Search with (multi-)vectors

Late interaction models for Search - Better relevance with fine-grained representative vectors

Colpali for Multimodal RAG

Late interaction models for RAG, advantages and trade-offs

Multimodal RAG

Don't miss the insights from the images, tables and charts of your PDF documents when building RAG pipelines

Rediscover your Lexical search

You always don't need vectors! Empower your Lexical Search with semantic match capabilities by applying Expansion, Enrichment and Rewriting techniques

Build RAG with AWS Lambda: Solution Architecture

Understand your user queries better using LLMs

Auto-extract filters from natural language queries of your users and use LLMs to write OpenSearch DSL queries

ML Search

Learn different ML Search types of OpenSearch

ML with Lexical Search

Text-expansion technique, a ML technique to empower Lexical Search with semantic matching capabilities

AI agents, the new users of Search

Using Agents in retrieval pipelines to build self reflective, iterative and relevant search

Explainable vector search using late interaction models

Multimodal RAG Part 2

Multimodal RAG using (i) Grounding to text approach (ii) Colpali approach, deepdive and tradeoffs

OpenSearch ML capabilites

Learn how OpenSearch has evolved by applying ML at every stage of a query lifecycle to improve relevance