Informatica, a leader in enterprise AI-powered cloud data management, announced a broad innovation agenda addressing its GenAI, Open Table and analytics capabilities built on AWS. The new capabilities will help customers build and deploy a new generation of generative artificial intelligence (GenAI) and analytics applications. The announcements were made at AWS re:Invent, AWS’s annual conference.
New developments include blueprints for enterprise-grade GenAI applications and automation recipes for Amazon Bedrock, a fully managed service that provides a single API to access and utilize various high-performing foundation models (FMs) from leading AI companies. Informatica is also a launch partner for Amazon SageMaker Lakehouse. Informatica supports AWS PrivateLink Resource Endpoints for enterprise customers so they can benefit from Informatica’s advanced serverless processing for data integration.
Also Read: boost.ai Launches Toolkit for Global Enterprise Gen AI
Data Management for AI Capabilities for Amazon Bedrock
Informatica’s data management capabilities for Amazon Bedrock support GenAI application development and deployment with architectural blueprints and automation recipes that help customers deploy and jumpstart their GenAI applications.
The recipes include:
- Connectivity to Amazon Bedrock Knowledge Bases
- Standardization of connectivity and orchestration for reuse across an enterprise
- Support for Retrieval Augmented Generation (RAG) through no-code development and deployment
- Data Quality and Master Data Management (MDM) to ensure GenAI applications are grounded and enriched
- Use of business glossary metadata and data governance to ensure GenAI applications are optimized for an enterprise or industry
These blueprints and recipes simplify the creation of data pipelines and data orchestration for RAG use cases in Amazon Bedrock, giving enterprises an advanced and better way to create AI applications that leverage their internally managed private data with Large Language Models (LLMs) safely and responsibly.
SOURCE: Businesswire
