This repository contains both the source code for article generation and the blog itself.
This project is an ongoing exploration of how LLMs can be used to extract meaningful insights out of complicated bureaucratic processes to serve the public interest. It is open-ended, with lots of room for ideas and experiments.
Contributions or comments of any sort are welcome.
The blog is a Hugo static site.
git submodule update --recursive # Install hugo themes
cd blog/
hugo server -D
Content in blog/content/posts
is auto-generated. To add any corrections to a summary, edit the file in archive/summary
and re-run generation.
Article generation is scripted in Python and uses Anthropic's Claude AI.
- You'll need to install Python and Poetry.
- You'll also need access to the Anthropic API
- LIMS scraping and Transcript scraping do not require authentication.
cp .env.example .env
# Add your Anthropic API key
poetry install
poetry run generate