FinRAG

A production-grade, citation-enforced financial research assistant over SEC filings and earnings call transcripts.

What This Does

FinRAG answers questions about SEC filings (10-K, 10-Q, 8-K) and earnings call transcripts. Every answer is grounded in a specific paragraph from a specific filing, with company, period, section, and page attached. When evidence doesn't support a claim, the system refuses to answer.

Stack

• Orchestration: LangGraph
• Vector Store: ChromaDB
• Retrieval: BM25 + sentence-transformers + cross-encoder reranking
• Evaluation: RAGAS + LLM-as-Judge
• Observability: Langfuse
• API: FastAPI with SSE streaming
• Guardrails: guardrails-ai + NeMo Guardrails

Setup

# Clone the repo
git clone <repo-url>
cd finrag

# Create virtual environment
python -m venv .venv
.venv\Scripts\activate  # Windows

# Install dependencies
pip install -e ".[dev]"

# Copy and configure environment
cp .env.example .env
# Edit .env with your details

Quick Start

# Download a filing
python scripts/ingest.py --ticker AAPL --filing-type 10-K --count 1

Project Status

🏗️ Under active development. See DEBT_LEDGER.md for known shortcuts.