AWS re:Invent 2025 - Build & Scale Enterprise Agentic Workflows on Trusted Data Foundations (AIM239)

Summary of AWS re:Invent 2025 - Build & Scale Enterprise Agentic Workflows on Trusted Data Foundations (AIM239)

Introduction to Informatica

• Informatica is a data management and analytics company recently acquired by Salesforce
• They focus on solving critical data challenges around data discovery, integration, quality, and governance
• Informatica's Intelligent Data Management Cloud (IDMC) is a cloud-native, microservices-based platform that leverages AI and ML at scale

Evolution of Informatica's AI Capabilities

• Informatica's native AI engine is called CLA (Cognitive Learning Agent)
• CLA includes a co-pilot for data engineers/developers and a GPT-based conversational interface
• Informatica has also introduced purpose-built data management agents for automation of tasks like data discovery, ETL, data lineage, and data quality

Informatica's Agentic AI Blueprint for AWS

• Key enterprise requirements for building AI agents:
  • Ability to bring data from across the enterprise
  • Incorporating business context and metadata
  • Ensuring data quality for trustworthy AI applications
  • Enabling no-code/low-code agent development
  • Enforcing data governance policies
• Informatica's approach:
  • Embedding enterprise data into formats understood by large language models
  • Leveraging Informatica's Master Data Management (MDM) to provide high-quality reference data
  • Dynamically selecting data based on quality scores and providing responses in business context
  • Enabling no-code/low-code agent development using Informatica's platform
  • Integrating with AWS Bedrock Agent Core via MCP (Metadata Connectivity Protocol) connectors

Customer Case Study: Citizens Bank

• Citizens Bank is a large US bank focused on building a trusted, enterprise-wide data ecosystem
• Key pillars of their data strategy:
  • Reducing time-to-insights for data and events
  • Anchoring on Informatica for reference and customer master data management
  • Expanding their predictive AI capabilities to include generative AI and agentic workflows
• Citizens Bank's Agentic AI Platform Architecture:
  • Leverages Informatica's data capabilities as a foundation
  • Integrates large language models, MCP services, and code repositories
  • Emphasizes observability, governance, and a feedback loop for the AI agents

Key Takeaways

• Informatica is enabling enterprises to build agentic AI workflows by leveraging their trusted data foundations
• Customers like Citizens Bank are incrementally expanding their data ecosystems to incorporate generative AI and agentic capabilities
• The focus is on complementing existing investments, not starting from scratch, to drive immediate business value
• Tight integration with AWS services like Bedrock Agent Core allows enterprises to build secure, governed, and explainable AI agents