT DEFAULT '',
+  user_id UUID REFERENCES auth.users(id),
+  created_at TIMESTAMP WITH TIME ZONE DEFAULT NOW()
+);
+
+-- Enable Row Level Security
+ALTER TABLE projects ENABLE ROW LEVEL SECURITY;
+
+-- Users can only access their own projects
+CREATE POLICY "Users can CRUD own projects"
+  ON projects FOR ALL
+  USING (auth.uid() = user_id);
+```
+
+## 🚢 Deployment
+
+### Frontend (Vercel)
+
+1. Push to GitHub
+2. Import project in [Vercel](https://vercel.com)
+3. Add environment variables
+4. Deploy!
+
+### Backend (Railway/Render)
+
+1. Push to GitHub
+2. Import in [Railway](https://railway.app) or [Render](https://render.com)
+3. Set environment variables
+4. Deploy!
+
+### Supabase
+
+Already hosted! Just keep your project running.
+
+## 💡 Hackathon Tips
+
+1. **Start with the AI demo** - It already works!
+2. **Use Shadcn/ui components** - Don't build from scratch
+3. **Leverage Supabase** - Auth and DB are ready
+4. **Keep it simple** - One cool feature > many half-done ones
+5. **Test the happy path** - Edge cases can wait
+
+## 📖 Learn More
+
+- [Next.js Docs](https://nextjs.org/docs)
+- [FastAPI Docs](https://fastapi.tiangolo.com)
+- [Supabase Docs](https://supabase.com/docs)
+- [Shadcn/ui Components](https://ui.shadcn.com)
+- [OpenAI API](https://platform.openai.com/docs)
+
+---
+
+Built for **Hack Nation 2026** 🏆 template
 
diff --git a/backend/.env.example b/backend/.env.example
new file mode 100644
index 0000000..3cae430
--- /dev/null
+++ b/backend/.env.example
@@ -0,0 +1,43 @@
+# =============================================================================
+# ENVIRONMENT VARIABLES
+# =============================================================================
+# Copy this file to .env and fill in your values
+# NEVER commit .env to version control!
+
+# -----------------------------------------------------------------------------
+# SERVER CONFIGURATION
+# -----------------------------------------------------------------------------
+PORT=8000
+ENVIRONMENT=development
+
+# -----------------------------------------------------------------------------
+# SUPABASE CONFIGURATION
+# -----------------------------------------------------------------------------
+# Get these from: https://app.supabase.com/project/_/settings/api
+SUPABASE_URL=https://your-project.supabase.co
+SUPABASE_SERVICE_KEY=your-service-role-key-here
+# Note: Use SERVICE KEY (not anon key) for backend - has admin access
+
+# -----------------------------------------------------------------------------
+# AI PROVIDER CONFIGURATION
+# -----------------------------------------------------------------------------
+# Choose your AI provider and add the appropriate key
+
+# OpenAI (GPT-4, GPT-3.5)
+OPENAI_API_KEY=sk-your-openai-key-here
+
+# Anthropic (Claude)
+# ANTHROPIC_API_KEY=sk-ant-your-anthropic-key-here
+
+# Default AI provider to use: "openai" or "anthropic"
+AI_PROVIDER=openai
+
+# Default model to use
+AI_MODEL=gpt-3.5-turbo
+# Other options: gpt-4, gpt-4-turbo, claude-3-opus-20240229, claude-3-sonnet-20240229
+
+# -----------------------------------------------------------------------------
+# CORS CONFIGURATION
+# -----------------------------------------------------------------------------
+# Comma-separated list of allowed origins
+CORS_ORIGINS=http://localhost:3000,http://127.0.0.1:3000
diff --git a/backend/config.py b/backend/config.py
new file mode 100644
index 0000000..d0b1aca
--- /dev/null
+++ b/backend/config.py
@@ -0,0 +1,83 @@
+"""
+=============================================================================
+CONFIGURATION MODULE
+=============================================================================
+
+This module handles all configuration using Pydantic Settings.
+
+Benefits of using Pydantic Settings:
+- Type validation for environment variables
+- Auto-loading from .env files
+- IDE autocomplete and type hints
+- Clear documentation of all required config
+
+Usage:
+    from config import settings
+    print(settings.OPENAI_API_KEY)
+"""
+
+from pydantic_settings import BaseSettings
+from functools import lru_cache
+
+
+class Settings(BaseSettings):
+    """
+    Application settings loaded from environment variables.
