vibe engineering
In plain wordsBuilding serious software with AI coding assistants while still taking full responsibility for its quality — the careful, professional opposite of vibe coding.
The careful, professional use of coding agentscoding agent, n.: An AI agent that writes and fixes software by itself: it reads a codebase, edits files, runs commands and tests, and keeps working until a feature is built or a bug is fixed. and other AI tools: experienced engineers let the tools speed up work on production software but still answer for the quality of what ships. Simon Willison, who proposed the name as a counterweight to vibe codingvibe coding, n.: Programming by conversation: you ask an AI for what you want, try out what it produces and ask again until it works, trusting the code without really reading it. (accepting AI-written code without understanding it), argued that agents reward the habits of strong engineering teams rather than replacing them, among them thorough automated tests, written plans and documentation, disciplined version control and code review, and hands-on testing, plus newer skills such as judging which tasks to hand off, re-estimating how long work takes and managing agents almost like staff.
Vibe engineering means the agent writes the code, but you still own every line that ships.
Origin
Proposed by Simon Willison in a blog post of 7 October 2025 as a deliberately cheeky counterpart to vibe coding.
Usage note
In 2026 'agentic engineering' (see agentic codingagentic coding, n.: Writing software by directing AI agents that plan, edit and test code themselves, with the human acting more as a reviewer and manager than as the person typing every line.) became a more common name for the same disciplined practice, including in Willison's own writing.
Related
- vibe codingvibe coding, n.: Programming by conversation: you ask an AI for what you want, try out what it produces and ask again until it works, trusting the code without really reading it.
- agentic codingagentic coding, n.: Writing software by directing AI agents that plan, edit and test code themselves, with the human acting more as a reviewer and manager than as the person typing every line.
- coding agentcoding agent, n.: An AI agent that writes and fixes software by itself: it reads a codebase, edits files, runs commands and tests, and keeps working until a feature is built or a bug is fixed.
- evals-driven developmentevals-driven development, n.: Building an AI product by first writing tests that show what good answers look like, then accepting changes to prompts, models or code only when the test scores improve.
- Ralph Wiggum loopRalph Wiggum loop, n.: A blunt way of using AI coding assistants: feed the same instructions to the AI again and again in an endless loop until the software gets built.
- dark factorydark factory, n.: A way of building software in which AI agents write, test and ship all the code and no human writes or reads it, like a factory run with the lights off.
- vibe codingvibe coding, n.: Programming by conversation: you ask an AI for what you want, try out what it produces and ask again until it works, trusting the code without really reading it.
Sources
- Willison, S. (2025). Vibe engineering. Simon Willison's Weblog, 7 October 2025. simonwillison.net/2025/Oct/7/vibe-engineering
Mentioned in
vibe codingvibe coding, n.: Programming by conversation: you ask an AI for what you want, try out what it produces and ask again until it works, trusting the code without really reading it., Simon WillisonSimon Willison, prop. n.: A British programmer and blogger, co-creator of the Django web framework, whose widely read writing on language models coined 'prompt injection' and the 'lethal trifecta'., spec-driven developmentspec-driven development, n.: Building software with AI coding agents by first writing a detailed description of what it should do, which the AI then follows to plan and write the code.