Vibecoding, defined

164 of 164 terms

• activation steering

  Also called steering vector, persona vector

  Changing a model's behavior by nudging its internal activations directly at runtime — no retraining, no prompt changes.

• Agent authentication and secrets

  Also called agent credentials

  Managing agent identity and credentials under least privilege.

• Agent automation in CI

  Also called CI agents

  Running versioned agent workflows from delivery infrastructure.

• Agent code execution

  Also called code interpreter

  Running model-generated commands or programs under controlled authority.

• Agent constraints

  Also called constraint-based prompting

  Boundaries that define what an agent may do and what success must satisfy.

• Agent data governance and lifecycle

  Also called AI data governance, agent retention

  Policies and controls for data collection, use, storage, location, retention, and deletion.

• Agent experience design

  Also called agent UX, AI UX

  Interface design that makes agent work understandable, steerable, and recoverable.

• Agent failure analysis

  Also called agent debugging

  Tracing a bad outcome to the system component that produced it.

• agent harness

  Also called harness

  The scaffolding around a model that turns it into a working agent — the loop, the tools it can call, and the rules for when to stop.

• agent hook

  Also called hook

  A script the harness runs automatically at a fixed point in an agent's loop — before a tool call, after an edit, on session start.

• Agent incident response

  Also called AI incident response

  The operating process for containing and learning from harmful agent behavior.

• Agent latency and cost engineering

  Also called agent economics

  Managing time and spend across model calls, tools, retries, and concurrency.

• Agent loop

  Also called reason-act-observe loop

  A cycle in which an agent observes, decides, acts, and interprets the result.

• Agent memory

  Also called long-term agent memory

  Information an agent can persist and retrieve beyond the immediate context.

• Agent mode

  Also called coding agent

  A mode in which one agent can inspect, edit, run commands, and iterate toward an outcome.

• Agent observability

  Also called agent tracing

  The ability to inspect an agent's actions, state, outputs, and resource use.

• Agent orchestrator

  Also called multi-agent orchestrator

  A coordination layer that assigns work and manages state across agent workers.

• Agent planning

  Also called plan mode

  Turning a goal into an inspectable sequence that may change as evidence appears.

• Agent production monitoring

  Also called agent monitoring

  Tracking agent quality and operations after deployment.

• Agent rate limits

  Also called concurrency limits

  Controls that bound request volume and simultaneous work.

• Agent regression testing

  Also called behavioral regression suite

  Repeating stable behavior checks after model, prompt, tool, or retrieval changes.

• Agent sandbox

  Also called sandboxed environment

  An isolated runtime with explicit limits for agent-executed work.

• Agent security threat modeling

  Also called agent threat model

  Systematically identifying how an agent could be misled, abused, or given excessive authority.

• Agent session management

  Also called session hygiene

  Keeping agent work coherent across growing history, compaction, handoffs, and restarts.

• agent skill

  Also called skill, SKILL.md

  A reusable instruction file that teaches an agent how to do one job well — the procedure, the tools, and what counts as done.

• Agent task decomposition

  Also called work decomposition

  Splitting work into bounded units that can be owned and verified independently.

• Agent tenant isolation

  Also called multitenant agent isolation

  Enforced separation of customer data and authority across every agent subsystem.

• Agent tool use

  Also called tool calling

  Giving a model structured actions for reading or changing external systems.

• Agent trust boundary

  Also called agent security boundary

  A separation that limits what an agent or untrusted input can access or change.

• agentic loop

  Also called tool loop, agent loop

  The cycle an agent runs in: decide, call a tool, read the result, decide again — repeating until the goal is met or a stop condition fires.

• Agentic software factory

  Also called software factory

  A repeatable system connecting specifications, agents, verification, review, and delivery.

• AI agent

  Also called agent

  An AI system that doesn't just answer once but works toward a goal in a loop — taking actions, reading the results, and deciding what to do next.

• alignment

  The work of making AI systems actually pursue what their builders and users intend, rather than something subtly or dangerously different.