+    
+    All settings can be overridden via:
+    1. Environment variables
+    2. .env file in the backend directory
+    """
+    
+    # -------------------------------------------------------------------------
+    # Server Configuration
+    # -------------------------------------------------------------------------
+    PORT: int = 8000
+    ENVIRONMENT: str = "development"  # "development", "staging", "production"
+    
+    # -------------------------------------------------------------------------
+    # Supabase Configuration
+    # -------------------------------------------------------------------------
+    SUPABASE_URL: str = ""
+    SUPABASE_SERVICE_KEY: str = ""  # Service key for admin access
+    
+    # -------------------------------------------------------------------------
+    # AI Configuration
+    # -------------------------------------------------------------------------
+    OPENAI_API_KEY: str = ""
+    ANTHROPIC_API_KEY: str = ""
+    AI_PROVIDER: str = "openai"  # "openai" or "anthropic"
+    AI_MODEL: str = "gpt-3.5-turbo"
+    
+    # -------------------------------------------------------------------------
+    # CORS Configuration
+    # -------------------------------------------------------------------------
+    CORS_ORIGINS: str = "http://localhost:3000"  # Comma-separated list
+    
+    @property
+    def cors_origins_list(self) -> list[str]:
+        """Parse CORS_ORIGINS string into a list."""
+        return [origin.strip() for origin in self.CORS_ORIGINS.split(",")]
+    
+    class Config:
+        # Load from .env file
+        env_file = ".env"
+        env_file_encoding = "utf-8"
+        # Make field names case-insensitive
+        case_sensitive = False
+
+
+# Use lru_cache to create a singleton settings instance
+@lru_cache
+def get_settings() -> Settings:
+    """
+    Get cached settings instance.
+    
+    Using lru_cache ensures we only read the .env file once,
+    improving performance.
+    """
+    return Settings()
+
+
+# Export settings instance for easy import
+settings = get_settings()
diff --git a/backend/main.py b/backend/main.py
new file mode 100644
index 0000000..1884528
--- /dev/null
+++ b/backend/main.py
@@ -0,0 +1,130 @@
+"""
+=============================================================================
+FASTAPI MAIN APPLICATION
+=============================================================================
+
+This is the entry point for the FastAPI backend.
+
+FastAPI Features:
+- Automatic API documentation (Swagger UI at /docs)
+- Request validation with Pydantic
+- Async support for high performance
+- Easy dependency injection
+
+To run the server:
+    uvicorn main:app --reload --port 8000
+
+Or use the provided run script:
+    python run.py
+"""
+
+from fastapi import FastAPI
+from fastapi.middleware.cors import CORSMiddleware
+from contextlib import asynccontextmanager
+
+# Import configuration
+from config import settings
+
+# Import routers (API endpoints grouped by feature)
+from routers import ai, projects, health
+
+
+# =============================================================================
+# APPLICATION LIFECYCLE
+# =============================================================================
+
+@asynccontextmanager
+async def lifespan(app: FastAPI):
+    """
+    Manage application lifecycle events.
+    
+    This runs code on startup (before yield) and shutdown (after yield).
+    Great for:
+    - Initializing database connections
+    - Loading ML models
+    - Setting up caches
+    """
+    # Startup
+    print("🚀 Starting Hack Nation Backend...")
+    print(f"📍 Environment: {settings.ENVIRONMENT}")
+    print(f"🤖 AI Provider: {settings.AI_PROVIDER} ({settings.AI_MODEL})")
+    
+    yield  # Application runs here
+    
+    # Shutdown
+    print("👋 Shutting down...")
+
+
+# =============================================================================
+# CREATE APPLICATION
+# =============================================================================
+
+app = FastAPI(
+    title="Hack Nation API",
+    description="""
+    Backend API for Hack Nation hackathon template.
+    
+    Features:
+    - AI text generation (OpenAI / Anthropic)
+    - Supabase database integration
+    - User authentication
+    """,
+    version="1.0.0",
+    lifespan=lifespan,
+)
+
+
+# =============================================================================
+# MIDDLEWARE
+# =============================================================================
+
+# CORS - Allow frontend to call API
+app.add_middleware(
+    CORSMiddleware,
+    allow_origins=settings.cors_origins_list,  # Origins allowed to call API
+    allow_credentials=True,                     # Allow cookies/auth headers
+    allow_methods=["*"],                        # Allow all HTTP methods
+    allow_headers=["*"],                        # Allow all headers
+)
+
+
+# =============================================================================
+# ROUTERS
+# =============================================================================
+
+# Include routers with prefixes
+# Each router handles a specific feature area
+app.include_router(
+    health.router,
+    prefix="/api",
+    tags=["Health"],
+)
+
+app.include_router(
+    ai.router,
+    prefix="/api/ai",
+    tags=["AI"],
+)
+
+app.include_router(
+    projects.router,
+    prefix="/api/projects",
+    tags=["Projects"],
+)
+
+
+# =============================================================================
+# ROOT ENDPOINT
+# =============================================================================
+
+@app.get("/")
+async def root():
+    """
+    Root endpoint - useful for health checks and API info.
