Inspiration Construction contracts often span hundreds of pages, making manual review slow, expensive, and prone to human error. Missing a single high-risk clause can lead to financial losses, legal disputes, or project delays. We wanted to build an AI-powered solution that helps legal teams, contractors, and project managers identify risks instantly and make contract reviews faster and more reliable. What it does The Autonomous Contract Risk Analyzer is an Agentic RAG-powered application that automatically analyzes construction contracts, retrieves relevant clauses, identifies potential legal and financial risks, and generates clear explanations and recommendations. Instead of spending weeks reviewing contracts manually, users receive comprehensive risk insights within seconds. How we built it We built the system using a Retrieval-Augmented Generation (RAG) pipeline combined with an AI agent that understands legal documents. Contract PDFs are processed, converted into embeddings, stored in a vector database, and retrieved based on user queries. A Large Language Model then analyzes the retrieved content to identify risky clauses, summarize findings, and recommend actions through an intuitive web interface. Challenges we ran into 1)Processing very large contract documents efficiently. 2)Retrieving only the most relevant legal clauses. 3)Reducing hallucinations by grounding responses in retrieved contract text. 4)Presenting legal insights in a simple and understandable format. Accomplishments that we're proud of 1)Reduced contract review time from weeks to seconds. 2)Built an autonomous AI workflow for legal document analysis. 3)Improved accuracy by combining RAG with intelligent agent reasoning. 4)Created a scalable solution that can be adapted to multiple legal domains. What we learned This project strengthened our understanding of Agentic AI, Retrieval-Augmented Generation, vector databases, prompt engineering, and document intelligence. We also learned how AI can effectively support professionals in complex legal workflows while keeping humans in the decision-making loop. What's next for Autonomous Contract Risk Analyzer 1)Support contracts from multiple industries. 2)Add multilingual contract analysis. 3)Integrate compliance checking against legal regulations. 4)Generate automated risk reports and negotiation suggestions. 5)Enable real-time collaboration for legal teams.

Built With

  • agentic
  • ai
  • chromadb
  • docker
  • face
  • faiss
  • fastapi
  • generation
  • git
  • github
  • gpt
  • hugging
  • langchain
  • openai
  • pdfplumber
  • pymupdf
  • python
  • rag
  • retrieval-augmented
  • sentence
  • streamlit
  • transformers
Share this project:

Updates