LLM

Exploring the Randomness of LLMs Large Language Models (LLMs) have become central to modern development—powering everything from RAG systems to AI agents that assist with our daily tasks. Yet, most developers who rely on LLMs rarely pause to explore their behavior. Instead, they depend on benchmarks, anecdotal usage, or viral posts on X to form their mental models of what LLMs can (or cannot) do. I believe this is risky—it creates a gap between people who use these models and those who deeply evaluate them. To help bridge that gap, I decided to run a simple but fun experiment: ...

Last year, the Anthropic team introduced the MCP API protocol, an open standard for connecting external tools to Large Language Models. The major contribution of this protocol is that it converts an N (Number of different LLMs) × N (Number of different services) problem into a 1 × N problem. I am currently working in the Agentic Solution Space and became curious about the protocol. I would like to share my journey and experience. ...