+    """
+    return {
+        "name": "Hack Nation API",
+        "version": "1.0.0",
+        "docs": "/docs",
+        "health": "/api/health",
+    }
diff --git a/backend/requirements.txt b/backend/requirements.txt
new file mode 100644
index 0000000..29701ad
--- /dev/null
+++ b/backend/requirements.txt
@@ -0,0 +1,22 @@
+# =============================================================================
+# FASTAPI BACKEND - REQUIREMENTS
+# =============================================================================
+# Install with: pip install -r requirements.txt
+# Or: pip install fastapi uvicorn openai python-dotenv supabase pydantic
+
+# Web Framework
+fastapi>=0.115.0
+uvicorn[standard]>=0.30.0
+
+# AI Integration
+openai>=1.50.0                    # OpenAI API client
+anthropic>=0.39.0                 # Anthropic Claude API (alternative)
+
+# Database (Supabase)
+supabase>=2.10.0                  # Supabase Python client
+
+# Utilities
+python-dotenv>=1.0.0              # Load .env files
+pydantic>=2.9.0                   # Data validation
+pydantic-settings>=2.5.0          # Settings management
+httpx>=0.27.0                     # Async HTTP client
diff --git a/backend/routers/__init__.py b/backend/routers/__init__.py
new file mode 100644
index 0000000..6b5dfc7
--- /dev/null
+++ b/backend/routers/__init__.py
@@ -0,0 +1,10 @@
+"""
+Routers Package
+
+This package contains all API route handlers organized by feature.
+Each router is responsible for a specific domain of the application.
+"""
+
+from . import ai, projects, health
+
+__all__ = ["ai", "projects", "health"]
diff --git a/backend/routers/ai.py b/backend/routers/ai.py
new file mode 100644
index 0000000..e8748db
--- /dev/null
+++ b/backend/routers/ai.py
@@ -0,0 +1,171 @@
+"""
+=============================================================================
+AI ROUTER
+=============================================================================
+
+This router handles all AI-related endpoints.
+
+Features:
+- Text generation with OpenAI or Anthropic
+- Streaming responses for real-time output
+- Multiple AI models support
+
+Architecture:
+- Endpoints are thin - they just validate input and return output
+- Business logic is in the services module
+- This separation makes testing easier
+"""
+
+from fastapi import APIRouter, HTTPException
+from fastapi.responses import StreamingResponse
+from pydantic import BaseModel, Field
+from typing import Optional
+
+# Import our AI service
+from services.ai_service import generate_text, stream_text, chat_completion
+
+router = APIRouter()
+
+
+# =============================================================================
+# REQUEST/RESPONSE MODELS
+# =============================================================================
+
+class GenerateRequest(BaseModel):
+    """Request body for text generation."""
+    prompt: str = Field(..., min_length=1, max_length=10000)
+    max_tokens: Optional[int] = Field(default=500, ge=1, le=4000)
+    temperature: Optional[float] = Field(default=0.7, ge=0, le=2)
+
+
+class GenerateResponse(BaseModel):
+    """Response from text generation."""
+    text: str
+    model: str
+    usage: Optional[dict] = None
+
+
+class ChatMessage(BaseModel):
+    """A single message in a chat conversation."""
+    role: str = Field(..., pattern="^(user|assistant|system)$")
+    content: str = Field(..., min_length=1)
+
+
+class ChatRequest(BaseModel):
+    """Request body for chat completion."""
+    messages: list[ChatMessage] = Field(..., min_items=1)
+    max_tokens: Optional[int] = Field(default=500, ge=1, le=4000)
+    temperature: Optional[float] = Field(default=0.7, ge=0, le=2)
+
+
+class ChatResponse(BaseModel):
+    """Response from chat completion."""
+    message: ChatMessage
+    model: str
+
+
+# =============================================================================
+# ENDPOINTS
+# =============================================================================
+
+@router.post("/generate", response_model=GenerateResponse)
+async def generate_endpoint(request: GenerateRequest):
+    """
+    Generate text from a prompt.
+    
+    This is a simple completion endpoint - give it a prompt,
+    get back generated text.
