The OpenSearch agent server is an experimental multi-agent orchestration platform that enables users to build and deploy specialized AI agents for specific tasks within OpenSearch.…
OpenSearch 3.5 introduces support for asymmetric embedding models, which enhance semantic search relevance by applying distinct encoding strategies to short queries and long, infor…
The OpenSearch team has developed four AI agents—Atlas, Ralph, Nitro, and Sentinel—to automate repetitive tasks throughout the software development lifecycle, ranging from knowledg…
OpenSearch agentic memory addresses the lack of conversational context in AI contact center agents by providing persistent, personalized memory across multiple interactions. By uti…
The OpenSearch Hadoop connector 2.0 has been released, introducing support for Apache Spark 3.5 and 4, as well as official compatibility for OpenSearch 3.x. The update also adds su…
This article benchmarks three approaches for multimodal document search in OpenSearch—ColPali, BDA modality-aware embedding, and text-only chunking—evaluating their trade-offs in s…
OpenSearch 3.5 introduces a new context management system for OpenSearch Agents designed to prevent context window overflow and performance degradation during long AI conversations…
OpenSearch is adopting gRPC as a high-performance alternative to its traditional REST APIs, addressing limitations with JSON's parsing overhead and payload size for data-intensive …
This article details the evaluation of OpenSearch's agentic search capabilities, which uses LLMs to enable natural language interaction with data. Testing focused on two key areas:…
This article details how the authors overcame limitations in previous date histogram aggregation optimizations in Elasticsearch. Previously, these optimizations (filter rewrite & m…
This article highlights the challenges of deploying and maintaining agentic AI applications in production, specifically the lack of visibility into their decision-making processes.…
OpenSearch has significantly improved the performance of vector search using 16-bit floating point numbers (FP16) over the last three releases. Starting with memory-optimized searc…