Search is no longer just about matching keywords modern applications need to understand meaning. From e-commerce to chatbots, recommendation engines to documentation search, AI powered vector search is redefining how information is discovered.
This session dives into how to build your own search engine using
1. Vector embeddings for semantic understanding
2. BM25 keyword search for precision
3. Hybrid Search in OpenSearch that combines the strengths of both
Understanding the Basics of Modern Search
The Limits of Vector Search
How BM25 Works
Why Hybrid Search is the Future
Building a Search Engine in OpenSearch
Real-World Use Cases
Helping find truly open source embeddings can be difficult, addressing this challenge and what is available will be helpful to give people understanding in the ever-changing AI landscape