• attention

  Also called self-attention, attention mechanism

  The mechanism that lets a model weigh which earlier words matter for the word it's currently processing — the core operation of a transformer.

• Autocomplete

  Also called code completion

  A model predicts the next code while the developer remains the author.

• Automated verification

  Also called executable verification

  Using executable checks to judge whether agent-produced work meets its contract.

• Autonomous agent team

  Also called agent swarm

  Multiple agents that coordinate work with limited direct human or central control.

• Background agent

  Also called asynchronous agent

  An agent that continues scoped work without a person attending the live loop.

• Backpressure

  Also called agent feedback constraint

  Signals and constraints that force an agent to stop or correct a failing path.

• benchmark

  A standard public test set for comparing AI models — the shared scoreboards behind every "model X beats model Y" claim.

• Browser automation

  Also called browser agent

  Programmatic operation and inspection of a rendered web interface.

• calibration

  Also called calibrated

  How well a model's confidence matches reality — a calibrated model saying "90% sure" is right about 90% of the time.

• chain-of-thought

  Also called CoT

  Having a model write out intermediate reasoning steps before its answer, which markedly improves performance on hard problems.

• Chat assistant

  Also called AI chat

  A conversational model that answers questions without directly owning repository changes.

• chunking

  Also called text splitting

  Splitting documents into passages small enough to embed and retrieve individually — the step that quietly determines whether retrieval works at all.

• CLI agent

  Also called terminal agent

  An agent operated from a command line with access to repository tools and shell commands.

• CLI tool

  Also called command-line tool

  A command-line capability an agent can discover and compose through shell commands.

• cloud model catalog

  Also called model catalog

  A selection of first-party and partner models offered inside a larger cloud platform's security and billing boundary.

• Compounding engineering

  Also called compounding workflow

  Improving the environment after each task so future agent work becomes more reliable.

• context compaction

  Also called compaction, context compression

  Summarizing an agent's earlier conversation to free room in the context window so a long session can keep going.

• context engineering

  Deciding what an agent sees at each step — the discipline that replaced prompt engineering once context windows got large enough to fill badly.

• context window

  Also called context length, context

  The maximum amount of text a model can consider at once — its working memory for the current conversation or task.

• data residency

  Also called regional data residency

  A requirement that data be processed or stored within a specified country or geographic region.

• Database and SQL

  Also called relational database, SQL

  Persistent structured storage queried and changed through explicit operations.

• dedicated endpoint

  Also called dedicated deployment

  A model deployment with capacity reserved for one customer instead of shared across many users.

• Delegated agent identity

  Also called agent delegation, agent authorization

  A traceable, limited identity an agent uses on behalf of a sponsor.

• determinism

  Also called reproducibility

  Whether the same input reliably produces the same output — something LLM systems mostly lack, which changes how you test and debug them.

• direct lab API

  Also called first-party model API

  An API operated by the same lab that creates the model family it serves.

• distillation

  Also called distilled

  Training a small, cheap model to imitate a big one's outputs, keeping much of the capability at a fraction of the cost.

• Document chunking

  Also called chunking

  Splitting source material into units that can be indexed and retrieved.

• Durable agent execution

  Also called durable execution

  Persisting agent state so long-running work can resume after interruption.

• embedding

  Also called embeddings, vector embedding

  A list of numbers representing a piece of text's meaning, so that similar meanings end up numerically close and can be searched.

• eval

  Also called evals, evaluation

  A repeatable test for AI quality — a set of tasks plus scoring — used the way software teams use test suites, because model output is too variable to judge by eyeballing.

• Evaluation dataset

  Also called eval set, golden set

  A repeatable set of representative cases used to measure system behavior.

• failover

  Also called fallback routing

  Automatically sending a failed request to another model endpoint or provider so the application can keep working.

• fine-tuning

  Also called finetuning, fine-tune

  Taking a trained model and training it a bit more on your own examples so it gets better at one specific job.

• Frontier model landscape

  Also called top models, state of the art models

  The changing set of models that lead meaningful workloads at a point in time.

