Google just announced a new interoperability protocol for AI agents called Agent 2 Agent (A2A). In this post I will share some of my learnings from playing with some of the early bits of A2A in the official Google repo. Specifically, I will show how I ported one of my previous agent implementations to use A2A. I also changed the A2A sample to work with local LLMs (qwen and Llama) instead of the default gemini implementation.
In this article I will show how to create a model context protocol (MCP) server and integrate it with a local LLM to implement a multi-step tool calling workflow.
In this article I will show how to use LangGraph to create an AI agent that works with multiple local LLMs. Through a simple example I will implement a news agent that loads a rss feed and provides short news summaries and categorization of the stories.
In a previous article I wrote about an experiment where I trained a neural network to play a card game. As a follow up to this project, I figured it would be fun to see if I could get an LLM to play the game instead. This post is a write up of how I arrived at a successful LLM based implementation of the original card game.
In this article I will show how to use a local LLM with RAG and tool calling to extract structured data from a document. My first example will show how to use tool calling to expand content from http links in the original prompt. The second example will show how to use tool calling to extract structured content that requires more detailed analysis of the content.
In the following blog post I will show how to use machine learning to teach a computer to play cards. The game used for these experiments is called “Palace”, but some of you may also know the same game as “Idiot”
In the following blog post I will show how to create a neural network for classifying various traffic signs.