AI coding can feel like it's just for experts. Unexplained jargon. Mysterious failures. Bills that don't seem to match the work.
It isn't, really. A lot of the confusion is manufactured: there's a whole VC-funded economy that benefits from keeping it hard to understand.
The basic terms of engagement are learnable in an afternoon. Once you have them, the whole thing stops feeling like guesswork.
Why does context degrade? Why is the bill so high? Why does the same prompt behave differently from one day to the next?
Each has a clean answer, once someone tells you the words to use.
That's what this dictionary is for. The vocabulary of AI coding, translated into plain English.
Want more than the vocabulary? Join 62,000+ developers at aihero.dev/newsletter for my latest skills, thinking on AI engineering, and the resources that'll keep you ahead of the curve.
A moving label, not a technology. "AI" doesn't name a fixed thing the way model or token does — it points at whatever computers can newly, impressively do. Right now it points at large language models. It has pointed at very different things before:
Expert systems — thousands of hand-written if-then rules encoding human expertise.
1990s
Game-tree search — Deep Blue beating Kasparov (1997). Researchers avoided the word "AI" entirely
2000s
Statistical machine learning — spam filters, recommenders. Still sold as "machine learning", not "AI"
2010s
Deep learning — image recognition (AlexNet, 2012), AlphaGo (2016).
2020s
Large language models — ChatGPT (2022) made "AI" mean chatbots
The pointer moves by a known mechanism, sometimes called the AI effect: once a technique works reliably, it gets renamed — it's "just" search, "just" statistics — and "AI" slides forward to the next unsolved thing. The observation is old. Bertram Raphael put it this way in 1971: "AI is a collective name for problems which we do not yet know how to solve properly by computer." Larry Tesler's version, from around 1979: "Intelligence is whatever machines haven't done yet."
This is why conversations about AI so often talk past each other. A claim like "AI can't reason" or "AI is overhyped" carries a hidden timestamp — it may be about expert systems, about 2010s image classifiers, or about last month's LLM, and each reference supports a different conclusion. When a discussion about AI stalls, the fix is usually to swap the word for whichever precise term is actually meant: the model, the harness, the agent, the context it was given.
Avoid: "AI" in any technical claim — name the part you mean instead. "AI coding" as a label for the practice is fine; "the AI is hallucinating" is not.
Usage:
"The CTO wants to know whether AI could handle the triage queue."
"Translate that before scoping it — she means an LLM in a harness with access to the ticket system. 'AI' on its own isn't a spec."
Models can't read files, run commands, browse the web, or remember yesterday — it takes tokens in and predicts tokens out, once per model provider request. Everything that feels like an agent working — choosing tools, reading results, looping until the task is done — is the harness orchestrating many of those predictions in a row.
Model providers ship models in tiers: a large one that's smartest but slow and expensive, and smaller ones that are faster and cheaper but less capable. Picking a tier is a real decision — heavyweight for planning and hard debugging, lightweight for mechanical changes — and harnesses let you switch mid-session.
Being strict about the word also sharpens diagnosis. "The model is bad at this" is a specific claim — the same model in a different harness, or with a different context, often behaves completely differently. Before blaming the model, check what it was given: most disappointing output traces back to context or harness, not parameters.
Usage:
"Should we switch the model from Sonnet to Opus for the planning step?"
"Try it — but the harness is doing most of the lifting on this task. The model swap won't help if the system prompt and tools are wrong."
Parameters
The numbers inside a model — often billions of them — tuned during training. Everything the model "knows" lives in them. Training sets them; inference uses them unchanged. Also called weights.
Mechanically, the parameters are what turn input into output. Next-token prediction is a giant calculation: the tokens in the context window go in, get multiplied through the parameters, and a prediction for the next token comes out. There is no database of facts inside the model, no code lookup table — just these numbers, arranged so that the calculation tends to produce useful output. Facts the model can recite from training, like a standard library API, are parametric knowledge: stored in the parameters, not retrieved from anywhere.
The detail worth internalising is that parameters are frozen after training. Nothing you do in a session changes them — no correction you make, no codebase you show it, no mistake it learns from. Every session runs on the same numbers. This is why the model is stateless, why its built-in knowledge stops at the knowledge cutoff, and why anything project-specific has to arrive via context instead. The only way parameters change is more training — which produces, in effect, a different model.
Usage:
"Can we fine-tune it on our codebase?"
