Pattern recognition with a fancy suit on. AI predicts the next token from what you give it — say "twinkle, twinkle, little", it returns "star." Better input, better prediction. Use four parts every time:

[ROLE]    Act like a world-class strategist focused on conversion.
[CONTEXT] Product, audience, current copy, ICP notes …
[COMMAND] Rewrite the homepage hero to maximise booked demos.
[FORMAT]  Table — Headline · Subhead · CTA · Reasoning.

Output never exceeds input quality. The prompt caps the ceiling.

The context window is the river your output flows from. Pollute it and the output is polluted. Curate ruthlessly — every extra token steers the next one.

Length

Token budget is finite. Past it the model truncates or gets worse at finding the relevant bits. Pin what matters — don't dump entire docs.

Pollution

Irrelevant or contradictory info derails confident answers. One wrong fact buried in a long context will surface in the output.

Tone

Aggressive or negative wording produces statistically worse outputs. Models mimic register — feed it craft, not contempt.

Don't argue

When the answer is wrong, don't reply "you're wrong, redo." Edit the original prompt or branch a fresh chat — corrections + the mistake all stay in context and skew everything that follows.

Reset

Don't keep one chat going forever. Old goals bleed into new tasks. New goal → new session.

Edit the prompt, not the response.

Like a GPS, not turn-by-turn directions. Engineers call this spec-driven development.

1. Set the role and context.
2. State the outcome — be specific. "A sequence that converts cold leads into booked calls."
3. Add: "Ask all the questions you need. Return as <format>."
4. Answer conversationally — voice-to-text saves typing on long question lists.
5. If off-target: "refine — ask more questions." Iterate.

Spec-driven — destination first, route second.

Use AI to scaffold its own work. Two reusable patterns for the same problem — quality output without grinding 30 iterations.

1. "Write a plan first. Steps, dependencies, edge cases. Don't execute yet."
2. Review the plan. Refine in plain English: "add this, restructure that."
3. Say "go" when it's right. Catches wrong assumptions before any output is generated.

1. "Senior prompt engineer. Goal: <X>. Write me a prompt — role, context, constraints, format."
2. Tweak the draft to fit your situation. Run it.
3. If output disappoints: "Why did this fall short? Rewrite the prompt." Save the polished version, reuse it.

You direct — the model plans and drafts.

A recipe that runs on repeat — same ingredients, same sequence, predictable result instead of 17 minutes of back-and-forth.

Agent

A system prompt packaged for reuse — a Custom GPT, Claude Project, Gemini Gem. Define role, constraints, tone and tools once; talk to it forever.

Skill

A smaller, composable capability the agent invokes when relevant — "summarise contract", "extract invoice", "format SQL." Skills stack inside agents.

1. "Expert prompt engineer. Create a system prompt that does <X>. Ask what you need."
2. Answer, test, refine. You code system prompts with words.
3. Save as an Agent. Compose Skills inside it. Hand the link to your team.

Pro tip — browse aitmpl.com for ready-made Agents and Skills.

Sharp output comes from sharp thinking — yours, and the model's. Push both. Machines optimise what exists; humans imagine what doesn't, and the model only catches its own blind spots when you make it look.

Taste

Stand under a waterfall of excellence. Curate your feed, follow the best, subscribe to the newsletters practitioners actually read.

Vision

Schedule thinking blocks. Plan, doodle, daydream. Your brain doesn't think in words — say "pink elephant" and you saw one.

Care

Use reclaimed time for connection. Celebrate clients and team. Be present with family.

1. After any answer: "What's wrong with this? Argue against it as a skeptical senior reviewer."
2. For code or output: "Code-review what you just produced — bugs, edge cases, weak assumptions, missed requirements."
3. Iterate: "Where would this fail in practice? Now fix it." Don't ship the first reply.

Critical on both sides — AI becomes your weapon, not your replacement.