Claude Code Journey

I have been using Claude Code for a few months now, first on small personal code bases that urgently needed refactoring and then professionally on small internal tools then bigger code bases that are actually in production but are legacy code that haven’t been updated in a few years.

Now I’m finally working on my own attempt at making sense of Juridical Data (65GB) of PDF’s and media coverage of the OKC judicial case.

I’m basically creating a Juridically oriented LLM/RAG solution to ingest PDF content and then use pgvector and other technologies to be able to interact with the whole Archive from my browser interface.

This will never be published publicly obviously but it’s a very interesting use case and learning curve.