"That'd update the parameters — different model afterwards. For one project it's almost always cheaper to load the codebase as context than to retrain."
Training
The process that sets a model's parameters, by exposing it to vast amounts of text and adjusting parameters to improve next-token prediction. A one-time, expensive process done by the model provider. Encompasses both pre-training (the bulk run) and post-training (later refinements like instruction-following and safety); the distinction doesn't matter at this glossary's level.
The mechanism is repetition at scale: show the model a stretch of text, have it predict the next token, nudge the parameters toward whatever the actual next token was, and repeat across trillions of tokens. Nothing is stored as facts or rules — everything the model "knows" is a side effect of getting better at prediction, compressed into the parameters as parametric knowledge.
Two consequences matter day to day. Training ends at a point in time, so the model has a knowledge cutoff — it hasn't seen the library version you upgraded to last month. And training is not something you can do: when the model doesn't know your codebase, your conventions, or your internal APIs, the fix is never "teach the model" — it's putting that material into context, the one input you control.
Usage:
"Can we get it to know our internal API?"
"Not via training — that's a months-long process by the model provider. Load the API docs into context instead, that's the lever you actually have."
Runs the frozen parameters over your context to generate tokens
Read-only
Nothing you do at inference time writes back to the parameters — that's the reason a correction you make today doesn't stick tomorrow. The model that makes the same mistake next session, after you carefully explained the fix, hasn't ignored you; it's incapable of learning from the exchange. The model is stateless — continuity has to come from outside it — from the context window or a memory system.
This mechanism also explains how you're billed. Every request runs the model over the full context, so cost scales with input tokens and output tokens, and an agent making dozens of tool calls pays for inference on each round trip. This is why context size is a cost question as well as a quality one.
Usage:
"Why does the bill scale with usage instead of being a flat license?"
"You're paying for inference — every model provider request runs the model on the provider's hardware. Training already happened, but inference costs accrue per request, and a single turn can expand into many requests when tools are called."
Effort
Effort is a dial for how much reasoning a model does before it answers. Set per model provider request, it controls the length of the thinking the model works through before it starts writing the response you see. That thinking is generated at inference time like everything else; the harness often hides it, but it's real work the model is doing.
Higher effort costs more and runs slower. The reasoning is emitted as tokens, billed as output tokens even when you never see them, and produced one token at a time — so turning effort up lengthens the wait before the answer arrives and adds to the bill. The trade is more deliberation against speed and cost.
Most harnesses expose effort as a small ladder:
Level
What it's for
Low
Mechanical edits, lookups, well-specified changes with one clear path.
Medium
Everyday coding — the usual default.
High
Tricky bugs, design decisions, multi-step plans.
Max
The hardest problems, where a wrong answer is expensive to unwind.
The symptom of getting it wrong cuts both ways. Set effort too low on a hard problem and you get a confident, shallow answer that skipped the reasoning the problem needed — it reads fine and is wrong in a way that costs you later. Set it to max for a one-line rename and you sit through a long think that produces nothing the lowest setting wouldn't have.
Match effort to the task, not the session. Turn it up for the part that's genuinely hard to reason about, and back down for the rote work around it.
Usage:
"It keeps botching this concurrency fix — I've re-explained it three times."
"Bump the effort up. That's a reasoning-heavy bug, and on the default setting it's not thinking long enough before it commits to an approach."
Token
The atomic unit a model reads and writes. Roughly word-sized but not exactly — common words are one token, rare or long ones split into several. Context window size, cost, and latency are all counted in tokens.
Text becomes tokens via a tokenizer: a fixed vocabulary of tens of thousands of fragments, learned before training, that splits any input into a sequence of vocabulary entries. The model never sees characters or words — every piece of text is converted to tokens on the way in, and next-token prediction produces output one token at a time on the way out.
As a rule of thumb, a token is about three-quarters of an English word, so a thousand tokens is roughly 750 words. Code is less predictable: common keywords and idioms tokenize compactly, while generated identifiers, hashes, base64 blobs, and minified output split into many tokens per "word". The pattern: text that appeared often in the tokenizer's source material gets short, efficient encodings; text that didn't gets chopped into many small pieces. A hash like a3f9c2e1 never appeared anywhere, so it splits into many tokens, while function is one. This is why a small-looking file full of unusual strings can occupy a surprising share of the context window.
