Prompt Writing

How to ask a machine for something and get back an answer you can use.

This page lists what a learner is able to do at the end of this class. It does not carry the class itself. The lessons, exercises and answers stay in the classroom and in the book.

What you will be able to do

What the class is really about

Most people's first disappointment with AI comes from asking badly. They type four words, get something generic back, decide the tool is not for them, and never return. The gap is not in the machine. It is in the asking.

This class is about the asking. A weak prompt and a strong prompt go to the same model and come back with completely different answers. By the end you can tell which one you have written before you press enter. Three things do most of the work. Telling it who it should be. Giving it the context it cannot see. Saying what shape you want the answer in.

The second half is that the conversation does not end at the first answer. Most people take whatever appears and move on. The useful skill is saying what was wrong with it and asking again. Two or three rounds of that produce something you would actually be willing to send.

We work on documents you have a real reason to write. A leave application. A complaint to a supplier or a service provider. A job application. A long document that needs cutting down to a paragraph. Something in English that has to be in Hindi, or the other way round. The point is that you leave with the thing you came to write, finished.

We also cover the difference between a chatbot and a search engine, because that difference decides what you should trust each of them for. A search engine points you at a page somebody wrote. A chatbot writes you a new one, and it will do that just as readily when it has nothing to go on.

Careful use

Verification. A confident answer is not the same thing as a correct answer.

Careful use is not a separate lesson at the end of the course. A piece of it belongs to every one of the eight classes, and this is the piece that belongs to this one.

How this class is taught

  • One core concept. The class carries a single idea. Nothing is covered twice and nothing is rushed.
  • Hands before theory. You do the thing first. The explanation lands better once you have seen what happened.
  • Practice in company. Weekday slots in the lab follow the weekend class, with somebody there to help.
  • Mixed ages at one table. The youngest in a batch may be ten and the oldest past fifty. Each takes something from the other.

Chapter 1, What is AI? · Chapter 3, How Machines Learn

All eight chapters

How to apply