I’ve been working on Trainly, a Graph-RAG API that helps developers go from raw documents → working RAG endpoints in minutes.
Instead of wiring LangChain, vector DBs, and retrievers manually, Trainly lets you: 1. `pip install trainly` 2. Initialize with your API key 3. Upload your docs 4. Query your endpoint