A repeatable process for solving real problems with agents.
Part 1
For this exercise, each of you is the CEO of your own life and work. CEOs get work done by stating goals and intentions, evaluating others’ work, and giving feedback. That is how we work with agents, too.
On their computer, the CEO should open the ChatGPT desktop app—not ChatGPT in a web browser—and start a new chat. Turn on the microphone and record the entire conversation while they are CEO. At the end of the turn, stop recording. The full transcript should now be in the CEO’s desktop app chat, ready to send to ChatGPT in Part 2.
Choose something that really matters to the CEO’s life or work. Understand why it matters. It must be something a computer or agent could plausibly help solve.
What is actually stopping them? Look for multiple problems, hidden problems, symptoms, and false assumptions. Get enough detail that you could take responsibility for solving it.
You should now understand the CEO’s most important problem. Shift from diagnosis to solutions.
Brainstorm until the CEO agrees the solution would definitely unblock the goal. The best case: they can fully describe a solution and hand it to an agent to automate.
If you have time, remove steps and complexity without weakening the result.
The instructor will tell you when time is up. Choose another person to be the CEO, restart the timer, and run the whole interview again.
Who found this difficult? Why?
Who found this super easy? Why?
Part 2
Open the CEO transcript from your turn in the ChatGPT Desktop app. Use GPT 5.6 Sol High, to make sure you're using an intelligent enough model.
Based on this transcript, what do you understand my goals and priorities to be? What would solve my problems?
Did it find a solution that feels like “hell yes, that would be awesome”?
Save the goal, problem, constraints, priorities, and agreed solution from this conversation in a file named GOALS.md. Put it in my personal folder. If “personal folder” is ambiguous, ask me where that is before saving. Write it so that a future agent can immediately understand what I am trying to accomplish, why it matters, what is blocking me, the solution I want, what success looks like, my constraints and preferences, and what remains uncertain. Show me the completed file and tell me exactly where you saved it.
I don't have a good enough problem statement or solution yet. Help me clarify my goals until we identify something that can plausibly be solved or automated with a computer. Interview me one question at a time. Help me distinguish symptoms from the actual problem, identify hidden constraints, and make the goal concrete. Once the problem is clear, brainstorm possible solutions with me. Keep refining them until I experience a “hell yes, I would love to have that solution” reaction. Do not start building until I explicitly agree that we have found that solution.
Part 3
This is one of the most important system-design questions. Local software is often simpler, cheaper, and more private. Cloud software can stay online when your computer is off and serve other people. Choose based on what the solution needs.
Do you think that this solution can be built entirely locally on my computer? Or will I need to run some software in the cloud somehow? Explain your recommendation in plain language. Compare whether it must work while my computer is off, whether other people need access, which apps or online services it must connect to, privacy, security, complexity, cost, and maintenance. Recommend the simplest option that can reliably solve the problem. Do not begin building yet.
If you agree the solution can be built entirely on your computer, use the local build prompt. If you agree it needs software in the cloud, use the cloud build prompt.
Hey, I have ChatGPT Desktop (formerly Codex) and Claude Desktop on my computer. If you think any of the built in features of these tools could be useful, feel free to propose using them. Tell me how you would build the solution. Prefer the simplest local design that can solve my problem. Explain what you would build, the tools you would use, what information or permissions you need, what happens when I use it, what could go wrong, what a useful first version contains, and what can realistically be built today. Do not start building until I approve the proposal.
Before sending: Replace the bracketed apps field before sending.
Hey, I have a Zo Computer, as well as [Replace this with any apps that you already use that might help with your goals]. Feel free to also look for prebuilt solutions that could help me. You can use those if you want. Tell me how you would build the solution. Prefer the simplest reliable design. Explain what would run in the cloud, which apps it connects to, what information or permissions you need, what happens automatically, cost, maintenance, risks, a useful first version, and what can realistically be built today. Do not start building until I approve the proposal.
You do not have to accept the agent’s way of doing things. This is your computer. This is your life.
Tell the agent how you feel and why. The model cannot read your mind or know what is best for you unless you tell it.
I am having a “hell no” reaction to this proposal. Interview me briefly to determine what feels wrong. Consider scope, complexity, technical difficulty, tools, cost, privacy, maintenance, workload, parts I do not understand, and whether it solves the problem I care about. Ask one focused question at a time, then revise the proposal. Continue until I tell you it feels like “hell yes, that would be amazing.” Do not start building before I approve it.
Part 4
HELL YEA, LET'S BUILD IT. Build the approved solution now. Start with the smallest version that demonstrates the solution is possible. Test your work, show concrete evidence of what is working, tell me when you need a decision or permission or when you need me to do something for you, try safe alternatives when something fails, track what remains unfinished, and do not silently expand scope. Continue building, testing, and improving for this work session.
Before sending: Replace the bracketed description.
I am blocked, stuck, confused, or unsure what to do next. Here is what is happening and how I feel: [DESCRIBE WHAT IS HAPPENING] Help me understand the blocker in plain language. Resolve it directly if you can. If you need something from me, ask for the smallest specific input necessary. Then tell me the single best next action and help me take it. Do not abandon the goal because this approach failed.
If your solution works, test it. Build, improve, and iterate. The first version rarely needs to be perfect; it should give you evidence that a solution is possible.
Test the solution we built and verify that it solves the original problem in GOALS.md. Check the important user flow from beginning to end, or tell me how to test this myself. Report what works, evidence that it works, what is incomplete or broken, risks or limitations, whether it meets our definition of success, and the most valuable next improvement. Fix safe, in-scope problems you discover, then test again.
Part 5
A retrospective helps you continuously improve your intuition and explicit process every time you solve—or try to solve—a problem. This is how you get better at solving problems forever.
Return to your original group. Choose the first CEO; the other two people are their team. Start a new transcription chat in the ChatGPT Desktop app for each CEO. The team helps the CEO identify what went well, what did not, and what to do better next time.
Do you have the same goal as when you started?
Did your understanding of the problem or goal change?
What went well?
What did not go well?
What else remains before the problem is solved?
Do you feel excited about continuing?
What did you learn?
What is one thing you can do better next time?
What is the next concrete action?
Reset the timer and run the full retrospective conversation again. Continue until everyone in your group has completed a 10-minute CEO turn.
After everyone has been the CEO
Based on this conversation, the work we attempted, and GOALS.md, help me run a retrospective. Ask me about anything you cannot infer. Then summarize whether I still have the same goal; how my understanding changed; what went well; what did not; what remains; whether I feel excited about continuing; what I learned; one improvement for next time; and the single best next action. Save the final retrospective as RETRO.md in my personal folder. If that location is ambiguous, ask me where it is. Show me the file and tell me exactly where you saved it.