Live blog: Code w/ Claude 2026

2026-05-06 · Source: Simon Willison's Weblog · Field: Technology & Digital — Artificial Intelligence & Machine Learning, Software Development & Engineering, Robotics & Autonomous Systems · Depth: Intermediate, medium

Summary

Anthropic's "Code w/ Claude 2026" keynote on May 6, 2026, highlighted significant platform advancements and developer tools rather than new model releases. Chief Product Officer Ami Vora noted a 17x year-on-year API volume increase and announced doubled rate limits for Pro, Max, and Enterprise customers on Claude Code and the API, facilitated by a partnership with SpaceX's Colossus data center. Dianne Na Penn, Head of Product for Research, emphasized enhanced tool use, long context, and multi-agent coordination, citing Opus 4.7's capabilities in visual design and planning. Katelyn Lesse and Angela Kiang introduced new Claude Managed Agents features: multi-agent orchestration, "Outcomes" for goal-setting, and "Dreaming" for self-improvement and memory creation. Cat Wu, Head of Product for Claude Code, detailed new surfaces like the Desktop app, Code Review, Remote Agents, CI auto-fix, and Security Reviews, noting customers like Mercado Libre aiming for 90% autonomous coding. Boris Cherny demonstrated async coding with Claude Desktop and "Routines" for automated PR generation.

Key takeaway

For engineering leaders evaluating AI integration, Anthropic's focus on multi-agent orchestration, self-improving agents, and developer tooling like Claude Code Desktop and "Routines" suggests a path to significantly increased developer velocity and autonomous coding. You should explore these new features, particularly "Dreaming" and "Outcomes," to potentially achieve "frontier model quality at 5x lower cost" and aim for higher percentages of autonomous coding within your teams.

Key insights

Anthropic is enhancing developer productivity and model autonomy through advanced agentic capabilities and platform integrations.

Principles

• Multi-agent orchestration improves complex task completion.
• Automated evaluations and simple scaffolding maximize model utility.

Method

Claude Managed Agents use multi-agent orchestration, "Outcomes" for success criteria, and "Dreaming" for self-improvement by inspecting past sessions and creating new memories.

In practice

• Utilize Claude Managed Agents for complex, multi-step tasks.
• Explore "Routines" for async code automation and PR generation.
• Implement "Code Review" and "CI auto-fix" to streamline development.

Topics

• Anthropic
• Claude Code
• Managed Agents
• Multi-agent Orchestration
• AI Code Generation

Best for: CTO, VP of Engineering/Data, Director of AI/ML, AI Engineer, Machine Learning Engineer, Software Engineer

Related on AIssential

• Code with Claude: The 5 biggest updates explained — Anthropic's new features enhance agent autonomy, collaboration, and memory, alongside increased usage limits.
• Anthropic just disclosed that Claude now writes more than 80% of the production code it merges — AI agents are dramatically increasing developer productivity and code generation capabilities, while raising recursive self-improvement concerns.
• Anthropic launches AI-powered Code Review for Claude Code — AI-powered code review can significantly accelerate development cycles by focusing on logical errors in AI-generated code.
• Anthropic Introduces Routines for Claude Code Automation — Anthropic's Routines automate coding workflows via scheduled, API, or event-driven triggers on cloud infrastructure.
• [AINews] Anthropic raises $965B Series H, releases Opus 4.8 and Dynamic Workflows/ultracode — Anthropic's Opus 4.8 and Dynamic Workflows advance AI capabilities, particularly in agentic coding, supported by significant new funding.
• Claude Code, Codex and Agentic Coding #7: Auto Mode — Auto Mode enhances agentic coding safety by selectively requiring human approval for high-risk actions.

Open in AIssential →

Editorial summary, takeaway, and curation by AIssential. Original article published by Simon Willison's Weblog.