Generative AI Zero-Shot Prompting
Start with a direct instruction before adding demonstrations.
How it works
Zero-shot prompting gives a task without worked input-output examples. It is often a sensible baseline for a straightforward transformation. If the result is unreliable, inspect whether the instruction is ambiguous or the input lacks essential facts before adding more text. Keep a baseline so later changes can be compared against the same cases instead of relying on an impression that a longer prompt is better.
How to use this lesson
This is an authored prompt exercise with a human-written acceptance guide. It is not a captured provider response or a promise that a model will produce the target. No provider account, API, model weights or live inference was used for the course execution review. If you try it in an external service, use an authorised environment, check its actual access and data settings, and record the product/model and run date.
Bhylu does not run a generative model or send this prompt to an AI service. Copy and download preserve the text only. External tools may require accounts, paid access or compatible inputs; this course does not promise availability or prices.
Prompt exercise
Classify the message as schedule, supplies or unknown.
Message: "Which room is the 14:00 session in?"
Return the label only.
Acceptance guide — not a model transcript
The reference label is schedule.
The label set is explicit and includes an unknown category. That prevents the task from silently assuming every possible message belongs to one of the first two classes. Test messages that mix topics and messages outside the workshop domain before relying on the classification.