• generative engine optimization

  Also called GEO, AEO, answer engine optimization

  Making a site the source an AI assistant cites when answering, rather than a link in a list of ten blue results.

• Git and version control

  Also called Git, version control

  A system for recording and comparing changes to a codebase.

• Git worktree

  Also called worktree

  A separate working directory attached to the same Git repository.

• Grounded generation

  Also called grounding

  Generation tied to supplied evidence rather than unsupported model memory.

• grounding

  Also called grounded

  Tying a model's answers to checkable sources — retrieved documents, live data, tool results — instead of letting it answer from memory alone.

• guardrails

  The checks around a model that block bad inputs and outputs — filters, validators, and permission rules the model itself can't override.

• hallucination

  Also called hallucinate

  When a model states something false with full confidence — inventing facts, citations, or APIs that don't exist.

• Harness engineering

  Also called agent harness

  Designing the full environment and feedback system around an agent.

• headless agent

  Also called autonomous agent run, background agent

  An agent running with no one watching — in CI, on a schedule, or triggered by an event — with its output reviewed after the fact.

• HTTP and JSON API

  Also called web API, REST API

  A common way for software to request data or actions from another service.

• Human oversight

  Also called human in the loop, human on the loop

  Defined points where a person approves, monitors, or retains a judgment.

• Human-agent handoff

  Also called approval flow, agent interruption

  A structured transfer of control and state between an agent and a person.

• human-in-the-loop

  Also called HITL

  Requiring a person's approval at specific points in an automated process, chosen so the irreversible steps are the ones a human sees.

• hybrid search

  Combining keyword and semantic search in one ranked result, so exact terms and paraphrases both find what they should.

• Implementer–reviewer separation

  Also called independent agent review

  Using a separate role or model instance to evaluate an implementer's work.

• inference

  Running a trained model to get answers — the phase where AI is actually used, as opposed to trained.

• inference cloud

  Also called AI inference cloud

  A cloud platform specialized in running open or custom AI models efficiently at production scale.

• Inference provider

  Also called model provider, LLM API provider

  A service or runtime that executes a model and exposes it to an application.

• Inline edit

  Also called editor edit

  A model-authored change applied directly to a selected region or file.

• jailbreak

  A prompt crafted to make a model ignore its own guidelines — usually through roleplay, hypotheticals, or encoding rather than a direct request.

• knowledge cutoff

  Also called training cutoff

  The date after which a model saw no training data — everything later has to reach it through search, tools, or the context window.

• KV cache

  Also called KV caching, key-value cache

  The memory a model keeps about text it has already read, so generating each new token doesn't require reprocessing the whole conversation.

• Large language model

  Also called LLM

  A neural network trained to predict and generate sequences of tokens.

• LLM

  Also called large language model, language model

  A large language model — the neural network behind tools like Claude and ChatGPT, trained on huge amounts of text to predict what comes next.

• LLM evaluation

  Also called agent evaluation, evals

  Measuring model or agent behavior against an explicit success definition.

• LLM-as-judge

  Also called model-graded eval, AI judge

  Using one model to score another's output against a rubric, so quality can be measured at a scale human grading cannot reach.

• llms.txt

  Also called llms txt

  A file at a site's root that gives AI systems a curated, plain-text map of its most useful content — robots.txt's counterpart for comprehension rather than permission.

• LoRA

  Also called low-rank adaptation

  A cheap way to fine-tune a model by training a small add-on layer instead of changing all of the model's weights.

• Major model lab

  Also called frontier lab, AI lab

  An organization that trains and releases important foundation-model families.

• MCP

  Also called Model Context Protocol

  The Model Context Protocol — an open standard that lets any AI assistant plug into any tool or data source without custom integration code.

• mixture-of-experts

  Also called MoE, mixture of experts

  A model built from many specialist sub-networks where only a few activate per token, giving big-model capability at small-model running cost.

• Model adaptation

  Also called fine-tuning decision, prompting versus RAG

  Changing the system or model so it performs a target workload more reliably.

• Model Context Protocol

  Also called MCP

  An open protocol for connecting AI applications to tools and data sources.

