Lesson 25 of 32 · Part One, the principles
The Formation of the Master Mind
This is the chapter that gathers up all the others. Stop treating the machine as a vending machine that takes a request and dispenses an answer, and treat it as a junior partner: it holds more organised knowledge than you ever will and has no aim of its own, while you hold the aim, the facts of your own life, and the judgement about what is actually true and what will actually work. So say what each of you brings, invite it to ask you questions before it drafts anything, and read every draft as the senior partner - this part fits, this part does not, and here is the thing only I could know.
How to use The Formation of the Master Mind in Perplexity
A Master Mind is not limited to two members, and the honest way to place this tool is as the specialist in the alliance: the member who brings sourced outside knowledge and nothing else. It does not hold your aim, it will not remember last week's decision, and it has no judgement to lend you - but when the partnership needs to know what is actually true out in the world, this is the member to ask, because what comes back carries a link and a date. Use it for that single contribution. Bring it one question at a time, keep the findings in a Space so the alliance has a shared record, and take the verified facts back to the partner you are actually planning with.
- Treat it as the alliance's researcher, not its planner.
- One question per query; it holds no thread for you.
- Paste your aim and constraints each time - it remembers none.
From GenAI for Everyone: Prompt Engineering Handbook, First Edition 2025, by Raja Rajeshwari Seshadri and Shreeram Iyer.
A Master Mind is not limited to two members, and the honest way to place this tool is as the specialist in the alliance: the member who brings sourced outside knowledge and nothing else. It does not hold your aim, it will not remember last week's decision, and it has no judgement to lend you - but when the partnership needs to know what is actually true out in the world, this is the member to ask, because what comes back carries a link and a date. Use it for that single contribution. Bring it one question at a time, keep the findings in a Space so the alliance has a shared record, and take the verified facts back to the partner you are actually planning with.
- Treat it as the alliance's researcher, not its planner.
- One question per query; it holds no thread for you.
- Paste your aim and constraints each time - it remembers none.
- Keep the findings in a Space as the alliance's shared record.
- Reading about a paper is not reading the paper.
Instead of
Make a daily schedule for me and my son.
Perplexity version
The routine itself is not a research question - it is my life, my judgement and a conversation I will have with a planning tool. There are two pieces inside it, though, where a sourced answer beats a guess. I work in Bengaluru, 9:30 to 6, with an hour's commute each way, and my son is 8. Cite everything and give the date of each source. 1. What does published guidance from paediatric or public-health bodies say about how much sleep a child of 8 needs, and about screen time on school nights? Link the guidance itself, not an article about it. 2. What is documented about typical weekday commute times in Bengaluru, and about which hours are worst? Cite recent reporting or traffic studies, and say how much the figure varies. 3. What do after-school coaching or sports programmes in Bengaluru publish as their usual weekday timings for children of that age? Cite the programme pages. 4. Is anything published on school bus timing norms in Karnataka that might affect the 7:40 pick-up? Link the official source if it exists. If the evidence on anything is thin or contested, say so plainly instead of giving me one clean number. I will take whatever stands up to the tool I am planning the week in.
Then loop it
- Narrow the commute question to my own route rather than the city average, and tell me honestly if no route-level data is published.
- Now give me the sourced facts as a short list with links, and save the thread in a Space called "Our weekdays" so I can re-check when his school timings change.
Why it worksThe alliance's researcher supplies the few outside facts the routine rests on, while the aim, the constraints and the judgement stay with the person living the week.
Instead of
Tell me how to sell furniture online.
Perplexity version
This is the example in the chapter where an outside researcher earns its place: the vision is the carpenter's, and almost everything he needs to know changes year by year and is published somewhere. I am a carpenter in Saharanpur, hand-carving sheesham. My vision is to sell small carved pieces - jewellery boxes, wall panels, pooja stands - directly to customers in the big cities, without middlemen. I am not asking what I should want. I am asking what is actually there. Cite everything, prefer official sources, and give the date of each. 1. What artisan or handicraft seller programmes are currently open on the major Indian marketplaces and on government platforms? For each: what registration it needs, what commission it takes, and who handles shipping. Link the programme's own page. 2. Does Saharanpur wood carving carry a Geographical Indication registration, and if so, what does that allow a craftsman to claim and who may use it? Link the GI registry entry. 3. What do government handicraft bodies offer artisans for selling online - registration, an artisan card, training, exhibitions, help with photography? Link the official schemes and say which are currently open. 4. What is documented about the practical obstacles - packaging and breakage, courier weight charges for heavy wood, returns, GST on handicrafts? Cite each. 5. What recent coverage exists of Saharanpur's wood-carving trade and its export or online market? Cite it. Anything you cannot source, mark unverified. I will take this to the partner I am planning with, and the choice will be mine.