+    
+    Example:
+        POST /api/ai/generate
+        {"prompt": "Write a haiku about coding"}
+    """
+    try:
+        result = await generate_text(
+            prompt=request.prompt,
+            max_tokens=request.max_tokens,
+            temperature=request.temperature,
+        )
+        return GenerateResponse(**result)
+    except Exception as e:
+        raise HTTPException(status_code=500, detail=str(e))
+
+
+@router.post("/chat", response_model=ChatResponse)
+async def chat_endpoint(request: ChatRequest):
+    """
+    Chat completion with message history.
+    
+    Send a list of messages (conversation history) and get
+    the assistant's response.
+    
+    Example:
+        POST /api/ai/chat
+        {
+            "messages": [
+                {"role": "system", "content": "You are a helpful assistant."},
+                {"role": "user", "content": "Hello!"}
+            ]
+        }
+    """
+    try:
+        # Convert Pydantic models to dicts
+        messages = [m.model_dump() for m in request.messages]
+        
+        result = await chat_completion(
+            messages=messages,
+            max_tokens=request.max_tokens,
+            temperature=request.temperature,
+        )
+        return ChatResponse(
+            message=ChatMessage(**result["message"]),
+            model=result["model"],
+        )
+    except Exception as e:
+        raise HTTPException(status_code=500, detail=str(e))
+
+
+@router.post("/stream")
+async def stream_endpoint(request: GenerateRequest):
+    """
+    Stream generated text in real-time.
+    
+    Uses Server-Sent Events (SSE) to stream the response
+    token by token. Great for chat interfaces!
+    
+    The frontend can read this with:
+        const response = await fetch("/api/ai/stream", {...});
+        const reader = response.body.getReader();
+    """
+    try:
+        return StreamingResponse(
+            stream_text(
+                prompt=request.prompt,
+                max_tokens=request.max_tokens,
+                temperature=request.temperature,
+            ),
+            media_type="text/event-stream",
+        )
+    except Exception as e:
+        raise HTTPException(status_code=500, detail=str(e))
+
+
+@router.get("/models")
+async def list_models():
+    """
+    List available AI models.
+    
+    Useful for letting users choose which model to use.
+    """
+    return {
+        "openai": [
+            {"id": "gpt-4-turbo", "name": "GPT-4 Turbo", "description": "Most capable model"},
+            {"id": "gpt-4", "name": "GPT-4", "description": "High quality responses"},
+            {"id": "gpt-3.5-turbo", "name": "GPT-3.5 Turbo", "description": "Fast and efficient"},
+        ],
+        "anthropic": [
+            {"id": "claude-3-opus-20240229", "name": "Claude 3 Opus", "description": "Most capable"},
+            {"id": "claude-3-sonnet-20240229", "name": "Claude 3 Sonnet", "description": "Balanced"},
+            {"id": "claude-3-haiku-20240307", "name": "Claude 3 Haiku", "description": "Fast"},
+        ],
+    }
diff --git a/backend/routers/health.py b/backend/routers/health.py
new file mode 100644
index 0000000..f9230fc
--- /dev/null
+++ b/backend/routers/health.py
@@ -0,0 +1,62 @@
+"""
+=============================================================================
+HEALTH CHECK ROUTER
+=============================================================================
+
+Simple health check endpoints for monitoring and deployment.
+
+These are useful for:
+- Load balancer health checks
+- Kubernetes readiness probes
+- Monitoring systems
+"""
+
+from fastapi import APIRouter
+from config import settings
+
+router = APIRouter()
+
+
+@router.get("/health")
+async def health_check():
+    """
+    Basic health check endpoint.
+    Returns 200 if the server is running.
+    """
+    return {
+        "status": "healthy",
+        "environment": settings.ENVIRONMENT,
+    }
+
+
+@router.get("/health/detailed")
+async def detailed_health_check():
+    """
+    Detailed health check with service status.
+    Useful for debugging connection issues.
+    """
+    # Check AI provider configuration
+    ai_configured = bool(
+        settings.OPENAI_API_KEY or settings.ANTHROPIC_API_KEY
+    )
+    
+    # Check Supabase configuration
+    supabase_configured = bool(
+        settings.SUPABASE_URL and settings.SUPABASE_SERVICE_KEY
+    )
+    
+    return {
+        "status": "healthy",
+        "environment": settings.ENVIRONMENT,
+        "services": {
+            "ai": {
+                "configured": ai_configured,
+                "provider": settings.AI_PROVIDER,
+                "model": settings.AI_MODEL,
+            },
+            "database": {
+                "configured": supabase_configured,
+                "provider": "supabase",
+            },
+        },
+    }
diff --git a/backend/routers/projects.py b/backend/routers/projects.py
new file mode 100644
index 0000000..bbc9e00
--- /dev/null
+++ b/backend/routers/projects.py
@@ -0,0 +1,191 @@
+"""
+=============================================================================
+PROJECTS ROUTER
+=============================================================================
+
+Example CRUD endpoints for managing projects.