• Model family

  Also called model lineage

  A related line of models released under a shared name and technical lineage.

• Model inference

  Also called inference

  Running a trained model to produce an output.

• model router

  Also called AI router, LLM router

  A service that sends model requests across multiple labs or hosts through one API.

• model routing

  Also called model cascade, model fallback

  Sending each request to a model chosen by the difficulty of the task, rather than using one model for everything.

• Model selection

  Also called model choice

  Choosing a model using workload evidence and operating constraints.

• multi-agent

  Also called multi-agent system, subagent

  Using several AI agents on one problem — splitting work in parallel, checking each other, or filling different roles like planner and reviewer.

• multimodal

  Also called multimodality

  A model that works with more than text — reading images, audio, or video, and sometimes generating them too.

• Multimodal pipeline design

  Also called vision pipeline, media AI pipeline

  A typed flow that turns media into model-ready evidence and verifiable outputs.

• Multimodality

  Also called multimodal model

  The ability to understand or generate more than one kind of media.

• open weights

  Also called open-weight, open source model

  A model whose trained parameters are published for anyone to download and run — unlike API-only models you can access but never possess.

• Open-weight model

  Also called open model, local model

  A model whose trained weights can be downloaded and run outside the maker's hosted API.

• OpenAI-compatible API

  Also called OpenAI compatibility, OpenAI-compatible

  An API that accepts the common OpenAI request shape so existing clients can call another provider with fewer code changes.

• Parallel agents

  Also called multi-agent parallelism

  Multiple isolated agents working simultaneously on independent work streams.

• plan mode

  Also called planning mode

  A read-only agent mode that investigates and proposes a plan without editing anything, so you approve the approach before any code changes.

• Pre-trained model

  Also called foundation model

  A model trained broadly before it is adapted or prompted for a particular application.

• pretraining

  Also called pre-training

  The first, biggest phase of building a model: training it on enormous amounts of text so it learns language, facts, and reasoning in general.

• Product and use-case judgment

  Also called AI use-case selection, product judgment

  Choosing a real user problem where an AI system creates verifiable value at acceptable cost and risk.

• Programming fundamentals

  Also called coding basics

  The core ideas used to express behavior in source code.

• Prompt anatomy

  Also called prompt structure

  The instructions, context, examples, and output contract inside a model request.

• prompt caching

  Also called context caching

  Reusing the model's processed form of a repeated prompt prefix so subsequent calls skip re-reading it, cutting cost and latency substantially.

• prompt injection

  An attack that hides instructions in content an AI will read — a webpage, email, or document — tricking it into following the attacker instead of the user.

• quantization

  Also called quantized

  Shrinking a model by storing its numbers less precisely — like rounding — so it runs faster and fits on smaller hardware, at a small quality cost.

• RAG

  Also called retrieval-augmented generation

  Retrieval-augmented generation — fetching relevant documents first and pasting them into the model's context so it answers from your data instead of memory.

• Realtime conversational agent

  Also called voice agent, realtime agent

  An agent that processes streaming audio or visual input while a session is live.

• Reasoning effort

  Also called reasoning budget, test-time compute

  The inference-time budget a reasoning model spends before answering.

• reasoning model

  Also called thinking model

  A model trained to think — generating extended internal reasoning before answering — trading time and tokens for accuracy on hard problems.

• red teaming

  Also called adversarial testing

  Deliberately attacking your own AI system to find what makes it fail before someone else does.

• Repository instructions

  Also called AGENTS.md, project instructions

  Versioned guidance that tells agents how a repository is structured and verified.

• reranking

  Also called cross-encoder reranking

  A second pass that re-scores retrieved candidates by reading each one against the query, fixing the ordering that fast vector search got approximately right.

• Retries and idempotency

  Also called safe retries

  Recovering from failure without duplicating a completed side effect.

• Retrieval

  Also called information retrieval

  Selecting evidence relevant to the current request.

• retrieval crawler

  Also called answer-time fetch, AI search crawler

  A bot that fetches a page at the moment someone asks a question, so an assistant can cite it — distinct from a crawler collecting training data.

