AI / Claude Models Basics Interview Questions
What is Claude Code and how does it differ from using Claude directly via the API?
Claude Code is Anthropic's agentic coding tool — a command-line interface (CLI) and SDK that allows Claude to work autonomously on coding tasks in your terminal, with direct access to your file system, git, and development tools.
| Feature | Claude Code | Claude API (direct) |
|---|---|---|
| Interface | CLI tool in your terminal | HTTP REST API / SDK |
| Setup | npm install -g @anthropic-ai/claude-code | pip install anthropic or npm install @anthropic-ai/sdk |
| File access | Yes — reads/writes files in your project | No — you pass content in the prompt |
| Tool execution | Yes — can run commands, tests, git operations | Only if you build tool use yourself |
| Use case | Coding assistance, refactoring, debugging, code generation | Custom apps, chatbots, data processing |
| IDE integration | VS Code, JetBrains plugins available | N/A |
# Installing Claude Code npm install -g @anthropic-ai/claude-code # Using Claude Code in your terminal cd my-project claude-code "Add error handling to all async functions in src/" # Claude Code can: # - Read and write files in your project # - Run tests and build commands # - Make git commits # - Navigate and understand large codebases # - Work through multi-step tasks autonomously
Claude Code is built on the same underlying Claude models (using Opus-tier models for best results) but provides a ready-made agentic environment with tools already wired up. The API requires you to build the tool use and agentic loop yourself.
What is the key capability Claude Code has that the raw Claude API does not provide out of the box?
How do you install Claude Code?
More Related questions...
