Boris Cherny, the Anthropic engineer who created Claude Code, says his personal development workflow now revolves around fleets of autonomous AI agents that run unattended for hours at a time — a description he frames as the new default for software engineering rather than an experiment. Cherny, who leads the Claude Code effort at Anthropic, described running hundreds and sometimes thousands of agents overnight, each executing continuously for anywhere from 5 to 20 hours, and characterized this as simply how engineering is done now.
Continue reading
The rest of this article is for AI News Blitz readers. Choose an option below to keep reading.
Already purchased? Sign in✓ Signed in — this article isn’t included in your current plan.