4 What you need before you start

5 What you need before you start

This is the pre-flight check. Skim it; install what you don’t have; come back when the ticks below are all green. The rest of Part 0 walks you through each item in turn, so do not panic if half of this list looks unfamiliar.

5.1 What you will get from this chapter

A short list of accounts and software the book assumes you have.
A sense of which items are essential from page one and which can wait.
A pointer to the right Part 0 chapter for each item.

5.2 The list

You need six things. Three of them are accounts. Three of them are software.

Accounts.

A GitHub account. Free is fine. We use it to store your work and, in the capstone, to publish a small website. Sign up at github.com.
An AI chat account with at least one frontier provider. This edition was written against Claude (Pro tier) and ChatGPT (Plus tier). Either one will do for almost every chapter. The free tiers also work for a surprising amount of the book; you will hit limits in Parts III and VI.
A Google or Microsoft account if you want to try the free academic tiers some providers offer. Optional.

Software.

Python 3.11 or newer. Anything that can import pandas and run a small script.
A terminal. macOS Terminal, Windows Terminal (or PowerShell), or any Linux shell.
A code editor. VS Code is the default in this book; RStudio works too. Pick one and stick with it.

5.3 What you actually need on day one

If you only do three things before reading on, do these:

Make sure Python runs. Open a terminal and type python --version. You should see Python 3.11.x or later.
Sign up for one AI chat account. Five minutes. You can do the entire first half of the book with just this.
Sign up for GitHub. Two minutes. You will not use it for a few chapters; setting it up early is painless.

The rest can wait until you need it. Terminal commands come up in the next chapter. VS Code waits until you start writing real scripts. The CLI agent waits until Part III. API keys wait until Part VI.

5.4 A note for readers on locked-down laptops

If your computer is managed by your university’s IT and you cannot install software freely, you have two options. First, ask. Many universities will whitelist Python, VS Code, and Git for students who request it. Second, use a cloud workspace — Posit Cloud for R-flavoured work, Google Colab for Python notebooks. Cloud setups will get you through chapters 1–10 without local installs. They struggle with Part III, where the book asks you to use a CLI agent on real files; for that, you really do need a machine you control.

5.5 Where AI helps · Where AI bluffs

Helps. Diagnosing a failed install — paste the error into your chat AI and it usually finds the cause faster than Googling.

Bluffs. Telling you a package version exists when it does not. Recommending a deprecated install command from a 2022 blog post that lived in its training data. If a chat AI tells you to run a command and the command fails, do not run a second similar command on its say-so. Read the actual error.

5.6 Keep this with you, not the AI

The decision of whether to install something on a managed laptop. Ask your IT, not your AI.
Whether your data is sensitive enough to care about a free tier’s privacy policy. Read the policy yourself for the one provider you use.
The choice between local Python and a cloud notebook. It is a real trade-off. Make it deliberately.

5.7 Try this

Open a terminal right now. Type python --version. If it works, you are ready for Chapter 0d. If it does not — different command on Windows? — that is fine, Chapter 0g — terminal basics will get you sorted.

5.8 AI and me

After reading this chapter, write a one-line answer to each:

How did AI support me here?
How did AI fail me?
How did AI extend me?

5.9 Where to go next

If you have never used Git, go to Git for analysts. If you have, jump to Python environment and skim from there.

--- title: "What you need before you start" --- # What you need before you start This is the pre-flight check. Skim it; install what you don't have; come back when the ticks below are all green. The rest of Part 0 walks you through each item in turn, so do not panic if half of this list looks unfamiliar. ## What you will get from this chapter - A short list of accounts and software the book assumes you have. - A sense of which items are essential from page one and which can wait. - A pointer to the right Part 0 chapter for each item. ## The list You need six things. Three of them are accounts. Three of them are software. **Accounts.** - A **GitHub account.** Free is fine. We use it to store your work and, in the capstone, to publish a small website. Sign up at [github.com](https://github.com). - An **AI chat account** with at least one frontier provider. This edition was written against Claude (Pro tier) and ChatGPT (Plus tier). Either one will do for almost every chapter. The free tiers also work for a surprising amount of the book; you will hit limits in Parts III and VI. - A **Google or Microsoft account** if you want to try the free academic tiers some providers offer. Optional. **Software.** - **Python 3.11 or newer.** Anything that can `import pandas` and run a small script. - A **terminal.** macOS Terminal, Windows Terminal (or PowerShell), or any Linux shell. - A **code editor.** VS Code is the default in this book; RStudio works too. Pick one and stick with it. ## What you actually need on day one If you only do three things before reading on, do these: 1. Make sure Python runs. Open a terminal and type `python --version`. You should see `Python 3.11.x` or later. 2. Sign up for one AI chat account. Five minutes. You can do the entire first half of the book with just this. 3. Sign up for GitHub. Two minutes. You will not use it for a few chapters; setting it up early is painless. The rest can wait until you need it. Terminal commands come up in the next chapter. VS Code waits until you start writing real scripts. The CLI agent waits until Part III. API keys wait until Part VI. ## A note for readers on locked-down laptops If your computer is managed by your university's IT and you cannot install software freely, you have two options. First, ask. Many universities will whitelist Python, VS Code, and Git for students who request it. Second, use a cloud workspace — [Posit Cloud](https://posit.cloud) for R-flavoured work, [Google Colab](https://colab.research.google.com) for Python notebooks. Cloud setups will get you through chapters 1–10 without local installs. They struggle with Part III, where the book asks you to use a CLI agent on real files; for that, you really do need a machine you control. ## Where AI helps · Where AI bluffs **Helps.** Diagnosing a failed install — paste the error into your chat AI and it usually finds the cause faster than Googling. **Bluffs.** Telling you a package version exists when it does not. Recommending a deprecated install command from a 2022 blog post that lived in its training data. If a chat AI tells you to run a command and the command fails, do not run a second similar command on its say-so. Read the actual error. ## Keep this with you, not the AI - The decision of *whether* to install something on a managed laptop. Ask your IT, not your AI. - Whether your data is sensitive enough to care about a free tier's privacy policy. Read the policy yourself for the one provider you use. - The choice between local Python and a cloud notebook. It is a real trade-off. Make it deliberately. ## Try this Open a terminal right now. Type `python --version`. If it works, you are ready for [Chapter 0d](00d-python.qmd). If it does not — different command on Windows? — that is fine, [Chapter 0g — terminal basics](11-terminal-basics.qmd) will get you sorted. ## AI and me After reading this chapter, write a one-line answer to each: 1. How did AI **support** me here? 2. How did AI **fail** me? 3. How did AI **extend** me? ## Where to go next If you have never used Git, go to [Git for analysts](00b-git.qmd). If you have, jump to [Python environment](00d-python.qmd) and skim from there.