“Loop engineering” is a hot buzzphrase after mentions of it by Boris Cherny (Claude Code’s creator) and Peter Steinberger (OpenClaw's creator) went...
Andrew Ng describes three loops for AI-assisted product development: the agentic coding loop (AI iterates on code), the developer feedback loop (human steers product), and the external feedback loop (user testing informs vision). He notes engineers are taking on product management roles as AI accelerates development.
Background
Andrew Ng is a highly influential AI researcher and educator, co-founder of Coursera and founder of DeepLearning.AI. "Loop engineering" refers to structuring how AI coding agents (software that writes code autonomously) repeatedly test and refine their output without human intervention. Boris Cherny created Claude Code (Anthropic's coding agent); Peter Steinberger created OpenClaw (another AI coding tool). The three loops Ng describes — agentic coding loop (AI iterates on code in minutes), developer feedback loop (human steers in tens of minutes to hours), and external feedback loop (user testing over days) — represent a workflow where humans increasingly act as product managers while AI handles implementation. This matters because AI coding agents are dramatically accelerating software development, shifting the engineer's role from writing code line-by-line to defining product vision and reviewing AI-generated work.