Claude Code Creator Rejects 'Vibe Coding' Phrase
According to Business Insider, Boris Cherny, head of Claude Code, told Business Insider at Anthropic's Code with Claude conference in San Francisco on May 6, 2026, that he is "sick of" the phrase "vibe coding" and has begun searching for alternative terminology. Business Insider reports that Cherny asked Anthropic's Claude chatbot for suggestions; the chatbot suggested "agentic engineering," a phrase also used by Andrej Karpathy. Business Insider notes that Karpathy coined "vibe coding" in early 2025 and that the Collins Online Dictionary named "vibe coding" its word of the year in November. Business Insider also reports Anthropic's documentation describes Claude Code as an "AI-powered coding assistant," and the article characterizes both Anthropic's and OpenAI's code tools as generating substantial revenue and producing millions of lines of code.
What happened
Business Insider reports that, at Anthropic's Code with Claude conference in San Francisco on May 6, 2026, Boris Cherny, head of Claude Code, described himself as "sick of" the phrase "vibe coding" and has begun searching for alternative names. Business Insider reports that Cherny asked Anthropic's Claude chatbot for ideas and that Claude suggested "agentic engineering," a phrase also associated with Andrej Karpathy. Business Insider reports that Karpathy originally used "vibe coding" in early 2025 and that the Collins Online Dictionary named "vibe coding" its word of the year in November. Business Insider also reports that Anthropic's public documentation labels Claude Code an "AI-powered coding assistant," and frames both Anthropic's and OpenAI's coding products as generating substantial revenue and producing millions of lines of code.
Editorial analysis - technical context
Terminology such as "vibe coding" and "agentic engineering" reflects how practitioners and commentators package a set of capabilities: large-model code generation, automated agents, and assisted developer workflows. Industry reporting often uses short, evocative phrases to capture complex toolchains that combine code models, prompt engineering, and orchestration layers. For practitioners, these shorthand labels can shape expectations about debugging, reproducibility, and test coverage when AI produces or composes code.
Industry context
Observed patterns show that naming contests tend to follow visible product adoption cycles. When a term becomes mainstream, third-party documentation, press coverage, and conference talks reinforce it; conversely, pushback from prominent engineers or evangelists can spur alternative framings. Neutral naming changes do not alter underlying technical risks or integration work, but they can influence how engineering teams justify tooling choices to stakeholders.
What to watch
Industry observers should track whether alternative terms gain traction in developer documentation, conference programs, or social discourse. Watch Anthropic's public docs and blog posts for any shifts in phrasing, mentions of Claude suggestions in follow-up coverage, and how other prominent figures such as Andrej Karpathy reference these labels in talks or posts. For practitioner communities, monitor signal changes in expectations around agentic behavior, testing expectations, and recommended guardrails.
Scoring Rationale
This is a cultural and product-marketing story rather than a technical breakthrough. It matters to developer communities and product teams for messaging and framing, but it does not change models, APIs, or infrastructure.
Practice interview problems based on real data
1,500+ SQL & Python problems across 15 industry datasets — the exact type of data you work with.
Try 250 free problems
