Learning out loud,
one curious build at a time.

"Agent" is the buzzword everyone reached for the moment chatbots got good. Underneath the hype is a simple idea: an agent is a chatbot you handed a goal, some tools, and permission to keep working. Think of it as an intern.

An agent is a language model plus tools plus a loop. That sounds like a team project. It isn't. Here's the exact shape of that loop, in about a dozen lines of Python, using Claude.

The magic word in "AI agent" isn't the AI. It's the doing. Tools are what let the intern step out of the conversation and touch the real world, and adding one is far less code than you'd guess.

Skeptical that AI is coming for your career? Good. Once you see what a language model actually does (autocomplete that read most of the internet), both the hype and the fear get a lot easier to size up.

AI, ML, LLMs: the words get thrown around like they mean the same thing. Here's a simple mental model that finally made it click for me.

Getting good answers out of an AI is a skill. These are the prompting habits that made the biggest difference for me.

You don't need a fancy setup or a credit card to start playing with AI. Here are the free tools I'd hand a total beginner.

What people mean when they say a model learns, explained with zero equations and a lot of plain English.