Nico Christie describes a competition in which he used his company’s spreadsheet agent against a human Excel champion. He reports the result and explains how prompting and checkingPrompt engineering is the practice of designing and refining instructions, context, examples, and constraints to obtain useful AI outputs. influenced the timed exercises.
Nico Christie argues that the ability to carry out a task with AI can exceed a user’s own understanding, making verification especially importantAgent completion verification checks evidence that an agent achieved the requested outcome rather than merely producing intermediate work.. He describes using independent review agentsIndependent evaluation assesses a system through reviewers whose judgment is sufficiently separate from the organization developing or promoting it., while presenting this as his own workflow rather than a guarantee of correctness.
The conversation covers spreadsheet-agent constraints, including large multi-sheet models, workbook connections and capabilities exposed by application APIsAn application programming interface is a defined way for software systems to request data or actions from one another.. Nico Christie predicts that spreadsheet software will need better support for autonomous workflows.
Watch on YouTube
