Financial Document Intelligence System

AI-powered analysis of investment documents (CIMs, investment memos) using NLP and LLMs.

Problem

Investment professionals spend 20+ hours manually reviewing 100-300 page deal documents, creating bottlenecks in deal screening.

Solution

Automated system that extracts financial metrics, flags risks, and generates AI summaries.

Result: 20+ hours → 5 minutes (99.6% time reduction)

Tech Stack

• Frontend: Streamlit
• LLMs: OpenAI GPT-4
• NLP: spaCy, FinBERT
• PDF Processing: pdfplumber
• Database: DuckDB

Installation

git clone https://github.com/leaalonzo/financial-doc-analyzer.git
cd financial-doc-analyzer
python -m venv venv
source venv/bin/activate
pip install -r requirements.txt
python -m spacy download en_core_web_sm
export OPENAI_API_KEY=sk-your-key-here

Create .env file

cp .env.example .env

Edit .env and add your actual OpenAI API key

OPENAI_API_KEY="sk-your-actual-key-here"

Usage

streamlit run app.py

Upload a PDF and get analysis in ~1 minute.

Key Features (Completed)

• AI-Generated Executive Summary (GPT-4)
• Financial metric extraction
• Risk detection (15+ categories)
• Sentiment analysis (FinBERT)
• Entity recognition

Impact

• 20+ hours → 5 minutes per document
• 10x deal throughput increase
• Standardized analysis framework

License

MIT

---

Disclaimer: Educational project. Not financial advice.