# Anthropic

> Anthropic is an AI safety and research company building reliable, interpretable, and steerable AI systems. The company develops Claude, a large language model available through web interface and API, with emphasis on safe AI development and constitutional AI principles.

Anthropic's research focuses on AI alignment, interpretability, and responsible scaling. The company serves enterprise customers, developers, and the general public through Claude's various interfaces and capabilities.

## Products

- [Claude Code](https://www.anthropic.com/product/claude-code): Agentic coding system that understands codebases, executes multi-file changes, and completes development tasks autonomously.
- [Claude Opus](https://www.anthropic.com/claude/opus): Flagship large language model variant with advanced reasoning and task completion capabilities.
- [Claude Sonnet](https://www.anthropic.com/claude/sonnet): Balanced model variant optimized for speed and efficiency across a range of tasks.
- [Claude Haiku](https://www.anthropic.com/claude/haiku): Lightweight model variant designed for fast, cost-effective inference.

## Research & Safety

- [Constitution](https://www.anthropic.com/constitution): Claude's constitutional AI framework and principles guiding model behavior.
- [Responsible Scaling Policy](https://www.anthropic.com/responsible-scaling-policy): Anthropic's approach to safely scaling AI systems and managing risks at each stage.
- [RSP Updates](https://www.anthropic.com/rsp-updates): Latest developments and revisions to Anthropic's responsible scaling framework.
- [Transparency](https://www.anthropic.com/transparency): Anthropic's commitments and reports on AI safety, research, and operational transparency.
- [Research](https://www.anthropic.com/research): Published papers and technical research on AI safety, interpretability, and alignment.

## Engineering & Technical Guides

- [Building Effective Agents](https://www.anthropic.com/engineering/building-effective-agents): Best practices and patterns for constructing AI agents with Claude.
- [Claude Code Best Practices](https://www.anthropic.com/engineering/claude-code-best-practices): Guidance on using Claude Code effectively for software development tasks.
- [Claude Code Auto Mode](https://www.anthropic.com/engineering/claude-code-auto-mode): Documentation on autonomous operation capabilities in Claude Code.
- [Claude Code Sandboxing](https://www.anthropic.com/engineering/claude-code-sandboxing): Security and isolation mechanisms for Claude Code execution environments.
- [Claude Think Tool](https://www.anthropic.com/engineering/claude-think-tool): Extended reasoning capability allowing Claude to think through complex problems.
- [Contextual Retrieval](https://www.anthropic.com/engineering/contextual-retrieval): Technique for improving retrieval-augmented generation with context-aware embeddings.
- [Effective Context Engineering for AI Agents](https://www.anthropic.com/engineering/effective-context-engineering-for-ai-agents): Strategies for optimizing context windows and information flow in agent systems.
- [How We Contain Claude](https://www.anthropic.com/engineering/how-we-contain-claude): Technical measures and sandboxing approaches for safe Claude deployment.
- [Advanced Tool Use](https://www.anthropic.com/engineering/advanced-tool-use): Patterns for enabling Claude to effectively use external tools and APIs.
- [Code Execution with MCP](https://www.anthropic.com/engineering/code-execution-with-mcp): Integration of Model Context Protocol for code execution in Claude applications.

## Featured Projects

- [Project Glasswing](https://www.anthropic.com/glasswing): Anthropic's research initiative on AI safety and interpretability.
- [Claude on Mars](https://www.anthropic.com/features/claude-on-mars): Claude assisted NASA's Perseverance rover with the first AI-planned drive on another planet.
- [Project Deal](https://www.anthropic.com/features/project-deal): Marketplace experiment where Claude conducted autonomous buying, selling, and negotiation for employees.
- [Anthropic Interviewer](https://www.anthropic.com/about-anthropic-interviewer): Claude-powered research tool conducting conversational interviews at scale.
- [What 81,000 People Want from AI](https://www.anthropic.com/features/81k-interviews): Insights from large-scale user interviews about AI adoption, aspirations, and concerns.

## Learning & Community

- [Learn](https://www.anthropic.com/learn): Educational resources and guides for using Claude and understanding AI capabilities.
- [Engineering Blog](https://www.anthropic.com/engineering): Technical articles on AI agents, safety, evaluation, and system design.
- [Events](https://www.anthropic.com/events): Upcoming conferences, summits, and speaking engagements featuring Anthropic.

## Economic Research

- [Economic Index](https://www.anthropic.com/economic-index): Anthropic's research tracking economic indicators and AI's impact on productivity.
- [Economic Futures](https://www.anthropic.com/economic-futures): Program exploring long-term economic implications of AI advancement.
- [Economic Futures Program](https://www.anthropic.com/economic-futures/program): Details on Anthropic's initiative studying AI's economic effects.
- [Economic Futures Symposium Proposals](https://www.anthropic.com/economic-futures/symposium-proposals): Call for research proposals on AI and economic futures.

## Company

- [About](https://www.anthropic.com/company): Overview of Anthropic's mission, team, and organizational structure.
- [Careers](https://www.anthropic.com/careers): Open positions and information about working at Anthropic.

## Optional

- [News](https://www.anthropic.com/news): Press releases and announcements from Anthropic.
- [AI for Science Program Rules](https://www.anthropic.com/ai-for-science-program-rules): Guidelines for Anthropic's initiative supporting scientific research with AI.
- [Coordinated Vulnerability Disclosure](https://www.anthropic.com/coordinated-vulnerability-disclosure): Process for reporting security vulnerabilities in Claude systems.
