Beginner

Introduction to Claude

Discover what Claude is, what it can do, and why it stands out among AI assistants. This lesson introduces the fundamentals you need before diving deeper.

What is Claude?

Claude is an AI assistant created by Anthropic, a safety-focused AI company founded in 2021. Claude is designed to be helpful, harmless, and honest — a set of principles that guide how it interacts with users.

Unlike simple chatbots, Claude is a large language model (LLM) capable of understanding and generating human-like text across a wide range of tasks. You can talk to Claude through claude.ai, the Anthropic API, or tools like Claude Code.

💡
Good to know: Anthropic was founded by former members of OpenAI, including Dario and Daniela Amodei. The company's mission centers on AI safety research and building reliable, interpretable, and steerable AI systems.

Key Capabilities

Claude is a versatile AI that excels across many domains. Here are its core capabilities:

  • Conversation: Claude can engage in natural, multi-turn conversations. It remembers context within a conversation and can discuss complex topics in depth.
  • Analysis: Given data, documents, or code, Claude can analyze patterns, identify issues, summarize findings, and provide structured insights.
  • Coding: Claude can write, debug, explain, and refactor code in dozens of programming languages including Python, JavaScript, TypeScript, Rust, Go, Java, C++, and more.
  • Writing: From emails to essays to technical documentation, Claude produces clear, well-structured written content tailored to your audience and tone.
  • Math & Reasoning: Claude handles mathematical problems, logical reasoning, and step-by-step problem solving, showing its work along the way.
  • Vision: Claude can analyze images, read text from screenshots, interpret charts and diagrams, and describe visual content in detail.
Example Conversation 
User:
Explain the difference between a stack and a queue
in simple terms.

Claude:
Think of them like this:

Stack - Like a stack of plates. You add plates
on top and remove from the top. The last plate you
put on is the first one you take off.
This is called LIFO (Last In, First Out).

Queue - Like a line at a coffee shop. The first
person in line is the first person served.
This is called FIFO (First In, First Out).

Claude's Values: Helpful, Harmless, Honest

Anthropic designed Claude around three core principles, often referred to as HHH:

Helpful

Claude aims to be genuinely useful. It tries to understand what you really need, provides thorough answers, and follows instructions carefully. If your request is ambiguous, Claude will ask for clarification rather than guessing incorrectly.

Harmless

Claude is designed to avoid producing harmful, dangerous, or unethical content. It will decline requests that could cause harm while explaining why and often suggesting safer alternatives. This is guided by Anthropic's Constitutional AI approach, where Claude is trained with a set of principles rather than just rules.

Honest

Claude strives for accuracy and transparency. When it is uncertain, it says so. When it does not know something, it admits it rather than fabricating an answer. Claude distinguishes between facts and opinions, and will tell you when a topic is outside its knowledge or when its information might be outdated.

✅
Key takeaway: Claude's honesty about uncertainty is one of its most valuable traits. If Claude says "I'm not sure about this," that is often more helpful than a confident but wrong answer from another system.

How Claude Differs from Other AI Assistants

While several AI assistants exist today, Claude has distinct characteristics:

Feature Claude's Approach
Safety focus Built on Constitutional AI principles with safety as a core design goal, not an afterthought
Long context Supports up to 200K tokens of context — enough to process entire books or large codebases at once
Instruction following Excels at following complex, multi-step instructions precisely and consistently
Nuanced reasoning Handles ambiguity well, considers multiple perspectives, and provides balanced analysis
Transparency Openly acknowledges limitations, uncertainty, and knowledge boundaries
Multiple models Offers Haiku, Sonnet, and Opus to match different needs for speed, cost, and capability

What You'll Learn in This Course

This course takes you from complete beginner to confident Claude user. Here is the path ahead:

  1. Getting Started

    Access Claude through different interfaces and have your first conversation.

  2. Understanding Models

    Learn the differences between Haiku, Sonnet, and Opus and when to use each one.

  3. Prompting Fundamentals

    Master the basics of writing effective prompts with clear instructions and examples.

  4. Advanced Techniques

    Explore chain-of-thought reasoning, system prompts, XML tags, and prompt chaining.

  5. Real-World Use Cases

    See Claude applied to writing, coding, analysis, research, support, and education.

  6. API Integration

    Make API calls with curl, Python, and JavaScript to integrate Claude into your apps.

  7. Best Practices

    Learn production tips, cost optimization, safety guidelines, and common pitfalls.

💡 Try It: Your First Prompt Idea

Before moving on, think about what you would ask an AI assistant. Write a prompt idea below — you will use it when we get to the Getting Started lesson.

Your prompt idea will help you practice in the next lesson. Think about a real task you want help with!
📚
Prerequisites: This course assumes no prior AI experience. All you need is curiosity and a web browser. A free claude.ai account is helpful but not required to follow along.