Welcome & How to Use This Course

A from-scratch, hands-on path to building real applications with Claude — starting with zero assumptions about machine learning and ending with production-grade capstone projects. This page gets your editor, your account, and your first API call all working before Module 1.

Who this course is for

☺ Like you're 10: You don't need to know how a car engine works to learn to drive it — you just need to know where the pedals are. Same deal with Claude: you call an API, you don't build a brain.

This course assumes no machine learning background whatsoever. You do not need to know what a transformer is, how attention works, or anything about training neural networks — Claude is accessed entirely through an API, and this course treats it that way from the very first line of code.

Basic programming experience is genuinely useful, though. You should be comfortable reading and writing simple scripts in at least one of Python or TypeScript/JavaScript — variables, functions, loops, installing a package. If you can write a script that reads a file and prints something to the console, you already have enough to start.

How the course is organized

☺ Like you're 10: It's a video game with a tutorial level, a long main story you play in order, and then open-world bonus missions at the end.

The course has two parts. Foundations (modules 0–20) build up your understanding one concept at a time, in order — from your first API call through prompting, tool use, agentic patterns, retrieval, safety, and evaluation. Capstones (six of them) are larger, open-ended projects where you combine everything into something closer to a real production system.

Foundations are meant to be read in numeric order — later modules assume you've seen the SDK calls, terminology, and patterns from earlier ones. Capstones are more self-directed: each one is a project brief, not a tutorial, and leans on whichever foundation modules are most relevant to it.

Set up VS Code as your reference IDE

This course uses VS Code as its reference editor throughout — screenshots, keyboard shortcuts, and setup steps assume it, though everything you build works the same in any editor. Start by creating a dedicated workspace folder for the course, then open it in VS Code (File > Open Folder, or code . from the integrated terminal if you have the code CLI installed).

A few extensions are worth installing before you write any code:

• Python extension (from Microsoft) — if you're following the Python track, this gives you IntelliSense, the interactive Python/Jupyter cell runner, and integrated debugging.
• An .env-aware setup — VS Code's integrated terminal and most Python/Node tooling read a .env file automatically once you load it with a library like python-dotenv or Node's built-in --env-file flag; you don't strictly need an extension, but a syntax-highlighting one makes .env files easier to read.
• Claude Code extension ("Claude Code" / "Claude for VS Code," optional) — install it from the Extensions view by searching "Claude Code." It adds a graphical panel for agentic coding sessions directly inside the editor, with inline diffs and file @-mentions. It's not required for this course, but several later modules touch on Claude Code as an example of a production agent harness, and having it installed lets you try those examples firsthand.

Inside your workspace folder, create a .env file to hold your API key, and make sure it's excluded from version control:

# from your workspace folder
touch .env
echo ".env" >> .gitignore

ANTHROPIC_API_KEY=sk-ant-api03-your-key-here

Create a free Anthropic Console account and API key

☺ Like you're 10: An API key is like a house key that also has your name tag on it — it lets you in, and it proves it was you who came in.

The Claude API is separate from the claude.ai chat product — it's the pay-as-you-go developer platform that claude.ai itself is built on. To use it from code, you need an account and an API key.

1. Go to platform.claude.com and sign in or create an account. (If you land on console.anthropic.com or docs.claude.com, that's expected — both now redirect here; it's a single combined Console-and-docs site.)
2. Navigate to Settings → API keys.
3. Click Create key. Give it a name, optionally scope it to a workspace, and choose an expiration — 3 hours, 1 day, 7 days, 30 days, a custom date, or "Never" for a key you plan to store in a secrets manager and rotate yourself.
4. Copy the key immediately. It starts with sk-ant- and the Console shows it in full exactly once, at creation — if you lose it, you'll need to create a new one.

Once you have a key, set it as an environment variable named exactly ANTHROPIC_API_KEY — this is what both official SDKs look for automatically:

export ANTHROPIC_API_KEY="sk-ant-api03-..."

Install Python or Node.js and the official SDKs

This course's examples run in Python and TypeScript/JavaScript side by side — use whichever you're more comfortable with, or follow both. You'll need Python 3.9+ or Node.js 18+ installed before continuing; VS Code's integrated terminal is the easiest place to check your version and run the install commands below.

pip install anthropic

npm install @anthropic-ai/sdk

Both SDKs read ANTHROPIC_API_KEY from the environment the moment you construct a bare client — no explicit key argument required. Here's the smallest possible "hello world" call in each language:

import anthropic

client = anthropic.Anthropic()  # reads ANTHROPIC_API_KEY from env

message = client.messages.create(
    model="claude-opus-5",
    max_tokens=1000,
    messages=[
        {
            "role": "user",
            "content": "What should I search for to find the latest developments in renewable energy?",
        }
    ],
)

for block in message.content:
    if block.type == "text":
        print(block.text)

import Anthropic from "@anthropic-ai/sdk";

const client = new Anthropic(); // reads ANTHROPIC_API_KEY from env

const message = await client.messages.create({
  model: "claude-opus-5",
  max_tokens: 1000,
  messages: [
    {
      role: "user",
      content: "What should I search for to find the latest developments in renewable energy?"
    }
  ]
});

for (const block of message.content) {
  if (block.type === "text") {
    console.log(block.text);
  }
}

Peel back either SDK and there's the same plain HTTP call underneath: a POST to /v1/messages carrying a model, a max_tokens cap, and a messages array of {role, content} objects, with your key riding along in a header. The SDK just saves you from writing that request by hand. The response's content is a list of typed blocks — this example only expects text, so it checks block.type == "text" before reading block.text; later modules cover the other block types (tool use, thinking) you'll see once you start using tools.

You're ready when…

Check off each of these before moving on to Module 1:

• You have a workspace folder open in VS Code, with a .env file (gitignored) holding your key.
• You've created an API key at platform.claude.com → Settings → API keys, and copied it somewhere safe.
• ANTHROPIC_API_KEY is set in your environment and readable by your terminal or script.
• You have Python 3.9+ or Node.js 18+ installed, plus the anthropic or @anthropic-ai/sdk package.
• You've run the hello-world call above and seen real text come back from Claude in your terminal.