In the age of generative AI, search expectations have changed. Users now want more than just keyword matches - they expect intelligent, context-aware results. Whether you're building a recipe recommendation engine or enhancing e-commerce discovery, vector search offers a powerful upgrade to the overall search ecosystem.

This shift is driven by the rise of generative AI models that understand natural language. Users now expect search to behave more like a conversation and less like a command-line filter. Traditional keyword-based search is hitting its limits, as it relies on exact or fuzzy matches of words in metadata, missing opportunities to truly understand user intent.

In this talk, we'll explore a practical example of how you can enhance your keywords search with vector search, powered by embeddings and semantic understanding, can dramatically improve the search experience. For a query like "Quick, healthy dinner with tofu, no dairy", a vector search engine can go beyond simple keyword matching to:

Identify recipes labeled as "30-minute meals" as relevant for "quick"
Surface dishes that are nutrient-dense, even without an explicit "healthy"