Client Onboarding System - Estimation Bot API

🚀 Overview

A production-ready FastAPI backend that provides intelligent time and cost estimates for client onboarding systems. Uses OpenAI GPT-4 to generate ChatGPT-like responses based on client requirements.

✨ Features

• AI-Powered Estimation: Uses OpenAI GPT-4 to extract features and generate estimates
• Natural Language Processing: Understands client requirements in plain English
• ChatGPT-like Responses: Generates professional, detailed responses matching example format
• Comprehensive Breakdowns: Feature-by-feature time and cost estimates
• RESTful API: Clean, documented API endpoints ready for frontend integration
• Railway Ready: Pre-configured for one-click deployment on Railway

📁 Project Structure

Estimate_Bot/
├── app/
│   ├── api/              # API endpoints
│   ├── core/             # Configuration
│   ├── models/            # Pydantic schemas
│   ├── services/          # Business logic
│   │   ├── estimation_engine.py
│   │   ├── knowledge_base.py
│   │   └── openai_service.py
│   ├── utils/            # Utilities (PDF extraction)
│   └── main.py           # FastAPI app
├── requirements.txt       # Python dependencies
├── Procfile              # Railway deployment config
├── railway.json          # Railway config
├── DEPLOYMENT.md         # Deployment guide
└── .env.example          # Environment variables template

🚀 Quick Start

Local Development

# 1. Clone and navigate
cd Estimate_Bot

# 2. Activate virtual environment (already created)
source venv/bin/activate  # Windows: venv\Scripts\activate

# 3. Install dependencies
pip install -r requirements.txt

# 4. Set up environment variables
# Edit .env and add your OPENAI_API_KEY
# The .env file is already created

# 5. Run the server
uvicorn app.main:app --reload

Using Docker Locally (Optional)

# Build and run with Docker
docker build -t estimation-bot .
docker run -p 8000:8000 --env-file .env estimation-bot

API will be available at http://localhost:8000

Deploy to Railway

See DEPLOYMENT.md for detailed instructions.

Quick Deploy:

1. Push code to GitHub
2. Connect Railway to your GitHub repository
3. Railway will auto-detect Dockerfile
4. Set OPENAI_API_KEY environment variable in Railway dashboard
5. Deploy!

📡 API Endpoints

POST /api/v1/estimate

Generate time and cost estimate.

Request:

{
  "requirements": "I need a client onboarding system with user authentication, dashboard, and payment processing",
  "hourly_rate": 100.0,
  "include_breakdown": true
}

Response:

{
  "total_time_hours": 240.0,
  "total_cost": 24000.0,
  "breakdown": [
    {
      "feature": "User Authentication",
      "time_hours": 52.0,
      "cost": 5200.0,
      "complexity": "medium"
    }
  ],
  "assumptions": [...],
  "timeline": "Approximately 6 weeks",
  "summary": "AI-generated summary...",
  "next_steps": [...]
}

POST /api/v1/chat

Chat endpoint for conversational interaction.

GET /api/v1/features

Get list of available features.

GET /api/v1/health

Health check endpoint.

Interactive API Docs: Visit /docs for Swagger UI or /redoc for ReDoc.

🔧 Configuration

Environment Variables

OPENAI_API_KEY=your_key_here          # Required
DEFAULT_HOURLY_RATE=100.0             # Optional
BUFFER_PERCENTAGE=0.20                # Optional
OPENAI_MODEL=gpt-4-turbo-preview      # Optional

Default Settings

• Hourly Rate: $100/hour
• Buffer: 20%
• Model: GPT-4 Turbo Preview
• Temperature: 0.7

🧪 Testing

# Health check
curl http://localhost:8000/api/v1/health

# Get features
curl http://localhost:8000/api/v1/features

# Create estimate
curl -X POST http://localhost:8000/api/v1/estimate \
  -H "Content-Type: application/json" \
  -d '{
    "requirements": "User authentication and dashboard"
  }'

📚 Documentation

• DEPLOYMENT.md - Railway deployment guide
• PROJECT_WORKFLOW.md - Complete project workflow
• API Docs - Available at /docs when server is running

🛠️ Technology Stack

• Framework: FastAPI
• AI: OpenAI GPT-4 API
• Vector Database: Qdrant (in-memory or Qdrant Cloud)
• Embeddings: Sentence Transformers (all-MiniLM-L6-v2)
• PDF Processing: pdfplumber, PyPDF2
• DOCX Processing: python-docx
• Validation: Pydantic
• Deployment: Railway (Docker)
• Frontend: Vanilla HTML/CSS/JS (included)

📝 Notes

• The system extracts features from client requirements using AI
• Estimates include a 20% buffer for unforeseen complexities
• All responses are formatted to match ChatGPT example style
• Knowledge base is built from provided PDF/DOCX documents

🔐 Security

• API keys stored in environment variables
• CORS configured for frontend integration
• Input validation via Pydantic schemas
• Error handling and logging included

📞 Support

For deployment issues, see DEPLOYMENT.md.
For API usage, check /docs endpoint when server is running.