Ever wondered how to get those clever Large Language Models to actually do stuff beyond just chatting? This talk spills the beans on using Prompting LLMs with Model Context Protocol (MCP) to tap into tools and external resources. It's not just for fancy apps; even your own LLMs can learn to grab info and make magic happen, paving the way for truly 'agentic' software!
Dive in to discover how different models react to tool-use prompts, the good and bad of various data formats for those tool calls, and peek at real examples from MCP servers and popular agent frameworks. Get ready to level up your LLM game!
Watch on YouTube
Top comments (0)