+
+This demonstrates:
+- Supabase database integration
+- CRUD operations (Create, Read, Update, Delete)
+- Request validation with Pydantic
+- Error handling patterns
+
+You can adapt this pattern for any resource in your app.
+"""
+
+from fastapi import APIRouter, HTTPException, Header
+from pydantic import BaseModel, Field
+from typing import Optional
+from datetime import datetime
+
+# Import Supabase client
+from services.database import get_supabase_client
+
+router = APIRouter()
+
+
+# =============================================================================
+# REQUEST/RESPONSE MODELS
+# =============================================================================
+
+class ProjectCreate(BaseModel):
+    """Data required to create a new project."""
+    name: str = Field(..., min_length=1, max_length=100)
+    description: Optional[str] = Field(default="", max_length=1000)
+
+
+class ProjectUpdate(BaseModel):
+    """Data that can be updated on a project."""
+    name: Optional[str] = Field(None, min_length=1, max_length=100)
+    description: Optional[str] = Field(None, max_length=1000)
+
+
+class Project(BaseModel):
+    """Full project model (database representation)."""
+    id: str
+    name: str
+    description: str
+    user_id: str
+    created_at: datetime
+
+
+# =============================================================================
+# ENDPOINTS
+# =============================================================================
+
+@router.get("")
+async def list_projects(
+    authorization: str = Header(None, description="Bearer token from Supabase")
+):
+    """
+    List all projects for the authenticated user.
+    
+    Requires authentication - pass the Supabase access token
+    in the Authorization header.
+    """
+    if not authorization:
+        raise HTTPException(status_code=401, detail="Authorization required")
+    
+    try:
+        # Get Supabase client
+        supabase = get_supabase_client()
+        
+        # Extract user from token (Supabase handles this)
+        # In production, you'd verify the JWT properly
+        
+        # Query projects
+        response = supabase.table("projects").select("*").execute()
+        
+        return response.data
+    except Exception as e:
+        raise HTTPException(status_code=500, detail=str(e))
+
+
+@router.post("", response_model=Project)
+async def create_project(
+    project: ProjectCreate,
+    authorization: str = Header(None),
+):
+    """
+    Create a new project.
+    
+    Example:
+        POST /api/projects
+        {"name": "My AI App", "description": "An awesome AI project"}
+    """
+    if not authorization:
+        raise HTTPException(status_code=401, detail="Authorization required")
+    
+    try:
+        supabase = get_supabase_client()
+        
+        # In production, extract user_id from JWT
+        # For demo, we'll use a placeholder
+        user_id = "demo-user-id"
+        
+        response = supabase.table("projects").insert({
+            "name": project.name,
+            "description": project.description or "",
+            "user_id": user_id,
+        }).execute()
+        
+        return response.data[0]
+    except Exception as e:
+        raise HTTPException(status_code=500, detail=str(e))
+
+
+@router.get("/{project_id}", response_model=Project)
+async def get_project(project_id: str):
+    """
+    Get a specific project by ID.
+    """
+    try:
+        supabase = get_supabase_client()
+        
+        response = supabase.table("projects").select("*").eq("id", project_id).single().execute()
+        
+        if not response.data:
+            raise HTTPException(status_code=404, detail="Project not found")
+        
+        return response.data
+    except HTTPException:
+        raise
+    except Exception as e:
+        raise HTTPException(status_code=500, detail=str(e))
+
+
+@router.patch("/{project_id}", response_model=Project)
+async def update_project(
+    project_id: str,
+    project: ProjectUpdate,
+    authorization: str = Header(None),
+):
+    """
+    Update a project.
+    
+    Only send the fields you want to update.
+    """
+    if not authorization:
+        raise HTTPException(status_code=401, detail="Authorization required")
+    
+    try:
+        supabase = get_supabase_client()
+        
+        # Build update dict with only provided fields
+        update_data = project.model_dump(exclude_unset=True)
+        
+        if not update_data:
+            raise HTTPException(status_code=400, detail="No fields to update")
+        
+        response = supabase.table("projects").update(update_data).eq("id", project_id).execute()
+        
+        if not response.data:
+            raise HTTPException(status_code=404, detail="Project not found")
+        
+        return response.data[0]
+    except HTTPException:
+        raise
+    except Exception as e:
+        raise HTTPException(status_code=500, detail=str(e))
+
+
+@router.delete("/{project_id}")
+async def delete_project(
+    project_id: str,
+    authorization: str = Header(None),
+):
+    """
+    Delete a project.
