Artificial Intelligence on Fernando Ruiz

Agent vs. Assistant vs. Copilot: Three Ways to Delegate to AI

Mon, 01 Jan 0001 00:00:00 +0000

info Quick Definition

An AI assistant answers a request and leaves the next decision to the person. An AI copilot works alongside them inside their task and offers suggestions in context. An AI agent receives a goal, runs several steps with tools and only returns when it finishes or needs a human decision.

Where is the line?

The difference is not in the brand or the technology. It is in who runs the steps. In an assistant, the person does almost all the work and the AI answers. In a copilot, the person keeps carrying the work and the AI proposes the next step without taking it. In an agent, the AI decides and runs steps on its own, uses tools and adjusts what it does based on the results.

AI Product Strategy: How to Decide Where AI Goes

Mon, 01 Jan 0001 00:00:00 +0000

info Quick Definition

An AI product strategy is the plan that decides where AI is used inside a product, with which data, at what risk and against which success criteria. It is treated as a portfolio of bets reviewed regularly, not as a list of chat features.

What changes from a normal product strategy?

A traditional feature behaves in a fixed way: when the user clicks, something predictable happens. An AI feature produces outputs that vary and sometimes get things wrong. That changes three things: how you measure success, how you manage risk and how you plan the roadmap.

Claude Code: a programming agent that works in your repo

Mon, 01 Jan 0001 00:00:00 +0000

info Quick Definition

Claude Code is Anthropic’s programming agent. It reads your code, edits several files at once, runs commands and connects to external tools through MCP. It is used from the terminal, editor extensions, the desktop app or the browser, and it shares the same configuration across all of them.

What does it actually do?

Claude Code takes a task in natural language, for example “write tests for the authentication module, run them and fix the failures.” Then it plans, reads the files it needs, writes changes across several files and runs commands to verify them. It works directly with git: it stages changes, writes commit messages, creates branches and opens pull requests.

Loading states in AI interfaces: what to show while the model thinks

Mon, 01 Jan 0001 00:00:00 +0000

info Quick Definition

A loading state in an AI interface is what the screen shows while a model generates a response or runs a task. Its job is to tell people the system is still working, give a sense of how much is left, and stop them from thinking the app has frozen.

Why is it different in AI?

In a traditional app, a load usually lasts milliseconds or a few seconds, and it is known in advance. In AI, a response can take two seconds or twenty. The duration changes with the length of the text, the provider’s load and, in an agent, the number of steps it takes.

Response streaming: showing AI output while it is being written

Mon, 01 Jan 0001 00:00:00 +0000

info Quick Definition

Response streaming is the way to deliver an AI model’s output in chunks, as it is generated, so the person can start reading before it ends. It is usually sent over SSE (server-sent events). The model does not get faster: what changes is when the text appears.

What is it and how does it work?

A language model generates the response token by token (if tokens are new to you, see [[tokens|tokens]]). Without streaming, the server waits until it has the full text and sends it in one piece. With streaming, it sends each fragment as soon as it has it.

Tokens: The Unit AI Thinks (and Bills) In

Mon, 01 Jan 0001 00:00:00 +0000

info Quick Definition

A token is the smallest unit of text a language model reads and writes. It can be a whole word, a fragment of a word or a punctuation mark. AI providers measure the size of each request, and its cost, in tokens.

What is a token?

A model does not see letters or words the way we do. Before processing a text, a component called the tokenizer splits it into pieces from a fixed vocabulary. A frequent word is usually a single token; a rare, long or misspelled word is split into several. Numbers, spaces and emojis also take up tokens.

Vibe coding vs. AI-assisted engineering: the difference that decides quality

Mon, 01 Jan 0001 00:00:00 +0000

info Quick Definition

Vibe coding means programming with AI and accepting the code without reviewing it (the full definition is in [[what-is-vibe-coding|what is vibe coding]]). AI-assisted engineering uses the same tools inside a process with a specification, a plan, human review and tests. The difference is not the tool but what happens before and after the AI writes.

What separates one from the other?

Both start the same way: a description in natural language and an AI that generates code. The difference shows up at three moments.

What Is an AI Agent?

Mon, 01 Jan 0001 00:00:00 +0000

info Quick Definition

An AI agent is a system in which a language model receives a goal, decides which steps to take, uses tools to act (read files, run commands, call services) and reviews the results until the task is finished or it stops.

What does it mean that it “decides” its steps?

A traditional chatbot answers a question with text and waits for the next one. An agent receives a task such as “fix the failing tests in the authentication module” and works out the order by itself: first it runs the tests, then it reads the files that fail, then it edits the code and finally runs the tests again to check.

What Is an LLM? Explained for Designers

Mon, 01 Jan 0001 00:00:00 +0000

info Quick Definition

An LLM (large language model) is a neural network trained on enormous amounts of text to predict which token comes next after another. That simple ability, repeated over and over, produces answers, summaries, code and text that reads as if a person wrote it.

What is an LLM, in one sentence?

An LLM is a system that, given a text, calculates the probability of each possible next piece of text and picks one. It then repeats the process with the extended text. It does not look the answer up in a database or an internal encyclopedia: it generates it step by step.

What is vibe coding? Programming by describing what you want

Mon, 01 Jan 0001 00:00:00 +0000

info Quick Definition

Vibe coding is building software by describing in natural language what you want to an AI, and accepting the code it generates without a thorough review. Andrej Karpathy coined the term in February 2025, and in November 2025 Collins Dictionary named it Word of the Year.

Where does the term come from?

Andrej Karpathy, a co-founder of OpenAI and former AI director at Tesla, coined it in a post on X in February 2025. He described “a new kind of coding” in which you fully give in to the vibes, embrace exponentials and forget that the code even exists. In practice: talking to a coding tool, accepting all changes without reading the diffs and pasting error messages back. He himself said it was not too bad for throwaway weekend projects.