Claude Code Now Works While You’re Away — How to Run It Async

2026-06-10 · Source: Towards AI - Medium · Field: Technology & Digital — Software Development & Engineering, Artificial Intelligence & Machine Learning · Depth: Intermediate, medium

Summary

Claude Code's 2026 build, following its 2.1 release, introduces significant asynchronous capabilities, allowing developers to delegate complex coding tasks to AI agents that run independently. This update enables operations like 40-file refactors, pull request triage, and codebase bug hunts to proceed without constant human supervision. The platform, which achieved approximately \$1 billion in annualized revenue within six months, now supports "workflows" for parallel task execution, an "advisor" model for pre-execution plan review, and "subagents" for focused exploration. Developers can manage tasks remotely via claude.ai, schedule "routines" for automated daily tasks, and use /loop for continuous monitoring. Crucially, the system includes safety features like /branch, /rewind, and /recap, alongside fallbackModel configurations to prevent job stalls. While powerful, workflows are noted as compute-intensive, and invisible work requires diligent human verification.

Key takeaway

For AI Engineers managing large codebases or frequent, repetitive tasks, Claude Code's new async capabilities fundamentally change workflow delegation. You can now initiate complex refactors or bug hunts, then close your laptop, trusting the AI fleet to proceed. This frees your attention for higher-value problems. Always integrate human verification steps, like reviewing diffs or using the advisor, to mitigate risks associated with unsupervised work.

Key insights

Claude Code's async features enable AI agents to autonomously execute complex development tasks, freeing human developers.

Principles

• Delegate work, never verification.
• Parallel agents outperform single agents.
• Second opinions prevent costly errors.

Method

Use "use a workflow" to trigger parallel agent execution; employ an "advisor" for plan review before execution.

In practice

• Migrate logging libraries across 40 files.
• Triage overnight pull requests automatically.
• Monitor production deploys with /loop.

Topics

• Claude Code
• Asynchronous AI
• Multi-Agent Systems
• Code Refactoring
• Workflow Automation
• AI Safety

Best for: Software Engineer, AI Engineer, Director of AI/ML

Related on AIssential

• shanraisshan / claude-code-best-practice — Claude Code's agentic capabilities, including subagents, commands, and skills, significantly enhance developer productivity and workflow automation.
• How Anthropic’s Claude Code is Changing Enterprise Workflows — Anthropic's new Claude Opus 4.8 and Dynamic Workflows enable autonomous multiagent AI for complex enterprise coding tasks.
• v2.1.97 — Claude Code prioritizes security, performance, and extensibility through continuous iterative improvements and new feature introductions.
• v2.1.80 — Recent Claude Code updates enhance agent capabilities, user experience, and system stability through new features and critical bug fixes.
• Claude Code Adds Dynamic Workflows for Parallel Agent Coordination — Claude Code's Dynamic Workflows enable parallel AI agent coordination for complex software engineering tasks.
• Claude Code MASSIVE Update! Claude Code OS, Computer Use, /Schedule, & More! — Claude Code now offers autonomous computer control and remote task scheduling, enhancing AI agent capabilities.

Open in AIssential →

Editorial summary, takeaway, and curation by AIssential. Original article published by Towards AI - Medium.