Generative AI with Amazon SageMaker AI JumpStart and MongoDB Atlas Vector Search

Amazon SageMaker AIJumpStart provides pre-trained AI foundation models such as Retrieval Augmented Generation (RAG) for intelligent text applications. You can combine JumpStart with MongoDB Atlas Vector Search, which enables semantic similarity queries on text, image, and other data, to build powerful search experiences. For example, your developers can implement intuitive semantic search over customer conversations by using Atlas Vector Search, and use Amazon SageMaker AI RAG models to add interactive summarization and translation, as illustrated in the following diagram.

This unlocks a variety of AI-driven search use cases, including automated support, smart content management, content summarization, and enhanced recommendations. By implementing intuitive precision search with MongoDB and generative capabilities from Amazon SageMaker JumpStart, developers can rapidly deliver impactful cognitive search applications.

Key highlights:

For more information about this implementation, see the AWS blog post Retrieval-Augmented Generation with LangChain, Amazon SageMaker AI JumpStart, and MongoDB Atlas Semantic Search.