BillFlow

A full-stack contract intelligence platform demonstrating RAG architecture, AI-powered analytics, and modern web development.

Architecture

┌─────────────────────────────────────────────────────────────────┐
│                        React Frontend                           │
│              TypeScript · Tailwind CSS · Vite                   │
└─────────────────────────────────────────────────────────────────┘
                                │
                                ▼
┌─────────────────────────────────────────────────────────────────┐
│                        FastAPI Backend                          │
│                    REST API · Async · CORS                      │
└─────────────────────────────────────────────────────────────────┘
                                │
                ┌───────────────┼───────────────┐
                ▼               ▼               ▼
        ┌──────────────┐ ┌──────────────┐ ┌──────────────┐
        │   RAG Chain  │ │  Intelligence│ │   Revenue    │
        │   LangChain  │ │    Engine    │ │    Engine    │
        └──────────────┘ └──────────────┘ └──────────────┘
                │               │               │
                └───────────────┼───────────────┘
                                ▼
        ┌──────────────────────────────────────────────┐
        │              ChromaDB Vector Store           │
        │         Embeddings · MMR Retrieval           │
        └──────────────────────────────────────────────┘
                                │
                                ▼
        ┌──────────────────────────────────────────────┐
        │              Azure OpenAI                    │
        │       GPT-5-nano · text-embedding-3-small   │
        └──────────────────────────────────────────────┘

Features

RAG-Powered Document Q&A

• Natural language queries across 100+ contract documents
• MMR (Maximal Marginal Relevance) retrieval for diverse, relevant results
• Source citation with document references
• Conversation context maintained across queries

Contract Risk Analysis

• Automated risk scoring based on contract terms
• Churn probability prediction with contributing factors
• Side-by-side contract comparison
• What-if scenario modeling for term changes
• AI-assisted contract generation from natural language

Revenue Analytics Dashboard

• Multi-tab analytics interface (Leakage, Opportunities, Signals, Actions)
• Prioritized action queue with impact scoring
• Client signal detection and alerting
• AI-generated outreach content

Drill-Down Views

Leakage Detection	Opportunity Identification

Signal Monitoring	Action Queue

Tech Stack

Layer	Technologies
Frontend	React 18, TypeScript, Tailwind CSS, Vite
Backend	FastAPI, Python 3.12, Pydantic
AI/ML	LangChain, Azure OpenAI (GPT-5-nano), ChromaDB
Document Processing	PyPDF, tiktoken, text-splitters

Implementation Highlights

RAG Pipeline

• Chunking strategy: 1500 characters with 200 overlap
• Embedding model: text-embedding-3-small (1536 dimensions)
• Retrieval: MMR with k=60 for multi-document comparison queries
• Vector store: ChromaDB with persistent local storage

API Design

• RESTful endpoints with Pydantic validation
• Async request handling
• Structured error responses
• CORS configuration for frontend integration

Frontend Architecture

• Single-page application with client-side routing
• Real-time state management with React hooks
• Responsive design with Tailwind utilities
• Dark mode UI for analytics dashboards

API Endpoints

Method	Endpoint	Description
POST	/chat	Natural language contract queries
GET	/contracts	List contracts with metadata
GET	/metrics	Portfolio aggregations
GET	/intelligence/risk	Risk analysis results
GET	/intelligence/churn	Churn predictions
POST	/intelligence/simulate	What-if scenario modeling
POST	/intelligence/compare	Contract comparison
POST	/intelligence/generate	AI contract generation
GET	/revenue/command-center	Revenue analytics data
POST	/revenue/generate-outreach	AI-generated outreach content

Getting Started

Prerequisites

• Python 3.10+
• Node.js 18+
• Azure OpenAI API access

Setup

# Backend
python -m venv venv
source venv/bin/activate
pip install -r requirements.txt
cp .env.example .env  # Configure Azure OpenAI credentials

# Generate sample data
python generate_contracts.py
python ingestion.py

# Start API server
uvicorn backend_api:app --port 8001

# Frontend
cd frontend
npm install
npm run dev

Open http://localhost:5173

Project Structure

├── backend_api.py           # FastAPI application
├── rag_chat.py              # LangChain RAG implementation
├── contract_intelligence.py # Risk and churn analysis
├── revenue_intelligence.py  # Revenue analytics engine
├── ingestion.py             # Document processing pipeline
├── config.py                # Configuration management
├── generate_contracts.py    # Synthetic data generation
├── data/                    # Contract documents
├── frontend/
│   ├── src/
│   │   ├── App.tsx          # Main application
│   │   ├── main.tsx         # Entry point
│   │   └── index.css        # Tailwind styles
│   └── package.json
└── screenshots/

---