From prompt to role|INEMA.CLUB-PRO

Course · 8 lessons · prompt, context and role

Stop hunting for the perfect prompt: define the role

First, we wrote the request word by word. Then we learned to give context. Now the AI works like someone on the team, and the skill that counts is managing. In eight lessons you see what changed in each era, build an agent's role sheet, give it a goal it can check, train it in rounds, and keep an eye on the cost.

Start with lesson 1 Go straight to the role sheet
Dark office at night: a gray-haired manager watches humanoid robots working at separate desks.
  • 3eras: prompt, context, role
  • 3steps to delegate
  • 5–10rounds to train
  • 8lessons with practice

The lessons

From the three eras to the role sheet, from the measurable goal to training in rounds and cost. Each lesson has a diagram, a template to copy, practice with a "done when" and a commented quiz.

  1. Lesson 1 · 12 minThree eras: from prompt to roleIn just a few years, the way of working with AI has changed three times. First, the request written word by word. Then, the context the AI receives before the request. Now, the role: the AI works like someone on the team and you manage. Knowing which era each of your requests belongs to shows you what's missing.
  2. Lesson 2 · 14 minThe prompt era: what died and what stayedMany prompt techniques were born to make up for weak models: telling it to think step by step, inventing a 30-line persona, listing everything it must not do. Today's models do this on their own. Other techniques still work, but only under certain conditions. And some never get old. Telling one from the other makes your request smaller and better.
  3. Lesson 3 · 15 minThe context era: the AI knows before you askWhen the AI already knows who you are, what the company rules are and what went wrong last time, your request fits on one line. You don't build this base in every conversation: it lives in a place the AI reads on its own before it starts. It's what separates repeating everything every time from having someone who already knows the work.
  4. Lesson 4 · 16 minThe role era: brilliant, but new to the companyAnthropic, the company that makes Claude, recommends treating Claude like a brilliant but new employee who doesn't know your rules yet. Anyone who has hired people knows what that takes: a clear goal, the background of where it came from, and an example of work done well. In this lesson you put the three together into a role sheet and check your sheet right on the page.
  5. Lesson 5 · 15 minA goal you can measure: the M in SMARTManagers have been delegating with SMART for half a century: specific, measurable, achievable, relevant and time-bound. For an agent, the letter that matters most is the M. Without a measure, it can't check its own work and doesn't know when it's finished. "Improve my website" is not a goal. "The home page opens in under two seconds on a phone" is.
  6. Lesson 6 · 16 minTrain in a cycle: five, ten rounds and a listOnce the agent can do the work, the part almost nobody does comes next: training. Don't ask just once. Ask five, ten times, evaluate it the way you'd evaluate an employee, gather the corrections into a list and run it in a cycle until it fixes everything. It takes more time and more usage, but in the end you have a written standard procedure that works for any agent.
  7. Lesson 7 · 13 minThe downside: the bill, and how to manage itWith an agent running all day, usage grows fast: on a subscription you hit the limit; paying per use, it becomes a big bill. In my case, almost everything runs on the subscription. As with any employee, you need to audit and watch the cost. First what applies to everyone: the right model and keeping track of the bill. Then, for those who pay per use, cache and batch.
  8. Lesson 8 · 14 minThe same task in the three eras, and your planThe best proof that the way of working has changed is to do the same task three times: with just the prompt, with prompt and context, and with the full role. You see with your own eyes what each era delivers. Then you build the plan with dates to put your first agent to work, one step at a time.

Who it's for

  1. 1People who already use ChatGPT, Claude or Gemini and feel like they keep rewriting requests.
  2. 2People who took a prompt course and want to understand what still matters with today's models.
  3. 3People who have started using agents and want to delegate the way they would to a person.

What you'll be able to do

  1. 1Cut from an old prompt what the model already does on its own.
  2. 2Write an agent's role sheet, with a goal you can measure.
  3. 3Train the agent in rounds until you have a written standard procedure.
  4. 4Pick the right model for each task and keep track of the bill.

Where to do it

  1. 1In the chat (ChatGPT, Claude or Gemini), with a project of fixed instructions: this is the course's main path.
  2. 2In the terminal, with Claude Code or Codex, for those who already use it: lessons 3, 4 and 6 also show this path.

Who speaks in the course

  1. 1Nei Maldaner, from INEMA.CLUB: 40 years in technology, from programmer to company founder.
  2. 2When the text says "I" or "in my case," it is him telling how he works with AI agents today.