AI & ML interests

We build graph-inspired RAG systems that use document metadata to drive smarter retrieval. By blending vector search with lightweight traversal logic, we unlock more accurate, explainable, and context-rich responses from language models.