RAGnificent
Documentation scraper with a Rust parsing core and a Python async layer, built for RAG pipelines.
The idea
A project to learn foreign function interface patterns between Rust and Python, aimed at a real need: turning documentation sites into clean input for retrieval-augmented generation.
What I built
Rust handles HTML parsing and conversion to markdown, JSON and XML. Python calls the Rust core through FFI bindings and provides the async API integration layer and the RAG pipeline: chunking, embeddings and search.
License
Open source under MIT.