Syllabus Lesson 1 of 141 · The blind start
The blind start

How this course works

Why it matters You already use AI every day. This course is not about using it more. It is about noticing when its first answer is also your first answer, and what to do when that answer is not good enough.

Every graded problem here is a real engineering task: data that needs merging, a service that must stay inside a budget, a bug report with one example. Each one comes with the answers an AI assistant would most likely give, written out and sealed. You see them only after you pass.

The loop

  • Solve first. Read the rules, try it yourself.
  • Then compare. When you pass, the sealed answers open next to an expert's route and the principle behind it.
  • Name what you learned. One sentence, in your own words.

Sealed answer: a realistic first answer from an AI assistant, written well, kept hidden until you pass. Most of them fail a test, a budget or a hidden input that the exercise told you about.

Boring trap: a problem where the obvious answer is exactly right and the clever one loses. They are mixed in on purpose, so "always avoid the obvious" never works.

Two rounds of exams with the AI switched off

This module ends with four short exams, solved alone: no tutor, no hints. The last module has four more, matched one to one (same skill, new data, same difficulty). The course lines them up so you can see your own change, measured on your own work in this browser. Nobody else sees it.

Do the baseline now, before you learn anything here. A baseline taken after the lessons measures nothing.

Recap Solve, then compare, then name the principle. Some problems reward the plain answer. The baseline is only useful if you take it first and alone.

Spotted a problem in this lesson? Report it · Ask about this lesson

Code · runs in your browser
Output

        

      

      
      
      
      
      

        Syllabus
        
      

      
Your progress is counted anonymously. Change this