Kensho's Multi-Agent Framework with LangGraph for Trusted Financial Data

Kensho, S&P Global's AI innovation engine, leveraged LangGraph to create a unified agentic access layer called Grounding, solving the challenge of fragmented financial data retrieval at enterprise scale.

Details

Kensho, an AI innovation engine within S&P Global, developed a multi-agent framework using LangGraph to create the Grounding system - a unified agentic access layer for trusted financial data retrieval. Financial data is often fragmented and siloed across different systems and sources, making it challenging for enterprises to access and utilize this information. Kensho's Grounding framework aims to solve this problem by providing a consistent, reliable, and scalable way to access financial data from various sources through a single interface. The multi-agent architecture allows for specialized agents to handle different data types and sources, while LangGraph provides a common language understanding and reasoning layer to enable seamless data integration and querying.