+    """
+    if not authorization:
+        raise HTTPException(status_code=401, detail="Authorization required")
+    
+    try:
+        supabase = get_supabase_client()
+        
+        response = supabase.table("projects").delete().eq("id", project_id).execute()
+        
+        return {"message": "Project deleted", "id": project_id}
+    except Exception as e:
+        raise HTTPException(status_code=500, detail=str(e))
diff --git a/backend/run.py b/backend/run.py
new file mode 100644
index 0000000..542a266
--- /dev/null
+++ b/backend/run.py
@@ -0,0 +1,23 @@
+"""
+=============================================================================
+RUN SCRIPT
+=============================================================================
+
+Simple script to run the FastAPI server.
+Useful for development and debugging.
+
+Usage:
+    python run.py
+"""
+
+import uvicorn
+from config import settings
+
+if __name__ == "__main__":
+    uvicorn.run(
+        "main:app",
+        host="0.0.0.0",
+        port=settings.PORT,
+        reload=settings.ENVIRONMENT == "development",  # Auto-reload in dev
+        log_level="info",
+    )
diff --git a/backend/services/__init__.py b/backend/services/__init__.py
new file mode 100644
index 0000000..ec5d406
--- /dev/null
+++ b/backend/services/__init__.py
@@ -0,0 +1,10 @@
+"""
+Services Package
+
+This package contains business logic and external service integrations.
+Keeping business logic separate from routes makes the code more testable.
+"""
+
+from . import ai_service, database
+
+__all__ = ["ai_service", "database"]
diff --git a/backend/services/ai_service.py b/backend/services/ai_service.py
new file mode 100644
index 0000000..17f5df0
--- /dev/null
+++ b/backend/services/ai_service.py
@@ -0,0 +1,250 @@
+"""
+=============================================================================
+AI SERVICE
+=============================================================================
+
+This module handles all AI provider integrations.
+
+Supported Providers:
+- OpenAI (GPT-4, GPT-3.5)
+- Anthropic (Claude)
+
+Architecture:
+- Provider-agnostic interface (same API regardless of provider)
+- Easy to add new providers
+- Streaming support for real-time responses
+
+The actual API keys and configuration come from config.py
+"""
+
+from typing import AsyncGenerator, Optional
+import openai
+from config import settings
+
+
+# =============================================================================
+# INITIALIZE CLIENTS
+# =============================================================================
+
+# Initialize OpenAI client
+openai_client = openai.AsyncOpenAI(api_key=settings.OPENAI_API_KEY)
+
+# For Anthropic, you would:
+# import anthropic
+# anthropic_client = anthropic.AsyncAnthropic(api_key=settings.ANTHROPIC_API_KEY)
+
+
+# =============================================================================
+# TEXT GENERATION
+# =============================================================================
+
+async def generate_text(
+    prompt: str,
+    max_tokens: int = 500,
+    temperature: float = 0.7,
+    model: Optional[str] = None,
+) -> dict:
+    """
+    Generate text from a prompt using the configured AI provider.
+    
+    Args:
+        prompt: The input text to generate from
+        max_tokens: Maximum tokens in the response
+        temperature: Creativity level (0 = deterministic, 2 = very creative)
+        model: Override the default model
+    
+    Returns:
+        dict with keys: text, model, usage
+    
+    Example:
+        result = await generate_text("Write a haiku about Python")
+        print(result["text"])
+    """
+    model = model or settings.AI_MODEL
+    
+    if settings.AI_PROVIDER == "openai":
+        return await _generate_openai(prompt, max_tokens, temperature, model)
+    elif settings.AI_PROVIDER == "anthropic":
+        return await _generate_anthropic(prompt, max_tokens, temperature, model)
+    else:
+        raise ValueError(f"Unknown AI provider: {settings.AI_PROVIDER}")
+
+
+async def _generate_openai(
+    prompt: str,
+    max_tokens: int,
+    temperature: float,
+    model: str,
+) -> dict:
+    """Generate text using OpenAI."""
+    response = await openai_client.chat.completions.create(
+        model=model,
+        messages=[{"role": "user", "content": prompt}],
+        max_tokens=max_tokens,
+        temperature=temperature,
+    )
+    
+    return {
+        "text": response.choices[0].message.content,
+        "model": model,
+        "usage": {
+            "prompt_tokens": response.usage.prompt_tokens,
+            "completion_tokens": response.usage.completion_tokens,
+        },
+    }
+
+
+async def _generate_anthropic(
+    prompt: str,
+    max_tokens: int,
+    temperature: float,
+    model: str,
+) -> dict:
+    """
+    Generate text using Anthropic Claude.
