Text to SQL using LLMs #10 - Making the Semantic Layer Searchable | LangGraph

We have a semantic layer - now we make it searchable. In this video, we build the semantic embedder: converting YAML schema files into vector embeddings stored in ChromaDB, then retrieving only the relevant tables at query time using cosine similarity. This is RAG applied to schema context, and it lets us kill two nodes from our pipeline. 🔗 GitHub Repo: https://github.com/v1jaysundaram/text2sql 🎥 Full Playlist: https://youtube.com/playlist?list=PLEHeb1HGikpQJvPlZEyrMLOjopsV5Ui_3&si=8dqEScL-uSyIskqx 🤝 Connect with me on LinkedIn: https://www.linkedin.com/in/vijay-sundaram/