Skip to content
    Field Notes

    Notes from the field.

    Inspiration and learned experience while building, shared to be used.

    Explore by question

    1. 01 · Evaluate

      How do you know an agent is working?

      Learn what correct behavior means and how to test it.

      Read the method: What is an LLM evaluation harness? Graders, protocols and closed loops for AI agents →

      Applied inAetherixLore

    2. 02 · Set boundaries

      Where should an agent's responsibility end?

      Separate prediction, recommendation and authority.

      Read the method: A wrong forecast is absorbed. A wrong action reaches a guest. →

      Applied inAetherix

    3. 03 · Learn

      Whose truth should an agent learn from?

      Distinguish approval, prediction and observed truth.

      Read the method: Whose truth is the ground truth? →

      Applied inAetherixLore

    All entries

    3 pieces
    Contact

    Discuss an evaluation protocol

    These notes document real systems and real failure modes. Want to challenge an evaluation harness, or compare notes on agent observability?

    Questions are stored for 90 days · Privacy

    · Ivan writes the notes on the pages in Rose's voice. Her answers here are live.