+    
+    NOTE: Uncomment and install anthropic package to use.
+    """
+    # import anthropic
+    # client = anthropic.AsyncAnthropic(api_key=settings.ANTHROPIC_API_KEY)
+    # 
+    # response = await client.messages.create(
+    #     model=model,
+    #     max_tokens=max_tokens,
+    #     messages=[{"role": "user", "content": prompt}],
+    # )
+    # 
+    # return {
+    #     "text": response.content[0].text,
+    #     "model": model,
+    #     "usage": {
+    #         "prompt_tokens": response.usage.input_tokens,
+    #         "completion_tokens": response.usage.output_tokens,
+    #     },
+    # }
+    
+    raise NotImplementedError("Anthropic provider not configured")
+
+
+# =============================================================================
+# CHAT COMPLETION
+# =============================================================================
+
+async def chat_completion(
+    messages: list[dict],
+    max_tokens: int = 500,
+    temperature: float = 0.7,
+    model: Optional[str] = None,
+) -> dict:
+    """
+    Generate a chat response from message history.
+    
+    Args:
+        messages: List of message dicts with 'role' and 'content'
+        max_tokens: Maximum tokens in the response
+        temperature: Creativity level
+        model: Override the default model
+    
+    Returns:
+        dict with keys: message, model
+    
+    Example:
+        result = await chat_completion([
+            {"role": "system", "content": "You are a helpful assistant."},
+            {"role": "user", "content": "Hello!"},
+        ])
+        print(result["message"]["content"])
+    """
+    model = model or settings.AI_MODEL
+    
+    response = await openai_client.chat.completions.create(
+        model=model,
+        messages=messages,
+        max_tokens=max_tokens,
+        temperature=temperature,
+    )
+    
+    return {
+        "message": {
+            "role": "assistant",
+            "content": response.choices[0].message.content,
+        },
+        "model": model,
+    }
+
+
+# =============================================================================
+# STREAMING
+# =============================================================================
+
+async def stream_text(
+    prompt: str,
+    max_tokens: int = 500,
+    temperature: float = 0.7,
+    model: Optional[str] = None,
+) -> AsyncGenerator[str, None]:
+    """
+    Stream generated text token by token.
+    
+    This is an async generator that yields text chunks as they're generated.
+    Use this for real-time chat interfaces.
+    
+    Example usage in FastAPI:
+        @app.post("/stream")
+        async def stream_endpoint(request: Request):
+            return StreamingResponse(
+                stream_text(request.prompt),
+                media_type="text/event-stream",
+            )
+    """
+    model = model or settings.AI_MODEL
+    
+    response = await openai_client.chat.completions.create(
+        model=model,
+        messages=[{"role": "user", "content": prompt}],
+        max_tokens=max_tokens,
+        temperature=temperature,
+        stream=True,  # Enable streaming
+    )
+    
+    # Yield each chunk as it arrives
+    async for chunk in response:
+        if chunk.choices[0].delta.content:
+            yield chunk.choices[0].delta.content
+
+
+# =============================================================================
+# UTILITY FUNCTIONS
+# =============================================================================
+
+async def count_tokens(text: str, model: str = "gpt-3.5-turbo") -> int:
+    """
+    Count the number of tokens in a text string.
+    
+    Useful for:
+    - Estimating costs before making API calls
+    - Ensuring prompts don't exceed model limits
+    
+    Note: This is an approximation. For exact counts, use tiktoken library.
+    """
+    # Rough approximation: ~4 characters per token for English
+    return len(text) // 4
+
+
+def get_model_context_limit(model: str) -> int:
+    """
+    Get the context window size for a model.
+    
+    Useful for knowing how much text you can send/receive.
+    """
+    limits = {
+        # OpenAI
+        "gpt-4-turbo": 128000,
+        "gpt-4": 8192,
+        "gpt-3.5-turbo": 16385,
+        # Anthropic
+        "claude-3-opus-20240229": 200000,
+        "claude-3-sonnet-20240229": 200000,
+        "claude-3-haiku-20240307": 200000,
+    }
+    return limits.get(model, 4096)
diff --git a/backend/services/database.py b/backend/services/database.py
new file mode 100644
index 0000000..81e4b0b
--- /dev/null
+++ b/backend/services/database.py
@@ -0,0 +1,172 @@
+"""
+=============================================================================
+DATABASE SERVICE (SUPABASE)
+=============================================================================
+
+This module provides Supabase database integration.
