Best of Anthropic — September 2025

  1. 1
    Article
    Avatar of bytebytegoByteByteGo·32w

    How Anthropic Built a Multi-Agent Research System

    Anthropic built a multi-agent research system using an orchestrator-worker pattern with a Lead Researcher agent coordinating specialized subagents and a Citation Agent for accuracy. The system outperformed single-agent setups by 90% through parallel processing and dynamic adaptation, though it consumes 15x more tokens. Key engineering principles include proper prompt design, delegation strategies, effort scaling, and parallelization. Production challenges involve managing stateful agents, debugging non-deterministic behavior, and handling deployments without breaking running tasks.

  2. 2
    Article
    Avatar of wheresyouredWhere's Your Ed At·33w

    Why Everybody Is Losing Money On AI

    Generative AI companies are fundamentally unprofitable, with every major player from OpenAI to Anthropic losing billions of dollars annually. The economics are broken: companies like Cursor send 100% of their revenue to model providers, Perplexity spends 164% of revenue on compute costs, and OpenAI lost $5 billion in 2024 despite being the market leader. Inference costs are increasing rather than decreasing, making it impossible for AI startups to achieve profitability even with usage-based pricing models.

  3. 3
    Article
    Avatar of vibecodingVibe Coding·30w

    Vibe coding just got an upgrade with Claude Sonnet 4.5

    Anthropic released Claude Sonnet 4.5, their latest and most capable coding model that reportedly outperforms the previous Opus 4.1 version.

  4. 4
    Article
    Avatar of weprodevWeProDev·33w

    Anthropic AI courses

    Anthropic has launched an educational platform offering structured courses on AI and machine learning concepts. The platform provides learning resources for developers and professionals looking to understand and work with AI technologies, particularly focusing on Anthropic's approach to AI development and safety.

  5. 5
    Article
    Avatar of collectionsCollections·30w

    Anthropic Releases Claude Sonnet 4.5: State-of-the-Art AI for Coding

    Claude Sonnet 4.5 achieves a 77.2% score on SWE-bench, positioning it as a leading AI coding model. Available on GitHub Copilot, VS Code, and JetBrains IDEs, it features enhanced memory management for tasks up to 30 hours, improved tool orchestration, and autonomous task handling. The model integrates with Snowflake Cortex AI and Amazon Bedrock for enterprise deployment, with pricing at $3 per million input tokens and $15 per million output tokens. Safety improvements include reduced sycophancy and better resistance to prompt injection attacks.

  6. 6
    Video
    Avatar of t3dotggTheo - t3․gg·32w

    It's not just you (Claude did get dumber)

    Anthropic's Claude models experienced significant quality degradation and reliability issues over several months, with multiple bugs affecting model intelligence going unnoticed and unacknowledged for extended periods. The company's lack of transparency, poor infrastructure reliability, and prioritization of research over user experience has created trust issues. Alternative providers like Google Vertex and AWS Bedrock offer better uptime for Claude models, and services like OpenRouter provide failover capabilities to avoid single-provider dependency.

See all Anthropic archives