1.3 The Structured Prompt

Between a mediocre and a precise AI answer usually lies not a better model but a better prompt. Prompt engineering is not secret knowledge — it is the consistent application of what Chapters 1.1 and 1.2 established about how LLMs work.

The power gap made visible: naked prompt vs. structured prompt

(Examples contributed by Prof. Inwah Kim)

Example 1: Information Retrieval

Naked Prompt

Tell me about the effects of social media on teenagers.

Structured Prompt

Summarize the effects of social media on teenagers based only on peer-reviewed studies published after 2020. For each claim, cite the specific study. If you are not confident a claim is well-supported, explicitly state "Uncertain" instead of presenting it as fact. Present the results in a table with the following columns:

• Effect
• Supporting Evidence
• Confidence Level

Example 2: Image Generation

Naked Prompt

Draw a dog.

Structured Prompt

Draw a Golden Retriever puppy with the following specifications:

• Subject: Sitting upright, facing the camera
• Setting: Autumn park with grass and scattered fallen leaves
• Lighting: Soft natural golden-hour light (around 4:00 PM)
• Camera: Eye-level angle, shallow depth of field, blurred bokeh background
• Style: Photorealistic, DSLR photography
• Expression: Playful, mouth slightly open, looking directly at the viewer
• Quality: 4K resolution, ultra-detailed, sharp fur texture
• Aspect Ratio: 16:9

Closing the loop: why the probabilistic nature is exactly why structure matters

(Contributed by Prof. Inwah Kim)

Because AI predicts the most probable answer, not the correct one, structure is what narrows that probability space toward what you actually want.

That single sentence connects everything so far. A vague prompt spans a huge space of plausible answers — the model effectively averages over every conceivable response for every conceivable audience. Every precise element shifts probability mass onto exactly the answer class you need. And because Chapter 1.2 showed that models fill gaps plausibly when in doubt, every serious prompt should include an explicit rule for missing information — notice how Example 1 above builds in exactly that with its "Uncertain" instruction. Structure is not politeness toward the machine; it is probability steering.

The core concepts of prompt engineering

(Contributed by Prof. Inwah Kim)

1. Role — Assigning the model a specific persona (e.g., "As a customer support chatbot..." or "As a gardening expert...") narrows its response style and domain framing, producing more targeted, professionally appropriate output.
2. Context — Providing background information or a concrete scenario (e.g., explaining conduction/convection "using a frying pan and boiling water") helps the model ground abstract concepts in relatable terms, improving relevance and clarity.
3. Structure — How the prompt is organized — instructive tone, step-by-step framing, or prompt chaining (breaking a task into sequential sub-prompts: summarize → decide action → draft reply) — guides the model through complex reasoning rather than expecting one-shot completeness.
4. Examples — Few-shot demonstrations or specified formats (e.g., "three multiple-choice questions") show the model exactly what output shape is expected, reducing ambiguity compared to zero-shot requests.
5. Constraints — Explicit limits — length ("three sentences"), format, tone, or scope — bound the model's output space so it aligns with the user's actual requirements rather than defaulting to a generic, unconstrained response.

Core principle tying them together: each element works by reducing the model's guessing room — role and context tell it what domain/perspective to draw from, structure tells it how to reason, examples show it what "correct" looks like, and constraints define the boundaries of an acceptable answer. Together, they convert a probabilistic guess into a targeted, checkable output.

Prof. Kim's perspective: what makes a prompt actually work

(Contributed by Prof. Inwah Kim)

A good prompt is not necessarily a long prompt — it is a clear prompt. The following principles consistently lead to better AI responses.

1. Be specific. Vague prompts produce vague, generic answers. The more clearly you describe what you want, the more accurate and useful the output becomes.
2. Give clear, consistent instructions. Avoid conflicting requirements (e.g., "be comprehensive" and "use only 20 words"). Clear and realistic constraints help the AI produce better results.
3. Show an example when possible. If you have a preferred style, format, or tone, providing an example is often more effective than simply describing it.
4. Put the most important instruction first. Start with your main goal, then add context, constraints, and formatting requirements. This helps the AI prioritize what matters most.
5. Don't assume the AI knows your context. Explain any background information that a human might already know. Hidden assumptions are one of the biggest causes of poor responses.
6. Remember that confidence is not correctness. A well-written prompt can improve the quality of an answer, but it cannot guarantee that the answer is true. Important information should still be verified using reliable sources or external evidence.

Where good prompts actually have an impact

(Contributed by Prof. Inwah Kim)

1. Effective Work Performance — Good prompts help professionals complete tasks more efficiently and with higher quality.
   • Faster, higher-quality work
   • Clearer writing and communication
   • More efficient problem solving and decision making
2. Education and Learning — Structured prompts make AI a more effective learning partner by providing personalized guidance and meaningful feedback.
   • Personalized learning support
   • Better feedback and practice opportunities
   • Deeper understanding of concepts
3. Team Collaboration and Knowledge Sharing — Shared prompting practices make AI use more consistent, transparent, and reproducible across teams.
   • Reusable prompts and workflows
   • Consistent and transparent AI use
   • Easier collaboration and smoother handoffs

Build and deploy: your first structured prompts

Take a real task from your own studies — for instance, contextualizing a central-bank press release for a seminar paper. Write the naive one-sentence version first, then a version that uses the five core concepts above. Run both and compare along three criteria: Does the output hit the target audience? Is the format directly usable? Are unsupported claims marked as such?

An editorial note on this deliberate tension: Shay's practitioner view and Prof. Kim's principles are less contradictory than they appear. As models get better at inferring intent, the mechanical part of prompting shrinks — but what remains is precisely the substance of Prof. Kim's list: defining the objective (specificity), supplying the background the model cannot know (context), and never mistaking confidence for correctness (verification). The formula ages; the thinking behind it does not.

[CONCLUSION] A structured prompt is applied model theory: once you understand why context steers the probability distribution, you write better prompts automatically. The next question is no longer how to talk to the model — but what you may entrust to it at all. That is Chapter 1.4.