Tokens are the unit everything else is measured in. Cost is per token — providers bill input tokens and output tokens separately. Speed is tokens per second, since output is generated one token at a time. And the context window is a fixed number of tokens, so the token count of your files decides how much fits.
Avoid: "word" — token boundaries don't match word boundaries, and tokens-per-second / tokens-per-dollar are the units that actually matter.
Usage:
"How big is this prompt going to be?"
"Run it through the tokenizer — the schema's compact but the JSON keys are weird, so they'll split into more tokens than you think."
Next-token prediction
What the model actually does. Given a context, it samples one next token, appends it, and runs again. Every output — a sentence, a tool call, a thousand-line file — is built one token at a time. The model has no other mode of operation.
Each step works the same way: the tokens in the context window are run through the parameters, which produce a probability for every token in the vocabulary — this one is very likely next, that one less so. One token is sampled from those probabilities, appended, and the loop runs again with the slightly longer context. That sampling step is why the same prompt produces different output on different runs: non-determinism is built into the mechanism, not a bug layered on top.
Holding onto this mechanism explains behaviour that otherwise looks strange. The model never checks whether a token is true before emitting it — only whether it's likely — which is the root of hallucination. It commits to each token as it goes, so a confident-sounding opening sentence can steer the rest of the answer wrong. And because output tokens are produced strictly one at a time, generation speed puts a floor on how fast any agent can work.
Usage:
"How does the agent 'decide' to call a tool?"
"It doesn't — it's next-token prediction all the way down. The tool call is just a structured string the harness parses out of the output stream."
Non-determinism
The same input can produce different output. Run a model twice with identical context and you may get two different answers — sometimes a word, sometimes a completely different approach. Nothing in your code has to change for this to happen.
Что это такое?Бесплатный словарь, который объясняет жаргон ИИ-кодинга простым языком — термины вроде context window, harness и MCP — чтобы основы перестали казаться лотереей и осваивались за один вечер.
Это приложение, которое надо устанавливать, или просто что-то почитать?Просто материал для чтения: README в репозитории — это и есть сам словарь, без установки, без регистрации и вообще без каких-либо зависимостей. Более того, README генерируется из исходных файлов в репозитории с помощью скрипта сборки.
Как им пользоваться — просто читать сверху вниз?Да — он написан так, чтобы его можно было прочитать подряд примерно за один вечер: от базовых понятий о моделях до рабочих приёмов. Каждая статья сочетает определение простым языком с примером употребления термина в реальном разговоре.
Это полноценный курс по ИИ-кодингу или просто определения?Это словарь терминов, а не курс «от и до»: статьи объясняют и иллюстрируют термины, но не учат полным рабочим процессам. Если захочется идти дальше, README отсылает к рассылке автора.
Нужно ли разбираться в ИИ-инжиниринге, чтобы это читать?Нет — он рассчитан именно на тех, кого такой жаргон отталкивает, а не на экспертов. Вся идея в том, что ИИ-кодинг только кажется уделом специалистов, а рабочий словарь осваивается за один вечер.
Почему автор создал этот словарь?Его аргумент: путаница отчасти создана намеренно, потому что целая индустрия с венчурным финансированием выигрывает от того, что ИИ остаётся чем-то непонятным. Словарь — противоядие: базовый словарный запас осваивается за один вечер, и это лишает смысла «привратничество».
Что такое «harness»?Это слой оркестрации вокруг модели: системный промпт, инструменты, режимы разрешений, управление сессиями. Всё, что выглядит как работа агента — выбор инструментов, чтение результатов, циклы до выполнения задачи, — это harness, а не сама модель.
Можно ли навсегда обучить модели моему коду?Нет — параметры заморожены после обучения, а во время работы модель только читает данные, так что поправка, внесённая сегодня, завтра не сохранится. Реальные рычаги — это контекст (загрузить свои доки и соглашения) и системы памяти.
Он написан под конкретный инструмент вроде Claude Code или Codex?Нет — он не привязан к какому-то инструменту. Модели Claude и GPT упоминаются лишь как примеры, а определяемые термины (context window, harness, MCP, AGENTS.md) применимы к ИИ-кодингу с любым провайдером или агентом.
Бесплатный словарь, который объясняет жаргон ИИ-кодинга простым языком — термины вроде context window, harness и MCP — чтобы основы перестали казаться лотереей и осваивались за один вечер.