AI Agent
A language model given tools, context, and a goal — so it can take actions and check results across multiple steps, instead of producing one answer and stopping.
Production AI is a system: context, tools, permissions, traces, evals, and feedback loops around the model.
What it is
An AI agent is a language model put in a loop: it receives a goal, looks at the context, decides on an action (call a tool, ask a question, write a result), observes what happened, and decides again. A chatbot answers; an agent works. The word covers everything from a single automated step to a long-running process with memory and approvals.
Why it matters
Agents are where AI stops being a writing aid and starts being labor: triaging tickets, drafting replies, updating records, monitoring systems. That shift also raises the bar — an agent's mistakes are actions, so production agents need permissions, approval gates, traces, and evals around them.
How it works
The loop is: assemble context, present tools, let the model choose, execute the choice, feed the result back, repeat until done or escalated. The engineering lives in what surrounds the model — the agent harness, the tool registry, the runtime that makes the loop restartable and observable.
Related resources
The scaffolding around an AI agent — prompt construction, tool dispatch, retry logic, trace emission, state management — that turns a model into a workflow participant.
A workflow where one or more AI agents make decisions about what to do next — choosing tools, gathering context, escalating to humans — rather than executing a fixed script.
The mechanism by which a language model invokes external functions — APIs, databases, code execution, retrieval — and reads the results back to continue its work.
The execution engine that turns an AI agent from a chat-window demo into a long-running, event-driven, restartable process you can trust with real operations.