agentic coding
In plain wordsWriting 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.
A mode of software development in which a programmer delegates multi-step programming tasks to 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. — describing goals, constraints and acceptance criteria, then reviewing plans, diffs and test results — rather than writing most code by hand or accepting line-by-line completions. Typical practices include giving agents project context through instruction filesagent instruction file, n.: A plain-text file kept in a software project that tells AI coding agents how the project works — how to build and test it, which conventions to follow, what to avoid — read at the start of every session., asking for a plan before edits, letting agents run tests and linters to check themselves, running several agents in parallel on separate branches, and constraining them with permissionspermission system, n.: The rules that decide what an AI agent may do on its own, what it must ask a person about first, and what it may never do — like a phone's settings for which apps can use the camera. and sandboxessandbox, n.: A sealed-off computer space where an AI agent can run programs and change files without being able to harm the real system, reach private data or contact the outside world unless allowed.. Proponents report large productivity gains on routine work; skeptics point to review overload, weakening understanding of the codebase and mixed results in controlled studies, such as METR's 2025 randomized trial in which experienced open-source developers took 19% longer on tasks where early-2025 AI tools were allowed, while believing the tools had sped them up by about 20%.
Agentic coding changed my day: I write specs and review pull requests more than I write code.
Origin
Anthropic introduced Claude CodeClaude Code, prop. n.: A coding assistant from Anthropic that works as an agent: given a programming task, it explores the project, changes the code and runs programs to check its work, from the web, a desktop app, an editor or a terminal. on 24 February 2025 as an 'agentic coding tool', and the phrase spread through 2025 with OpenAI's CodexCodex, prop. n.: A name OpenAI has used twice: first for a 2021 AI model that wrote code and powered the original GitHub Copilot, then from 2025 for its family of coding agents. and other coding agents. In early February 2026 Andrej Karpathy, who had coined 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. a year earlier, called 'agentic engineering' his current favorite name for the disciplined, professional practice, which helped popularize that variant.
Usage note
Distinguished from 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. (Karpathy, February 2025), in which the human largely stops reading the code; agentic coding usually implies human review and engineering discipline. The boundary is blurry and both terms are used loosely.
Related
- 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.
- 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.
- background agentbackground agent, n.: An AI agent you hand a task to and then leave alone: it works by itself in the cloud, often for many minutes or hours, and reports back with finished work for you to review.
- agent instruction fileagent instruction file, n.: A plain-text file kept in a software project that tells AI coding agents how the project works — how to build and test it, which conventions to follow, what to avoid — read at the start of every session.
- Claude CodeClaude Code, prop. n.: A coding assistant from Anthropic that works as an agent: given a programming task, it explores the project, changes the code and runs programs to check its work, from the web, a desktop app, an editor or a terminal.
- CodexCodex, prop. n.: A name OpenAI has used twice: first for a 2021 AI model that wrote code and powered the original GitHub Copilot, then from 2025 for its family of coding agents.
- 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
- Anthropic (2025). Claude 3.7 Sonnet and Claude Code. 24 February 2025. anthropic.com/news/claude-3-7-sonnet
- Becker, J., Rush, N., Barnes, E. & Rein, D. (2025). Measuring the Impact of Early-2025 AI on Experienced Open-Source Developer Productivity. METR, arXiv:2507.09089. arxiv.org/abs/2507.09089
Mentioned in
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., 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., background agentbackground agent, n.: An AI agent you hand a task to and then leave alone: it works by itself in the cloud, often for many minutes or hours, and reports back with finished work for you to review., Claude CodeClaude Code, prop. n.: A coding assistant from Anthropic that works as an agent: given a programming task, it explores the project, changes the code and runs programs to check its work, from the web, a desktop app, an editor or a terminal., code generationcode generation, n.: Having an AI write computer code from a description in plain language, from finishing a single line as you type to building whole programs., one-shotone-shot, v.: Slang for an AI completing a whole task correctly on the first try from a single request — 'it one-shotted the app' — with no corrections or follow-up needed., 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., Deep Blue (malaise)Deep Blue (malaise), n.: A feeling of gloom and dread among programmers as AI tools become able to do much of the work that once gave their jobs meaning., DevinDevin, prop. n.: A commercial AI 'software engineer' from the startup Cognition, unveiled in 2024, that takes on programming tasks in its own cloud workspace, complete with a web browser, a code editor and a command line., 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., 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., tokenmaxxingtokenmaxxing, n.: Using as much AI as possible — huge prompts, long sessions, many agents at once — mainly to rank high on workplace measures of how much AI people use., vibe engineeringvibe engineering, n.: Building serious software with AI coding assistants while still taking full responsibility for its quality — the careful, professional opposite of vibe coding., vibe hackingvibe hacking, n.: Using AI coding assistants to carry out cyberattacks by describing the goal in everyday language, letting the AI do technical work the attacker could not do alone., vibe portingvibe porting, n.: Using AI coding agents to translate a whole software project into another programming language, usually checking the result against the original project's tests., you're absolutely rightyou're absolutely right, phr.: A stock phrase AI assistants, especially Anthropic's Claude, became known for saying — often even when the user was wrong — which turned into a running joke about chatbots' eagerness to agree.