2023-08-25-sagemaker-haystack-opensearch.markdown

authors	category	date	excerpt	feature_image	has_math	has_science_table	layout	meta_description	meta_keywords	odfeimport	title	twittercard
dtaivpp tuanacelik	partner-highlight	2023-08-25 00:00:01 -0700	Here we will show how to build an end-to-end generative AI application for enterprise search with Retrieval Augmented Generation (RAG) using Haystack, OpenSearch, and Sagemaker JumpStart.	/assets/media/blog-images/2023-08-25-sagemaker-haystack-opensearch/Architecture.png	false	false	post	Amazon Personalize launches a new integration with self-managed OpenSearch that enables customers to personalize search results for each user and predict their needs.	SageMaker Jumpstart, OpenSearch, OpenSearch Dashboards, RAG, retrieval augmented generation, generative ai	false	Partner Highlight: Build production-ready generative AI applications for enterprise search using Haystack pipelines and Amazon SageMaker JumpStart with LLMs	image_alt site type Haystack, Sagemaker, and OpenSearch Architecture for Retreval Augmented Generation @OpenSearchProj summary_large_image

{: .img-fluid}

Ever wanted to get started with generative AI but didn't know where to start? Check out the AWS Machine Learning blog post "Build production-ready generative AI applications for enterprise search using Haystack pipelines and Amazon SageMaker JumpStart with LLMs". In this article we discuss how retrieval augmented generation (RAG) works and how to set up a RAG pipeline with Haystack, OpenSearch, and SageMaker JumpStart.

The article first provides a bit of background on RAG and how these pipelines typically work. Next, it walks through deploying OpenSearch and models on Amazon SageMaker JumpStart. Then it walks through ingesting documents into OpenSearch using the Haystack Python library. Finally, it shows how to use the data in a RAG pipeline. Check out the full blog post, which includes a GitHub repo with code so you can follow along yourself.