About

The home of Astrobagel: articles and apps from noodling about with AI.

Most of the articles here are practical walkthroughs: what was built, how it works and what was learned along the way, usually with code and a demo. Others are shorter thoughts on how AI fits into the way we work.

Topics so far include AI-assisted research, document processing and search for humanitarian evaluation reports, Model Context Protocol (MCP) servers, satellite map embeddings and climate data, and checking whether the sources a research paper cites actually support what it says.

Projects

  • Evidence Lab, a free, open-source document processing and AI search platform, populated with about 20,000 United Nations humanitarian evaluation reports. The code is on GitHub.
  • Athetize, a Claude plugin that checks whether the sources a paper cites support what the paper says.
  • An MCP server for the Humanitarian Data Exchange (HDX), written up in this article.

Articles from Astrobagel have also appeared in TDS Archive, Data Science Collective and AI Advances.

Get in touch

LinkedIn is the best place for a message: send one here. You can also follow Astrobagel on Medium, and the code for most projects is on GitHub.

Articles are personal views, not professional advice, and code is shared as is, without warranty.