ngmisl/mojo-qa

The system ingests Mojo documentation, chunks it into embeddings stored in DeepLake's vector database, and uses LangChain's retrieval-augmented generation (RAG) pipeline to answer queries via OpenAI's language models. It leverages DeepLake's managed vector storage for efficient semantic search, enabling contextual answers grounded in official Mojo documentation rather than general model knowledge.

No commits in the last 6 months.