-
Book Overview & Buying
-
Table Of Contents
LLMs in Enterprise
By :
The realization of connected LLM systems in production environments depends critically on a suite of supporting technologies that handle the complex coordination between models, tools, and data flows. Recent research from Stanford’s Center for Research on Foundation Models highlights how these enabling technologies collectively address what they term the “orchestration gap” – the significant challenge of making multiple AI components work together reliably (Bommasani et al., 2023).
Two prominent approaches emerging to close this gap are the Model Context Protocol (MCP) and agent-to-agent (A2A) communication. MCP provides a standardized interface for transmitting structured context (including task intent, user state, and prior outputs) between models, enabling consistent behavior across different systems. A2A communication focuses on enabling autonomous agents or specialized LLMs to directly exchange messages, decisions, or...