Then loop it
- Narrow to the courier question: what do the major couriers publish for weight-based charges on wooden parcels from Uttar Pradesh to Delhi and Mumbai? Cite their own pages.
- Now compress everything official into one page with links, and save it in a Space called "Selling my carving" so my nephew and I can both work from it.
Why it worksThe vision and the final choice stay with the craftsman, while the outside knowledge he lacks arrives sourced and dated instead of guessed at.
Instead of
Make my journal club presentation.
Perplexity version
The division of labour in this example suits a sourced research member exactly - with one limit stated first: I have the paper and I will read it myself. Searching for a study's details is not reading it, and I am the one who will be questioned. I am a junior resident in Chennai presenting a 15-minute journal club. I will give you the title, authors, journal and year. Cite everything and date it. 1. Find the paper's own record - the journal page, the DOI, the abstract, and any registered protocol or trial registry entry. Link each. 2. What published commentary, editorials, correspondence or critiques followed it? Link them and say in one line what each argues. 3. How many times has it been cited, in which sources, and what do the major citing papers do with it - support, qualify, contradict? Cite them. 4. What do current major guidelines in this area say, and does this paper's finding sit with them or against them? Link the guidelines with their years. 5. Is there a later or larger study on the same question? Link it and state its size and result. Do not summarise the paper for me from secondary sources, and do not tell me whether the findings apply to our patients - that judgement is mine and my department's. Flag anything behind a paywall so I can get it through my institution.
Then loop it
- Narrow to the critiques: quote what the correspondence actually objects to, with the source, so I can decide whether it holds.
- Now give me the sourced context in half a page - critiques, guidelines, the later study - and save it in a Space called "Journal club" so the next presenter in my batch can use it.
Why it worksThe researcher member supplies the sourced surroundings - critiques, guidelines, later evidence - while reading the paper and judging its relevance stay with the resident.
Questions
What is The Formation of the Master Mind in prompt engineering?
This is the chapter that gathers up all the others. Stop treating the machine as a vending machine that takes a request and dispenses an answer, and treat it as a junior partner: it holds more organised knowledge than you ever will and has no aim of its own, while you hold the aim, the facts of your own life, and the judgement about what is actually true and what will actually work. So say what each of you brings, invite it to ask you questions before it drafts anything, and read every draft as the senior partner - this part fits, this part does not, and here is the thing only I could know.
When should I use this technique?
On the work that matters and will take more than one sitting - a routine for your family, a plan to sell what you make, a presentation your professor will question - where the answer has to carry your judgement and not just the average of everyone else's.
How do I apply it in Perplexity?
A Master Mind is not limited to two members, and the honest way to place this tool is as the specialist in the alliance: the member who brings sourced outside knowledge and nothing else. It does not hold your aim, it will not remember last week's decision, and it has no judgement to lend you - but when the partnership needs to know what is actually true out in the world, this is the member to ask, because what comes back carries a link and a date. Use it for that single contribution. Bring it one question at a time, keep the findings in a Space so the alliance has a shared record, and take the verified facts back to the partner you are actually planning with.
What are the limits of this technique in Perplexity?
Be clear about what it cannot be. There is no standing relationship here: each question is answered afresh against the web, so the continuity this chapter depends on - your aim, your constraints, what you decided last Tuesday - has to live elsewhere or be pasted in every time. It will not ask you five questions before drafting, because it does not draft. And it will not help you with the two parts of this chapter that matter most: the judgement about whether a study applies to your patients, and the vision that makes a carpenter's plan his own. There is also a limit worth stating on the medical example - searching for a paper's details is not the same as reading it. The alliance still requires you to open the PDF.
How do I improve the answer after the first try?
In a Master Mind, the loop is the partnership itself. Do not expect a finished product in one message. Share your aim and invite the AI to ask questions first. Read its draft as the senior partner: what fits your vision, what is wrong, what only you would know? Say so plainly - "This part works, this does not, here is why" - and let it rebuild. Each round adds your judgement to its knowledge. When a plan is settled, save it and the key facts so the next session starts where you left off.
Loop it
In a Master Mind, the loop is the partnership itself. Do not expect a finished product in one message. Share your aim and invite the AI to ask questions first. Read its draft as the senior partner: what fits your vision, what is wrong, what only you would know? Say so plainly - "This part works, this does not, here is why" - and let it rebuild. Each round adds your judgement to its knowledge. When a plan is settled, save it and the key facts so the next session starts where you left off.
This lesson on the other tools
The technique is the same; the way you write it is not. ChatGPT, Claude, Gemini.