+
+Supabase Features Used:
+- Database: PostgreSQL with auto-generated REST API
+- Auth: User authentication (handled in frontend mostly)
+- Real-time: Websocket subscriptions (used in frontend)
+- Storage: File uploads (not shown here, but easy to add)
+
+Why Supabase?
+- Free tier is generous (great for hackathons!)
+- Built-in auth that works with frontend
+- Real-time subscriptions out of the box
+- Full PostgreSQL (can run raw SQL)
+- Great documentation
+"""
+
+from supabase import create_client, Client
+from config import settings
+from functools import lru_cache
+
+
+@lru_cache
+def get_supabase_client() -> Client:
+    """
+    Get a cached Supabase client instance.
+    
+    Using lru_cache ensures we reuse the same client,
+    avoiding connection overhead.
+    
+    Note: We use the SERVICE KEY here (not anon key) because
+    the backend needs admin access to the database.
+    The frontend uses the anon key with Row Level Security.
+    """
+    if not settings.SUPABASE_URL or not settings.SUPABASE_SERVICE_KEY:
+        raise ValueError(
+            "Supabase not configured. "
+            "Set SUPABASE_URL and SUPABASE_SERVICE_KEY in .env"
+        )
+    
+    return create_client(
+        settings.SUPABASE_URL,
+        settings.SUPABASE_SERVICE_KEY,
+    )
+
+
+# =============================================================================
+# DATABASE HELPERS
+# =============================================================================
+
+async def get_user_by_id(user_id: str) -> dict | None:
+    """
+    Get a user by their ID.
+    
+    Example:
+        user = await get_user_by_id("123-456-789")
+        print(user["email"])
+    """
+    supabase = get_supabase_client()
+    
+    response = supabase.auth.admin.get_user_by_id(user_id)
+    
+    return response.user if response else None
+
+
+async def verify_jwt_token(token: str) -> dict | None:
+    """
+    Verify a JWT token and get the user.
+    
+    Use this to authenticate API requests that include
+    the Supabase access token.
+    
+    Example:
+        @app.get("/protected")
+        async def protected_route(authorization: str = Header()):
+            token = authorization.replace("Bearer ", "")
+            user = await verify_jwt_token(token)
+            if not user:
+                raise HTTPException(status_code=401)
+    """
+    supabase = get_supabase_client()
+    
+    try:
+        response = supabase.auth.get_user(token)
+        return response.user if response else None
+    except Exception:
+        return None
+
+
+# =============================================================================
+# GENERIC CRUD HELPERS
+# =============================================================================
+
+class DatabaseTable:
+    """
+    Helper class for common database operations.
+    
+    Example:
+        projects = DatabaseTable("projects")
+        all_projects = await projects.get_all()
+        project = await projects.get_by_id("123")
+        new_project = await projects.create({"name": "My Project"})
+    """
+    
+    def __init__(self, table_name: str):
+        self.table_name = table_name
+        self.client = get_supabase_client()
+    
+    async def get_all(self, filters: dict = None) -> list[dict]:
+        """Get all records, optionally filtered."""
+        query = self.client.table(self.table_name).select("*")
+        
+        if filters:
+            for key, value in filters.items():
+                query = query.eq(key, value)
+        
+        response = query.execute()
+        return response.data
+    
+    async def get_by_id(self, record_id: str) -> dict | None:
+        """Get a single record by ID."""
+        response = (
+            self.client.table(self.table_name)
+            .select("*")
+            .eq("id", record_id)
+            .single()
+            .execute()
+        )
+        return response.data
+    
+    async def create(self, data: dict) -> dict:
+        """Create a new record."""
+        response = (
+            self.client.table(self.table_name)
+            .insert(data)
+            .execute()
+        )
+        return response.data[0]
+    
+    async def update(self, record_id: str, data: dict) -> dict:
+        """Update an existing record."""
+        response = (
+            self.client.table(self.table_name)
+            .update(data)
+            .eq("id", record_id)
+            .execute()
+        )
+        return response.data[0] if response.data else None
+    
+    async def delete(self, record_id: str) -> bool:
+        """Delete a record."""
+        self.client.table(self.table_name).delete().eq("id", record_id).execute()
+        return True
+
+
+# =============================================================================
+# EXAMPLE: Projects Table Helper
+# =============================================================================
+
+# Create a helper instance for the projects table
+projects_db = DatabaseTable("projects")
+
+# You can now use:
+# await projects_db.get_all()
+# await projects_db.get_by_id("123")
+# await projects_db.create({"name": "Test", "user_id": "abc"})
+# await projects_db.update("123", {"name": "Updated"})
+# await projects_db.delete("123")