• Retrieval reranking

  Also called reranker

  Reordering retrieved candidates with a stronger relevance model.

• Retrieval-augmented generation

  Also called RAG

  Retrieving external evidence and placing it in context before generation.

• RLHF

  Also called reinforcement learning from human feedback

  Reinforcement learning from human feedback — training a model to prefer answers humans rate as better, which turns a raw text predictor into a usable assistant.

• Sampling controls

  Also called temperature, top-p

  Parameters that shape how deterministic or varied model generation becomes.

• sandbox

  Also called sandboxing, sandboxed

  An isolated environment where AI-generated code or agent actions run without being able to touch anything real.

• Self-improving agent system

  Also called self-optimizing agent

  A system that proposes and evaluates changes to its own operating components.

• serverless inference

  Also called serverless model API

  Model inference that scales on demand without the developer reserving or managing a continuously running server.

• Shared agent memory

  Also called multi-agent memory

  Governed durable knowledge shared among multiple agent workers.

• slash command

  A named, reusable prompt invoked with a leading slash, turning a workflow you repeat into one word you type.

• spec-driven development

  Also called SDD, spec-first

  Writing and agreeing the specification first, then having an agent implement against it — the spec, not the code, becomes the artifact you maintain.

• speculative decoding

  Also called speculative sampling

  A speed trick where a small model drafts several tokens ahead and the big model verifies them in one pass, often doubling generation speed.

• streaming

  Also called token streaming

  Sending a model's response token by token as it is generated, so the reader sees text immediately instead of waiting for the whole answer.

• structured output

  Also called JSON mode, constrained decoding

  Forcing a model's response to match a schema, so downstream code can parse it instead of guessing at prose.

• subagent

  Also called sub-agent

  A separate agent spawned by another to do one scoped piece of work in its own context, returning only the result.

• SWE-bench

  Also called SWE-bench Verified

  The standard benchmark for AI coding agents: real GitHub issues from real repositories, scored by whether the agent's patch passes the project's own tests.

• system prompt

  The standing instructions a model receives before any user input — defining its role, rules, tools, and tone for the whole conversation.

• Task framing

  Also called agent task brief

  Expressing an outcome, context, constraints, and definition of done for an agent.

• Terminal and shell

  Also called command line, shell

  A text interface for navigating files and running programs.

• test-time compute

  Also called inference-time compute, inference-time scaling

  Spending more computation when the model answers — thinking longer, trying multiple attempts — to buy accuracy without training a bigger model.

• Tests, builds, and deployment

  Also called software delivery

  The checks and release steps that turn source changes into running software.

• throughput

  Also called tokens per second

  How much model output a provider can generate over time, commonly measured in output tokens per second.

• time to first token

  Also called TTFT

  The delay between sending a model request and receiving the first piece of its response.

• token

  Also called tokens

  The chunk of text a model reads and writes in — roughly three-quarters of a word — and the unit AI usage is billed in.

• token budget

  A cap on how many tokens a task, session, or agent run may consume — the practical control on both cost and how long an agent will grind.

• tool permissions

  Also called permission model, allowlist

  The rules governing which tools an agent may call and which need confirmation — the boundary between a mistake and an incident.

• Tool schema design

  Also called function schema design

  Designing a tool interface so a model can select and call it correctly.

• tool use

  Also called function calling, tool calling

  A model's ability to call external functions — run code, search the web, edit files — instead of only generating text.

• transformer

  The neural network architecture behind modern AI models, built on attention — letting every word directly consider every other word in parallel.

• Vector search

  Also called semantic search

  Finding content by similarity between embeddings.

• vibe coding

  Also called vibecoding

  Building software by describing what you want to an AI in plain language and steering the result, instead of writing every line yourself.

• vibe engineering

  Using coding agents with the discipline of real engineering — tests, review, version control, specs — rather than accepting whatever the model produced.

• x402

  A payment protocol built on HTTP's long-unused 402 status code, letting a server charge per request so software agents can pay without an account.