Prompt as a control surface
Same model, same question, wildly different answers. What actually changed?
The idea inside
The prompt's real levers: context, examples, schema, tools, memory, temperature.
After this lesson
You can name the prompt's control surfaces (context, examples, schema, tools, memory, temperature) and what each one changes.
Where it leads
Inside this lesson
That's the real lesson stage, paused. Claim your pass to operate it.
See how AI actually works, end to end.
This lesson is one stop on the full arc. Unlock all of it, and keep it for life.
What you get
- The 34-lesson main path, a finishable route from a word to agents
- Goal tracks for using AI at work and building AI features
- Boss labs that make you apply a whole act, not just recognize it
- Spaced recall that brings each idea back before you forget
- Course memory: every term defined, with links to where it first appears
- A shareable capability card when you finish the main path
- Lifetime access on every device, every future lesson included
Not videos to watch. You predict, operate the machine, then prove it. That is why it stays.
99 interactive lessons and challenges. No videos, no code.
Free launch pass: lifetime access, no card needed
New here? The first lessons are free to try. Start with lesson 0.1
What this lesson shows
The prompt's real levers: context, examples, schema, tools, memory, temperature.
The question it opens with
Same model, same question, wildly different answers. What actually changed?
The walkthrough, in the lesson's own words
- This answer is a mess. Flip one lever below and watch the SAME answer change.
- Six levers, one model. Each maps to a lesson you've already seen.
- All levers off, temperature cranked high. The model rambles and guesses. Flip something.
- Why does the SAME model give such different answers?
- Nothing about the weights changed between these answers. Everything you flipped lives in the prompt: what context it can see, what examples it copies, what shape you demand, what tools it can call, what it remembers, and how much it's allowed to drift. The prompt is a control panel, not a wish.
- It's choosing which of these levers a problem needs. A flaky format? Reach for Schema. Wrong facts? Add Context or Tools. Drifting tone? Lower Temperature. You're not finding magic words, you're operating a control surface.
- Your assistant keeps returning answers your code can't parse and sometimes invents prices. A teammate wants to fine-tune. What do you try first?
- Reach for the prompt's control panel before touching the weights. Turn on a JSON schema so the output always parses, and attach the real pricing as context so it stops guessing. Each fix is a lever in the prompt, not a new model.
Key takeaway
You can see the prompt as a control panel and reach for the right lever per problem.
What you can do after this lesson
You can name the prompt's control surfaces (context, examples, schema, tools, memory, temperature) and what each one changes.
This is the written summary. The lesson itself is interactive: you predict, drag and operate the mechanism above, and the reveal answers you.