Hybrid search is a technique that combines keyword-based searches with vector (embedding) searches to improve search accuracy and relevance, particularly in large-scale applications where traditional methods may fall short. The approach involves using a mix of full-text search (FTS) algorithms like BM-25, which focus on syntactical matches, and vector search algorithms that find semantically similar results. By integrating these methods, hybrid search can leverage user metadata, such as preferences or location, to filter and rank results more effectively. Techniques like score fusion, re-ranking, and stacking results are employed to balance semantic and syntactic search outcomes, ensuring that search results are both precise and relevant to user queries. This approach is particularly useful in contexts where nuanced user preferences or language-specific queries can impact the quality of search results.