This tutorial guides you in building a coding agent from scratch to understand its inner workings. Prior Python experience and a Claude API key are necessary. The project setup involves using 'uv' to create a project and install 'python-dotenv' and 'anthropic' dependencies. Authentication is managed by placing your Claude API key in a .env file and adding it to .gitignore. The core of a coding agent is a Large Language Model (LLM). This tutorial specifically uses the Claude API due to its user-friendly SDK. You will learn to interact with the LLM through its API. The tutorial emphasizes the concept of tools as a way to extend LLM functionality. A coding agent is defined as an LLM enhanced with tools to interact with its environment. The practical implementation involves sending a message to the LLM using the 'anthropic' Python SDK. The provided code snippet demonstrates how to load environment variables and initialize the Anthropic client. Future steps will involve managing context and implementing tool calling. This hands-on approach demystifies the creation of sophisticated coding agents.