A Visual AI Search Flow Builder for OpenSearch: Build GenAI and Vector...- Tyler Ohlsen & Owais Kazi

A Visual AI Search Flow Builder for OpenSearch: Build GenAI and Vector Search Solutions Faster - Tyler Ohlsen & Owais Kazi, Amazon Web Services OpenSearch Flow is a low code/no code user interface that simplifies the creation of AI/ML workflows within OpenSearch. It enables builders to develop diverse search solutions including semantic, hybrid, and multimodal search, as well as retrieval-augmented generation (RAG). Users can create functional solutions in minutes by simply uploading sample data and selecting models. The platform offers flexible customization through iterative development, allowing users to incorporate data transformations and models precisely where needed. Completed workflows can be exported as templates, facilitating consistent deployment across multiple OpenSearch clusters. Complementing this interface is the OpenSearch Flow Framework plugin, which streamlines ML resource deployment with a suite of robust APIs. The plugin features pre-configured templates for various ML models from leading providers like Cohere, OpenAI, and DeepSeek. Users can establish ML resources and inference endpoints in OpenSearch through a single API call. Join us to learn how OpenSearch Flow and the Flow Framework plugin empower builders to leverage OpenSearch’s expanding suite of AI/ML capabilities.