FastMCP Podcast: LangChain on the New MCP Spec, Agent Evals, and Decision Models

Hacubu · x · 2026-09-29

The FastMCP podcast features FastMCP creators jlowin and zzstoatzz in conversation with Sydney Runkle of LangChain. Key topics: why LangChain chose FastMCP to connect agents to MCP servers, the new MCP spec, keeping open source collaborative when maintainers prefer coding agents, how to tell whether an MCP tool actually helps an agent, Sydney's prediction that agent development may mostly be eval design a year from now, and experiments with jev for tool selection, model routing, and evals.

Original post →

More from coding & agent

coding & agent channel →