Logo Lanfrica

st-vin/legislative-observer

Domain:

natural language processing

Record type:

software
Creator:
st-
Host:
A civic technology platform that transforms Kenya's official parliamentary Hansard into plain English summaries, searchable by date, topic, and MP empowering citizens to hold their representatives accountable. # Bunge Summary Bunge Summary is an AI-powered legislative observer designed to bring transparency, clarity, and accessibility to parliamentary proceedings. By leveraging natural language processing, it ingests, summarizes, and categorizes parliamentary Hansards, making civic engagement easier for everyone. ## Developer's Intent The core goal of this project is to bridge the gap between complex legislative documents and the everyday citizen. By combining a robust tech stack—including Django, React, and Gemini AI—Bunge Summary provides a streamlined platform to track bills, understand member contributions, and stay informed on what happens in parliament without needing to sift through hundreds of pages of raw transcripts. It is more than just a software tool—it's a step towards open governance and digital democracy. If you are passionate about civic tech and building applications that empower the public, you've found the right codebase. ## Quick Guide Navigating the project is simple: - **Frontend (`/frontend`):** A modern React application serving as the user interface to visualize summaries and explore legislative data. - **Backend (`/backend`):** A Django REST API powering the core logic, integrated with PostgreSQL (pgvector) and Celery for handling background AI processing tasks. - **Setup & Installation:** - Want to spin up the entire stack quickly using containers? See the detailed SETUP.md or DOCKER_COMPLETE_PROJECT_SETUP.md. - Prefer to run things natively? Check out SETUP_WITHOUT_DOCKER.md. Enjoy exploring the repository. Let's make democracy more accessible, one summary at a time!