Write a prompt that gives the AI model precise context, role, task, and expected format.

The model does not know what happened after its training cutoff. Know each model's cutoff.

Don't default to the biggest model. Choose by: task depth, cost, privacy, speed.

Set "system instructions" once in your account = every chat starts from your context automatically.

When he opens sensitive topic, say "tell me more" vs reply.

Don't ask "do we need another hire?" — ask "if we started today with nobody, who'd we hire first?".

In few-shot, deliberate variety (positive/negative/edge case) matters more than count.