Why This Matters

The claude.ai interface is more than a chat box. It has Projects (knowledge containers), Artifacts (interactive content), and a model picker. Knowing the layout multiplies your leverage.

Concept Explained

The claude.ai interface has four main areas: (1) chat panel (center) — the conversation; (2) Projects sidebar (left) — your knowledge containers; (3) Artifacts panel (right, opens when Claude generates one) — interactive content; (4) model picker (top-right) — switch between Opus/Sonnet/Haiku.

How It Works

Conversations live in the chat panel. Each conversation can be standalone or attached to a Project (for recurring context). When Claude generates code, documents, or SVGs, it can render them as Artifacts — interactive panels you can edit, share, or download. The model picker lets you switch models mid-conversation.

Step-by-Step Tutorial

1. Tour the interface

Open claude.ai. Identify: chat panel (center), Projects sidebar (left), model picker (top-right). The Artifacts panel opens on the right when Claude generates one.

2. Start a new conversation

Click "New chat" or press the keyboard shortcut. Each conversation has independent context. Start fresh per task.

3. Pick a model

Use the dropdown top-right. Default is Sonnet 4. Switch to Opus 4 for complex reasoning; Haiku for fast/simple.

4. Create a Project

Click "Projects" in the left sidebar → New Project. Name it. Add files or text as "knowledge" — Claude uses this for all conversations in that Project.

5. Generate an Artifact

Ask Claude: "Create an interactive HTML calculator." Claude should generate it as an Artifact — a live preview on the right side that you can interact with, edit, and share.

Real-World Example

A marketing manager set up a Project called "Brand Voice" with their style guide, past campaign examples, and tone preferences. Every new blog post draft started in that Project — Claude automatically adopted the brand voice without re-prompting. Saved 30 minutes per draft.

Example Prompts / Commands / Code

Layout:
+-------------------------------------------------------+
|  [Projects]  |  [Chat: title]            [Sonnet 4 v] |
|  - Brand     |                                         |
|  - Research  |  User: Explain X                         |
|  - ...       |                                         |
|              |  Claude: [response]                     |
|              |                                         |
|              |  [Artifact panel opens on right when     |
|              |   Claude generates code/SVG/etc.]        |
|              |                                         |
|  [New chat]  |  [Type a message...] [attach] [send]    |
+-------------------------------------------------------+

1. Click "Projects" in left sidebar
2. Click "New Project"
3. Name: "Brand Voice"
4. Add knowledge:
   - Upload: style-guide.pdf
   - Paste text: example blog posts
   - Write: "Tone: professional but warm"
5. Save
6. All conversations in this Project now have this context.

Common Mistakes

• Not using Projects — re-pasting context every conversation.
• Ignoring Artifacts — Claude's code generation is significantly better in Artifacts.
• Always using default model — Opus 4 is worth it for complex reasoning.
• Mixing unrelated tasks in one conversation — context bleeds.

Best Practices

• One conversation per task.
• Use Projects for recurring context (team, codebase, brand).
• Use Artifacts for code, diagrams, and interactive content.
• Pick the right model per task: Sonnet default, Opus for hard, Haiku for fast.
• Name conversations meaningfully for future search.

Troubleshooting

Practical Exercise

Frequently Asked Questions

Further Reading

• What are artifacts
• Projects

Official References

• Artifacts

Related lessons: BE-04, BE-06, BE-10