Prerequisite Track 0B — You've Never Used Claude

You may already be comfortable with technology — maybe you write code, maybe you've used other AI tools — but you've never actually opened Claude. This is your fast orientation. The goal is simple: in under an hour you'll have an account, you'll know which Claude "surface" to reach for, and you'll have run a productive first session.

Total time: about 45–60 minutes. This is a sprint, not a study track. When you're done you'll know where to go next.

Step 1 — Create your account · ⏱ ~5 min

Go to claude.ai and sign up (free tier is fine to start).
Optional but recommended: install the desktop app and the mobile app so Claude is one click away.
If you'll be doing developer work later, skim the setup-checklist now so you know what's coming — but you don't need any of it for this session.

Step 2 — Learn the surfaces (so you reach for the right one) · ⏱ ~5 min

The same Claude model family powers several different surfaces. Pick the one that fits the job:

claude.ai / desktop / mobile  → chat, Projects, Artifacts        (start here)
Claude Cowork                 → Claude works alongside you on a task
Claude Code                   → coding agent in your terminal/IDE  (developers)
Claude API                    → build Claude into your own apps    (developers)
Bedrock / Vertex AI           → Claude inside AWS / Google Cloud   (enterprises)

For this orientation, everything happens in claude.ai. The developer surfaces come later, in Tracks 3–7.

Step 3 — Your 15-minute first session · ⏱ ~15 min

Open a new chat and actually do these, in order. Don't just read them — type them.

A real task with context (4 min). Don't search; converse. Try:

"Write a 120-word friendly email to my team announcing that our Monday standup is moving to 10am. Warm tone, no jargon, end with a clear call to action." Notice how much better this is than typing "standup email."
Iterate, don't restart (3 min). Reply to that answer with: "Make it shorter and a little funnier." Watch it refine instead of starting over. This is the core skill.
Upload a document (4 min). Drag in any PDF you have and ask: "What are the three most important points, and what page is each on?" This is how you get grounded, trustworthy answers instead of guesses.
Try an Artifact (2 min). Ask: "Make me a reusable weekly meal-planning checklist." Claude builds it in a side panel you can edit and keep — that's an Artifact.
See a limit on purpose (2 min). Ask about something very recent or very specific to your life. Notice where Claude hesitates or could be wrong. This is the trust-but-verify habit: for facts, dates, and numbers, you check or you provide the source.

✅ If you did all five, you've already used the four most important features: conversation-with-context, iteration, file grounding, and Artifacts — plus the verification mindset.

Step 4 — Lock it in · ⏱ ~25 min

Course: claude-101 — the full front-door course. It deepens everything you just tried (the four prompt levers, model selection, Projects, Artifacts, safe use).
Project: p01-first-conversations — run your first 10 real conversations and keep a before/after of one prompt you improved.

You're ready to move on when…

You have a claude.ai account and have run a multi-turn conversation.
You can refine an answer with a follow-up instead of re-typing the whole request.
You've uploaded a document and asked questions grounded in it.
You can name the main surfaces and say which one fits a given job.

Where to go next

Now route yourself by what you want to do:

Comfortable with tech / want to build? Head into the tracks in order, starting at track-1-claude-foundations and continuing through prompting, the API, Claude Code, MCP, agents, and deployment. The full map is in README.
Not technical, and you'd like a gentler, no-code path? Switch over to track-0a-non-technical. It covers only the non-developer courses and reassures you at every step — a calmer ride than diving straight into the tracks.

Tick this off in progress when you're done.