ML//agent

An AI agent repeatedly chooses actions from observations in pursuit of a task. In an LLM agent, the model proposes the next response or tool call and an agent harness executes the surrounding loop.

An AI agent repeatedly chooses actions from observations in pursuit of a task. In an LLM agent, the model proposes the next response or tool call and an agent harness executes the surrounding loop.

Tools expand the available action and observation spaces beyond text generation.

Planning may be explicit, implicit in repeated inference, encoded in a workflow, or distributed between model and harness.

State does not live automatically inside the model. Conversation history, files, memory, and environment state persist only because software stores and presents them again.

The useful shift is from “model as oracle” to “model as policy component.” Agency belongs to the assembled system, not to one completion.

On the web the agent becomes a web agent (search, fetch, browse) and in practice a hybrid agent that routes between an index, direct navigation, browser automation and APIs by cost. It does not want to traverse the Internet; it wants to obtain something.