APEX Accelerator - Government Contracting AI Assistant

A beautiful, AI-powered knowledge base for government contracting built with React, FastAPI, and Google Gemini.

Features

• Intelligent Q&A: Ask questions in natural language and get expert answers
• Semantic Search: Find relevant information across 179 documents instantly
• Source Citations: Every answer includes references to source documents
• Beautiful UI: Apple-inspired design with smooth animations
• Dark Mode: Comfortable viewing in any environment
• RAG Architecture: Retrieval Augmented Generation for accurate responses

Tech Stack

Backend

• FastAPI: Modern Python web framework
• ChromaDB: Vector database for semantic search
• Google Gemini: AI embeddings and text generation
• python-docx: Word document processing

Frontend

• React with TypeScript
• Vite: Fast build tool
• Tailwind CSS: Utility-first styling
• Framer Motion: Smooth animations
• TanStack Query: Data fetching and caching

Prerequisites

• Python 3.9+
• Node.js 18+
• Google Gemini API key

Installation

1. Clone the Repository

cd /mnt/c/Users/CharlesMartin/projects/apex-accelerator

2. Set Up Backend

# Navigate to backend directory
cd backend

# Create virtual environment
python3 -m venv venv

# Activate virtual environment
source venv/bin/activate  # On Windows: venv\\Scripts\\activate

# Install dependencies
pip install -r requirements.txt

3. Configure API Key

The Google Gemini API key is stored securely in the .env file:

Location: backend/.env

Your API key has already been configured:

GOOGLE_API_KEY=AIzaSyA5JiLrp9LnJQpFxidL7YyjpdJcOvqcHaA

Important Security Notes:

• ✅ The .env file is in .gitignore and will NOT be committed to git
• ✅ Never share your API key publicly
• ✅ If you need to regenerate your key, visit: https://makersuite.google.com/app/apikey
• ✅ A .env.example file is provided as a template for others

4. Set Up Frontend

# Open new terminal
cd frontend

# Install dependencies
npm install

Running the Application

Step 1: Start the Backend

cd backend

# Activate virtual environment
source venv/bin/activate  # On Windows: venv\\Scripts\\activate

# Run FastAPI server
python -m uvicorn app.main:app --reload

The backend will be available at: http://localhost:8000

Step 2: Index Your Documents

Before using the chat, you need to index your 179 Word documents:

# In a new terminal or using a tool like Postman/curl
curl -X POST "http://localhost:8000/api/index?documents_path=/path/to/Government%20Contracting"

Important: Replace /path/to/Government Contracting with the actual path to your documents folder.

You mentioned your documents are in a folder called "Government Contracting". Please provide the full path so we can index them.

Monitor indexing status:

curl http://localhost:8000/api/index/status

Step 3: Start the Frontend

# In a new terminal
cd frontend

# Start development server
npm run dev

The frontend will be available at: http://localhost:5173

Usage

1. Open the application at http://localhost:5173
2. Click "Start Chatting" to access the chat interface
3. Ask questions about government contracting
4. View sources for every answer to verify information

Example Questions

• "What are the requirements for government contracts?"
• "How do I register for SAM?"
• "What is a NAICS code?"
• "Explain the FAR regulations"

Project Structure

apex-accelerator/
├── backend/
│   ├── app/
│   │   ├── main.py              # FastAPI application
│   │   ├── models.py            # Pydantic models
│   │   ├── api/
│   │   │   └── routes.py        # API endpoints
│   │   └── rag/
│   │       ├── document_processor.py  # Word doc extraction
│   │       ├── embeddings.py          # Gemini embeddings
│   │       ├── vector_store.py        # ChromaDB interface
│   │       └── query_engine.py        # RAG logic
│   ├── chroma_db/               # Vector database (generated)
│   ├── .env                     # API keys (DO NOT COMMIT)
│   └── requirements.txt
│
├── frontend/
│   ├── src/
│   │   ├── components/
│   │   │   ├── Landing.tsx      # Landing page
│   │   │   └── Chat.tsx         # Chat interface
│   │   ├── lib/
│   │   │   ├── api.ts           # API client
│   │   │   └── utils.ts         # Utilities
│   │   ├── App.tsx
│   │   └── main.tsx
│   └── package.json
│
├── documents/                    # Your Word documents go here
├── .gitignore
└── README.md

API Endpoints

GET /api/

Health check endpoint

POST /api/index

Index documents from a directory

• Query Param: documents_path - Path to documents folder

GET /api/index/status

Get indexing status

POST /api/chat

Send a chat message

• Body: { "message": "your question", "conversation_id": "optional" }

POST /api/search

Semantic search

• Body: { "query": "search term", "top_k": 5 }

GET /api/stats

Get knowledge base statistics

API Key Management

Where Your API Key is Stored

Your Gemini API key is securely stored in:

backend/.env

Accessing Your API Key

1. During Development: The key is automatically loaded from .env
2. For Deployment: Set the GOOGLE_API_KEY environment variable

Rotating Your API Key

If you need to change your API key:

1. Get a new key from: https://makersuite.google.com/app/apikey
2. Update backend/.env:

   GOOGLE_API_KEY=your_new_key_here
   

3. Restart the backend server

Security Best Practices

• ✅ Never commit .env to git (already in .gitignore)
• ✅ Never share your API key in screenshots or documentation
• ✅ Use environment variables in production
• ✅ Rotate keys regularly
• ✅ Monitor API usage in Google AI Studio

Building for Production

Backend

cd backend

# Install production dependencies
pip install -r requirements.txt

# Run with gunicorn
gunicorn app.main:app -w 4 -k uvicorn.workers.UvicornWorker --bind 0.0.0.0:8000

Frontend

cd frontend

# Build for production
npm run build

# Preview production build
npm run preview

The build output will be in frontend/dist/

Troubleshooting

Backend Issues

Problem: "GOOGLE_API_KEY not found"

• Solution: Check that backend/.env exists and contains your API key

Problem: "Error processing documents"

• Solution: Ensure documents are valid .docx files and path is correct

Problem: ChromaDB errors

• Solution: Delete backend/chroma_db/ and re-index documents

Frontend Issues

Problem: "Network Error" or "Connection Refused"

• Solution: Ensure backend is running on http://localhost:8000

Problem: TypeScript errors

• Solution: Run npm install again

Performance Tips

• Indexing: Takes ~1-2 minutes per 100 documents
• Query Speed: First query is slower (~3-5s), subsequent queries are faster (~1-2s)
• Embeddings: Cached locally in ChromaDB for instant retrieval
• Chunk Size: Adjust in document_processor.py for better performance

Future Enhancements

• Document upload via UI
• Conversation history
• Export chat to PDF
• Multi-user support
• Advanced filters
• Document preview modal
• Analytics dashboard

Support

For issues or questions, please refer to the documentation or check the logs:

• Backend logs: Terminal where FastAPI is running
• Frontend logs: Browser console (F12)

License

Private project for APEX Accelerator

---

Built with ❤️ using React, FastAPI, and Google Gemini