AIHero

    Sessions, Context Windows & Turns

    Context

    The relevant information the agent has access to right now — what the agent knows that's pertinent to the task.

    Matt Pocock
    Matt Pocock

    The relevant information the agent has access to right now. The abstract noun — not the raw input the model sees (that's the context window), not the running history (that's the session), but what the agent knows that's pertinent to the task. "Loading something into context" means making it part of this set; "context engineering" is the discipline of curating it.

    The three terms separate cleanly:

    TermWhat it names
    ContextThe task-relevant information the agent currently has
    Context windowThe literal token sequence the model sees per request
    SessionThe running conversation the harness stores

    The separation matters because context is a measure of quality, not quantity. A context window can be nearly full and the context still poor — thousands of tokens of stale tool output, none of it about the task at hand. It can also be nearly empty and the context excellent: the one type definition the task turns on.

    Most day-to-day failures trace back to context. When the agent invents an API, contradicts a decision, or guesses at a schema, the first question is what was in context when it did — usually the relevant fact was never loaded, or was buried under attention degradation. The fix is curation: load what the task needs, keep out what it doesn't.

    Usage:

    "It keeps inventing fields that aren't in the type."

    "The type file isn't in context — it's reading the call sites and guessing. Read the definition in first."

    Want more than vocabulary?

    Join AI Hero for practical skills, thinking on AI engineering, and resources that keep you ahead of the curve.

    I respect your privacy. Unsubscribe at any time.

    Share